Universal Emulators Return

webmilhouse writes "Wired has an article about Transitive Corporation that claims their software "allows any software application binary to run on any processor/operating system" without any performance hit. That would allow any program written for Windows to run on Linux or Mac, and vice-versa, which Wired likened to digital alchemy. The Transitive software is supposed to be released today. What do you think, vaporware or miracle?"

Comments

Not Vapor and not the arrival of Christ

[AKAImBatman]

Like many hyped up concepts, I don't think this product is really all they're making it out to be. At the same time, however, I don't think it's vapor. Instead, it's probably something in between that performs as advertised, but mitigating factors (300MHz CPU?) result in it not being everything everyone expected.

Re:Not Vapor and not the arrival of Christ

[Nos.]

Okay, but for a product that really is this good, why is the newest news on their site dated March 2003? (There's an article in 04, but it has nothing to do with what they're releasing)

Re:Not Vapor and not the arrival of Christ

[Atreide]

maybe because it's not vapor anymore ?
it's already evaporated !

Re:Not Vapor and not the arrival of Christ

[Kierthos]

So if it went from glossy brochues to vaporware, would it be product sublimation?

Kierthos

Re:Not Vapor and not the arrival of Christ

[Lesrahpem]

So many times with a product like this they will boast something amazing, but then in the small print you find something to the effect of "for this product to function as described, you must meet the following requirements....". Such requirements always seem to be very obscure or hard to meet, leading to a product which does not do exactly what it says it does.

A somewhat related example is Unreal Tournament 2003 for Linux. It works, but only if you have a GeForce. Granted, that's not an outlandish requirement since nVidia cards work well on Linux anyway, but you get the idea.

Re:Not Vapor and not the arrival of Christ

[Anonymous Coward]

Well, looking at their website, I'd say the vapour is not going to hold -- otherwise they'd have their webserver emulate a way faster machine with a significantly faster Internet connection... ;-)

Re:Not Vapor and not the arrival of Christ

[mattyrobinson69]

not to be an arse, but i dont know what you mean. there are many windows only games that require a recent gforce or recent radeon in order to use their transparancies, bump mapping or whatever. even the old open source tux racer has special features such as bump mapping which only works with gforce cards iirc. do you have any other examples of anything more obscure than that?

Re:Not Vapor and not the arrival of Christ

[skraps]

All of you naysayers take note:
This is not vaporware! I am on the beta test team for this product. In fact, right now, I just fired up a copy of Duke Nukem Forever on OS/2. Works like a champ.

Re:Not Vapor and not the arrival of Christ

[Anonymous Coward]

vapor is evaporated already

did you fail the earth science regents exam?

Re:Not Vapor and not the arrival of Christ

[Anonymous Coward]

Let me know how the Linux binary for Doom 3 turns out.

-Thanks

Re:Not Vapor and not the arrival of Christ

[grantsellis]

Yes, it's not all they're making it out to be. The poster read the teaser instead of the article, not that we're surprised :)

QuickTransit fully supports accelerated 3-D graphics and about 80 percent computational performance on the main processor.

and

Analyst Rob Enderle of the Enderle Group said Transitive benefits from the fact that most modern machines are fast enough to emulate each other without much affecting performance.

"Typically with emulation you take a big performance hit," he said. "Their big breakthrough is they are much more efficient ... but there's so much overhead anyway, you can pretty much put any software on any platform. The power user might notice the difference, but the other 95 percent won't notice."

so yes, it does affect performance. You take a 20% hit. The "almost no performance hit" means, in this context, "computers are fast enough that no one will notice unless they're doing something crazy like video editing. Go back to surfing slashdot."

Re:Not Vapor and not the arrival of Christ

[Diplo]

A somewhat related example is Unreal Tournament 2003 for Linux. It works, but only if you have a GeForce.

When it comes to gaming only two graphic card manufacturers count, and that is ATI and Nvidia. Nobody else makes cards that can run modern games (with the possible exception of the Matrox and the over-priced Parhelia). Of ATI and Nvidia only Nvidia have decent drivers for Linux. This is why you need a GeForce to run UT 2003/4, not because of anything inherent to the game. If ATI released decent Linix drivers with good OpenGL support, you could run it on Radeon etc.

Re:Not Vapor and not the arrival of Christ

[Anonymous Coward]

A couple of friends of mine work for transative. I know there are a lot of smart guys there (top of class masters, phds, lecturers). My mates have been very secretive when i've been asking them what there up to for more than a year now and have said it will be a big thing when it comes out. I also know getting binaries to run on different systems is their speciality. So i know the company isn't vapour and neither are their employees (or their swanky city centre flats). I believe the program isn't complete vapour either, but how well it lives up to it's claim of any binary on any machine...i don't know.

Re:Not Vapor and not the arrival of Christ

[Anonymous Coward]

Wait; they got Enderle for a quote?

It's all bullshit.

Re:Not Vapor and not the arrival of Christ

[Fyoozen]

I think this is one of those "show me" items, which many would greatly like to see succeed. The ability retain software and change OS's would be a great step to freeing users from OS bondage. Its potential is a bit frightening from a security aspect. Could we be looking at an new means of spreading malignant code across various OS platforms? It would require work, but when it comes time to mess over your fellow man, there are some that rejoice in doing it. You rightfully point out, also, that there may be some other caveats to be considered.

Re:Not Vapor and not the arrival of Christ

[ajs]

Yep, none of this is new tech. In fact, it's pretty old tech at this point. Emulators/translators and everything in between have been the subject of experimentation for decades.

Actually, I expect to see someone sit down and write this for Parrot sometime soon. Especially of interest would be an S/390 emulator written in Parrot.

Parrot, for those who don't know, is a VM that targets very high level languages, but it's flexible enough and has a sufficiently strong JIT compiler that a hardware emulator extension to Parrot could easily produce code that would perform as well as the described product.

The cool part about writing such an emulator for Parrot is that you get access to the resulting emulated code from a number of high-level languages, so you could port over your S/390 airline application written in TPF and call its routines from a Java, Perl, Scheme or Ruby program, jumping into and out of hardware emulation as you go. While high-level languages would only have gross access to data as opaque objects, the hardware emulator could provide the ported code with everything that it expects.

"Emulation" is a sophisticated art at this point, and it's going to get very interesting over the next few years.

Re:Not Vapor and not the arrival of Christ

[Anonymous Coward]

damit! you should have said runnning it on The Hurd ;)

Re:Not Vapor and not the arrival of Christ

[gasmasher]

I recently picked up an ATI 9800Pro 128MB and it works fine with the ATI drivers under Xorg (AMD XP 2000+, 1G SDRAM, kernel 2.6.8.1). The single card drives two CRTs (one with a DVI to VGA adapter) for normal desktop use. I play UT2K4 for about an hour everyday at 1600x1200 on a single monitor with the detail settings right in the middle with no problems.

Re:Not Vapor and not the arrival of Christ

[skraps]

LOL. I tried for a moment to come up with a good vaporware OS, but nothing came to me. :-)

Re:Not Vapor and not the arrival of Christ

[Anonymous Coward]

There's a release on their site dated September 13th 2004, which is today I believe.

Re:Not Vapor and not the arrival of Christ

[avgjoe62]

LOL. I tried for a moment to come up with a good vaporware OS, but nothing came to me. :-)

Longhorn? Or a secure MS-OS?

AmigaOS?

I think most of those are in the same world where I'm sitting on the fantail of my yacht, playing DukeNukem Forever! on my OQO while Natalie Portman is in the galley whipping up a batch of hot grits!

Re:Not Vapor and not the arrival of Christ

[Croaker]

Hmm... well, to prove their product actually works (and to confuse the hell out of Netcraft) they could host their site on IIS running under OS X.

Re:Not Vapor and not the arrival of Christ

[GoClick]

Yeah right.
We've seen lots of companies that pay hansomly to people who work on nothing. Their just bloated on VC. How many .coms have we all seen who went from 100 people living in swanky lofts to 2 guys working in a basement ass soon as everyone found out they didn't have a product or worse 25 people in jail and 75 living in boxes?

Re:Not Vapor and not the arrival of Christ

[ebh]

There's always the infamous S1 operating system from the 80's, which consisted solely of a bunch of ads in Byte with the tagline "Unix is a dinosaur, MS-DOS is a toy." I don't think they ever shipped.

Re:Not Vapor and not the arrival of Christ

[Riff6809]

After reading the article, I personally see the whole issue as much ado about nothing. The quotes provided in the article leave me with the impression that those who were involved with the article had little experience with emulators or were quoted out of context. I think this is obvious given remarks like "One of the key breakthroughs is an 'intermediate representation'..." that imply revolutionary thinking when in fact the ideas are not new.

I'm sure their product does whatever they designed it to do, but the article alludes to platform migration and operating system virtualization. This screams out to me that the emulated programs are going to be very well behaved out of necessity, and most hardware interfaces will not be accessable except through API calls. Additionally, desktop PC software and operating environments tend to be much more 'regular' than embedded systems like game consoles. It is much easier to describe the behavior of user-mode code on a platform with a generic memory space and API set than it is to describe the behavior of an embedded multiprocessor system with control registers, DMA, custom graphics and audio subsystems and banked memory.

I also have to question the allegation that "no one has successfully developed an emulator for multiple processors and operating systems." Dynamic recompilation is not new. Intermediate representations are not new. Surely there exist some emulators which are capable of emitting multiple native instruction encodings in the backend. If none exist, I doubt it is because they are not capable of doing so.

Describing a processor architecture and providing an API mapping is not a trivial task by any means. The Transitive tool doesn't just 'simply work,' its requires a massive undertaking to prepare the behavior descriptions that I imagine would be in some ways more difficult than writing an ad-hoc single-platform emulator. I think that calling their tool a "hardware virtualizer" is probably a good idea, but not because its faster than an "emulator," but more because its likely nowhere near as powerful as a system emulator.

Finally, I would also beware the performance claims. Dynamic recompilation is certainly the way to go for ultimate performance, but when you generalize architectures, you often lose the ability to take advantage of native features. Also related to processor capabilities, the overhead incurred by emulation is highly correlated to the disparity between the host platform and the emulated platform. Transmeta processors suffer about 20% overhead and thats using a flexible VLIW architecture designed with x86 emulation in mind and using a dynamic recompiler that supports *only* x86. Thats a huge performance penalty, even if programs are running as fast as needed. Given the generalized emulation approach, I think its clear that the feasibility of such an approach is going to depend heavily on the host platform being more powerful/flexible than the emulated platform.

FWIW, I am the author of Nuance, the Nuon emulator. Nuance currently performs all of the same feats listed in the article including block translation, optimization and 'OS' virtualization (native implementation of the Nuon BIOS). I'm currently working on emitting native code using a custom x86 run-time assembler and backpatch mechanism.

Re:Not Vapor and not the arrival of Christ

[Anonymous Coward]

Let's see:

Freedows - RS the Vapor King

Re:Not Vapor and not the arrival of Christ

[Anonymous Coward]

I wouldnt be at all surprised if this was just a deliberate attempt to create media hype surrounding the company in order to inflate its value prior to a sell off or something... I get the feeling that when the cloud of vapour finally lifts, a lot of people will be disappointed, and a couple of people will be much richer.

Re:Not Vapor and not the arrival of Christ

[wamatt]

Or you could just slice through the BS with occam's razor http://en.wikipedia.org/wiki/Occam's_Razor.
The fantastical and the logical explanation. Hmm tough choices..

Re:Not Vapor and not the arrival of Christ

[Anonymous Coward]

The greatest trick Transitive Corp ever unintentionally pulled was convincing the world its products were just vaporware.

Re:Not Vapor and not the arrival of Christ

[Squareball]

Freedows

Re:Not Vapor and not the arrival of Christ

[Kindaian]

Not vapor... just lack of investment...

In reality, the Alphas and Windows had something similar (called fx86! or something like that). I the end, the stuff was emulated "slowly" will the system would check what was going in with the code... and then would "morph" it to true "alpha" opcodes... plus lots of meta (so a roll-back would be possible).

With some time of usage, the program would perform as in a real environment (actually would perform as if it was pure alpha code - which wasn't that far fetched)...

Alphas... the memory... ;)

Re:Not Vapor and not the arrival of Christ

[OxyFrog]

Hurd?

Re:Not Vapor and not the arrival of Christ

[AnfieldSierra]

You only have to check out GrokLaw to find out how much credibility should be given to anything Rob Enderle says. He's a shill. "Snake oil. Get your snake oil here."

Re:Not Vapor and not the arrival of Christ

[drinkypoo]

You're going to suffer more than 20%. There's no way to reasonably reoptimize code such that it changes endianness, so you're going to have to emulate that. Basically you're going to have to emulate the whole processor. This is going to take more than 20% out of you, especially if you're running UltraSparc or Alpha or x86-64 code on a 32 bit chip, for example...

Re:Not Vapor and not the arrival of Christ

[TheoMurpse]

"maybe because it's not vapor anymore ? it's already evaporated !" see, but when something EVAPORATES, it becomes VAPOR! ^_^ eg ice--melt-->water--evaporate-->watervapor--condens e-->water--freeze-->ice

Re:Not Vapor and not the arrival of Christ

[TheoMurpse]

"maybe because it's not vapor anymore ?
it's already evaporated !"

see, but when something EVAPORATES, it becomes VAPOR! ^_^

eg ice--melt-->water--evaporate-->watervapor--condens e-->water--freeze-->ice

Re:Not Vapor and not the arrival of Christ

[fatphil]

fx32!

(alpha)linux tried to adopt it as em86, but that seems a dead project, I could never get it to run.

It was bloody clever. My 533MHz 21164 ran like a 266MHz P2 (which was the same vintage or newer).

However, getting a fully 64-bit processor to emulate a crappy 32-bit one was never a clever idea, in the grand scheme of things.

FP.

Re:Not Vapor and not the arrival of Christ

[edittard]

There's an echo in here.

Re:Not Vapor and not the arrival of Christ

[fatphil]

Why couldn't your post be nearer the top? I was nearly giving up on this story due to the unbelievable amount of misinformation and general gibbering!

I couldn't agree more with your points, the one thing that stood out most glaringly in the article was the one you pick on too:
"""
One of the key breakthroughs is an "intermediate representation," a kind of lingua franca that gives the software the flexibility to translate from one platform to another.
"""

So one of the breakthroughs is to use a bytecode in the same way that Java has since day 1. That's so not a breakthrough. The breakthrough is to get it _fast_. The claimed 80% is very impressive, unless that's emulating x86/windows on x86/linux, in which case it's crap.

FP.

Re:Not Vapor and not the arrival of Christ

[chawly]

OK, OK - if you're tellin' me I'll go along. But did Christ have a little something to do with this on his last visit here below ? If this works it may even have been His Father ....

Re:Not Vapor and not the arrival of Christ

[LSD-OBS]

What people seem to keep forgetting is that emulating the processor is approximately 10% of the job. There's all the rest of the hardware to emulate! The CPU is the EASY part.

Not vapor

[BoldAC]

If it's going to be released today, I am assuming it is not vapor...

Re:Not vapor

[Davak]

Yeah, obviously! :)

QuickTransit fully supports accelerated 3-D graphics and about 80 percent computational performance on the main processor. It requires no user intervention: It kicks in automatically when a non-native application is launched.

It sounds like it is software that translates one machine language to another? Pretty sweet idea!

It will still have some java-ish problems with each different form of hardware needing a unique version to translate. And then updating each of those versions as each change in the operating systems occur, etc.

Davak

Re:Not vapor

[Mysticalfruit]

I'll be a believer when I can download a demo and try it for myself.

Otherwise, I'll have to default to my age old motto that harkens from the BBS days.

"In God(s) we trust, all others... voice verify"

Re:Not vapor

[RetiredMidn]

It sounds like it is software that translates one machine language to another? Pretty sweet idea!

There was a translation solution in place back in the early 90's: Apple was working with a company called Echo Logic (probably not in existence today; please don't /. the logical URL!), a spin-off of Bell Labs, that could convert 68K binaries to PowerPC as an approach to migration to PowerPC.

I worked with them for a while to see if we could port our application (which would have required tons of work to re-compile for PowerPC); the technology was impressive, but aspects of our code gave it fits (trap patching, and dispatch tables that were effective self-modifying code).

The EL technology identified code blocks in the binary, built an intermediate representation of all the effects of each code block, and translated it back to binaries in the target architecture. Theoretically feasible, but computationally very expensive. In some test cases, the translated code was in fact more efficient from the original, because the software was able to detect unused output of a code block, and re-code the block to eliminate the unused "side-effects."

Ten+ years later, maybe somebody has more of the gnarly problems worked out. But I would bet there are issues that can't be solved with technology; i.e., the binary software on the "source" system. Presumably you can find and translate the system binaries to build a translated app, but wouldn't this constitute "reverse engineering" that most software licenses prohibit?

Re:Not vapor

[julesh]

Presumably you can find and translate the system binaries to build a translated app, but wouldn't this constitute "reverse engineering" that most software licenses prohibit?

The right to reverse engineer software for the purposes of enabling interoperability (which this must qualify for) is legally protected in the EU and many other regions. A lot of us have the freedom to ignore that aspect of our EULAs.

Re:Not vapor

[hyc]

There's nothing special about this technology at all. It's just a standard compiler whose source language is object code. I wrote the same thing for my Atari ST back in 1987, except it took DOS executables as input and spit out Atari TOS programs as output. In those days, the resulting Atari program always ran faster than the original DOS code. Besides the machine language parser you just needed a memory map and interrupt map, all of which were pretty simple back then.

Re:Not vapor

[nusuth]

It sounds a lot like what UAE, the Amiga emulator, has been doing for ages. Identifying blocks of code, and translating all of them at once is the only way to get decent emulation speed. AFAIK the original JIT compiler author never got around to write an optimizer for blocks of code (since the emulator was already faster than any AMiga ever built) ut it should not be too hard. IOW, the EL technolgy you speak of is far from unique, it even made its way into a open source emulator some time in the early 2000s.

Re:Not vapor

[Anonymous Coward]

Don't forget the Alpha program that would re-compile regular win32 binaries (and all associated libraries) into Alpha binaries, then allow you to run regular windoze stuff. I'm sure it existed for NT 3.5 and maybe 4?

I'll believe it ...

[Anonymous Coward]

when I see it.

Remember...

[Steve G Swine]

... if nothing runs at all, everything runs equally well.

Re:Remember...

[hype7]

very true :)

take a look at what they're demonstrating, too. Linux Quake 3 on a Powerbook... and Linux GIMP on a Windows machine. These aren't really things that can't be done already today.... but that may be just that the article doesn't go into a lot of depth. Show me Windows Quake 3 running on a Powerbook, now that would be something a little more impressive.

It will be interesting to see the software in any case, and see whether it really does live up to the promise. Because if it does, they're right, it's comp sci's equivalent of turning base metal into gold.

-- james

Re:Remember...

[Arker]

Show me Windows Quake 3 running on a Powerbook, now that would be something a little more impressive.

Actually, not. Emulating x86 on a PPC chip is easy.

What would be truly impressive would be running, say, Wolfenstein3d Mac on an x86 box, with reasonable speed. That would be far more difficult.

Reading the article, it sounds like a lot of hype, and I suspect the product behind it, even if it's pretty well done, will never live up to the hype.

Re:Remember...

[SkywalkerOS8]

Windows Quake 3 on a Powerbook is still also something that can be done today with VirtualPC. Show me Mac OS X on a Wintel box, that's what hasn't been done.

Re:Remember...

[Erwos]

Do a Google search for PearPC. People have most certainly gotten MacOS X running on their x86 boxes.

-Erwos

Re:Remember...

[aldousd666]

that's not done by translating stuff, that's actually emulating another processor... Vitrual PC is not the same thing here. Same goes for vmware.

This product claims to translate the code block by block into native code, thusly not emulating another processor per se. Metaphysically we could say that the two are equivelent, but it's not built around the fake processor as a unit like the other popular emulators are.

Re:Remember...

[Zorilla]

What would be truly impressive would be running, say, Wolfenstein3d Mac on an x86 box, with reasonable speed. That would be far more difficult.

Thanks to Basilisk II, I think I'm on level 20 on the Mac version of Wolfenstein 3D while I play it on my x86. The JIT compiler seems to work right in the Windows version, so I get a decent speed (about 30-40 fps) in the game. It crashes every once in a while, but the Linux version seems to be better behaved, although much slower since the JIT compiler doesn't behave well with the assembly that draws the walls in the game.

Re:Remember...

[SkywalkerOS8]

I'm aware that VirtualPC is an emulator, I meant that demoing Windows on a Mac is something that can be accomplished without writing a line of code. If this is truely not vaporware, I'd want to see an interactive demo of something that hadn't yet been accomplished such as Mac on Windows (though I see, from another poster, that too has been nearly accomplished).

Re:Remember...

[JPM NICK]

here is the link to wired.com talking about exactly this Wired.com

Re:Remember...

[damiangerous]

Show me Mac OS X on a Wintel box, that's what hasn't been done.

Here you go. Painfully slow though.

Re:Remember...

[matthewg42]

Similarly, my sig... :)

Re:Remember...

[cymen]

I agree... I loaded up PearPC on my AMD 64 3200+ CPU and with v0.4-something, it can run MacOS X at 1024x768 without too much lag. PearPC is impressive but the slowdown is massive.

Re:Remember...

[CmdrTHAC0]

This product claims to translate the code block by block into native code...

Nothing new to see here, then. Psyco, Transmeta, and valgrind all operate on variations on this same theme (dynamic recompiling).

Re:Remember...

[CritterNYC]

Do a Google search for PearPC. People have most certainly gotten MacOS X running on their x86 boxes.

I have. PearPC has been great for firing up Safari and Mac IE on my AMD 64 3200+ box to test websites. It's no speed demon, though. It's sluggish even on my system. The PearPC folks state that performance is roughly 1/40th that of native.

Re:Remember...

[Anonymous Coward]

This product claims to translate the code block by block into native code, thusly not emulating another processor per se. Metaphysically we could say that the two are equivelent, but it's not built around the fake processor as a unit like the other popular emulators are.

Uh... that looks to me like an attempt to explain to a layman what JIT is. Most emulators of modern systems use similar technology; it's been standard practice since the late 90s (remember UltraHLE, the first usable N64 emulator?).

In other words, it sounds suspiciously like this is doing exactly what other popular emulators are doing.

Re:Remember...

[Anonym0us Cow Herd]

But Basillisk II emulates the 68K processor, not the PowerPC processor.

I think what the ancestor post was trying to say was that demoing a PowerPC emulator running on x86, at indecent speed, would be far less unimpressive.

Re:Remember...

[Saeger]

The PearPC folks state that performance is roughly 1/40th that of native.

Sounds about right. It took over 8 hours to do a minimal MacOSX 10.3 install on my 1.2Ghz athlon system (running SuSE9.1), and it takes about 5 minutes just to boot it up. Still, it's great for testing Safari compat even at a snails pace.

--

Re:Remember...

[Anonym0us Cow Herd]

Nevermind.

Instead of Basillisk II, try PearPC .

Re:Remember...

[Trejkaz]

I would rather see even just OSX running on an x86 box at reasonable speed first.

Re:Remember...

[drinkypoo]

It has been accomplished, it's just been accomplished very slowly. One expects PearPC on hammer-core to do better than PearPC on x86. Seppel says that register starvation is not a real problem (though I don't see how that can be, what with some x86 instructions requiring that data be in specific registers and all) but there are some real issues that make it hard to do this stuff. Hopefully some of the problems can be solved through the use of SSE/SSE2/3dnow! in the future, along with altivec support being added. Anyway read this from the above link:

rgoodwin* Seppel: so in terms of performance gain, you say mmu, jitc timing/bookkeeping code, and FPU (most gain to least)

MMU emulation is not there yet. The timing problem is, well, a problem. And the FPU is not yet implemented using the FPU, I guess. OSX seems to do a lot of floating point math (probably related to the gui, I wonder if some of that will go away when altivec is implemented?) and one of the issues raised in the above link which I found from the pearpc wiki is that the x86 fpu is stack-based, while the powerpc is register based.

Re:Remember...

[Krach42]

--- author of AltiVec for PearPC

Yes, it's very slow. VERY VERY slow.

But remember, PearPC is a total hardware system emulator. It can't take any "speed hacks" that don't actually keep all the same side-effects. PearPC is much more compatible, and is responsible for emulating every feature of the hardware.

Read this well: PearPC is slow because it must emulate the hardware, while this project (and SoftPear) are merely emulating the behavior of the program.

This program "emulates" the graphics of the process by making native calls, PearPC emulates the graphics by calling the PowerPC OS system calls, which call the PowerPC graphics driver, which writes to a framebuffer, which we use to update our window, which goes through native graphics calls, through native graphics drivers, and finally are effecting the native hardware.

PearPC does so much MORE than this project must do, that it's just plain apparent that it must be slower.

Also, PearPC does all mmu operations through software right now, which means that there's about a 20-30 times overhead for memory operations. Yes, 20-30 times overhead. Sebastian is working at this time on a HWMMU implementation that will bring this much closer to 5-10. (But since we must still emulate the full PowerPC MMU, this will still create more overhead than this program that can just translate all these MMU operations into native calls, and native operations)

Also, right now there's no graphics emulation... so it's not going to run anything that uses OpenGL quickly at all.

Vapour.

[Anonymous Coward]

Seriously.

Let me be the first to welcome...

[bobetov]

... our vaporware overlords!

No f'n way this is legit.

Games Games Games

[Gr8Apes]

If true - we'd have any game worth playing on Linux or Macs, and life would be good, most likely, too good to be true.... :(

Re:Games Games Games

[Anonymous Coward]

There is NO F-IN way that BIG BUSINESS would have EVER let this even GET to the starting gate, much less be this close to release.

Think of it....

This would put a date-of-expiration on MS's BILLIONS.

I predict:

1 - Windows will team up with hardware vendors (Dell, HP, IBM) and give them incentives to make this NOT work
2 - Office prices will skyrocket (don't know how that's possible, but it will to prevent this) UNLESS you have a valid Windows license

There are so many GREAT things that can happen with this, but I predict that too many politicians and big-business big wigs are wise to what this will do and will NEVER let this happen.

Just my .02.

Re:Games Games Games

[BoldAC]

In demonstrations to press and analysts, the company has shown a graphically demanding game -- a Linux version of Quake III -- running on an Apple PowerBook.

Quake, linux, and Apple--all at one time. It's like a slashdot user's nirvana.

QuickTransit fully supports accelerated 3-D graphics and about 80 percent computational performance on the main processor. It requires no user intervention: It kicks in automatically when a non-native application is launched.

20% performance hit is not all that painful even... especially the way that CPUs are ramping up all the time! I'll keep my fingers crossed.

Re:Games Games Games

[BobTheAtheist]

Do you think MS honestly cares if you ran Windows on a different processor? You still do or don't pay MS for Windows depending on your idealogy but I can see why MS would want to stop this.

Re:Games Games Games

[nkh]

and maybe we could install WindowsXP on Linux and it would be fast and reliable!

Re:Games Games Games

[Anonymous Coward]

I'm skeptical about its ability to use 3D acceleration. Wouldn't it have to translate the GPU calls as well as the CPU calls?

For example, a lot of PC games use DirectX for manipulating the gpu. The programmers make calls to directx functions and directx manipulates whatever video card is installed. It seems conceivable that there could be video cards in Apple machines that are not compatible with directx and, therefore, the directx manipulation of the gpu wouldn't really be translatable. Some development studios use OpenGL for their gpu manipulation and those calls would probably translate, but opengl users are still a minority among game developers. Am I missing something?

Re:Games Games Games

[Negatyfus]

From the article:

In demonstrations to press and analysts, the company has shown a graphically demanding game -- a Linux version of Quake III -- running on an Apple PowerBook.

If their claims were really as true as they say, they would have been brave and they would have chosen Doom III or something like that. Quake III on a Mac-- not so very impressive.

Re:Games Games Games

[illumina us]

Yes, however, the DirectX libraries and binaries would also be executable on those systems if the program works in the way it is claimed to work. Moreover, the hardware requirements for the game are still the same. If you don't have a DirectX compatible card you can't expect to play the game on a PC.

Either way, the top two video card manufacturers make all their cards DX compliant and most of Apple's new systems are running ATi Radeon cards.

Re:Games Games Games

[ChairmanMeow]

Fast and reliable Windows XP? Now [i]that[/i] is vaporware!

Re:Games Games Games

[ChairmanMeow]

Damn... Always use preview button.

And always wait for two minutes to be up.

Re:Games Games Games

[aldousd666]

yes you are. DirectX and OpenGL are both abstractions from the hardware layer, at which the GPU itself performs. The DRI, for example, in linux does not emulate direct X, but rather performs hardware level commands. via it's own api -- which openGL and other implementations of GL have actually tied into. Comparing Direct X to OpenGL is like comparing C++ to Pascal. They are different languages that are both translated into a common instruction set in the end.

Re:Games Games Games

[Anonymous Coward]

Hey man, this is ./, not your average bb forum. "Our code runs longer than any other. And we didn't touch it since '98".

Re:Games Games Games

[gl4ss]

it's not the machine code that's the trouble with playing windows games on x86 linux machines, rather the problem lies in the supporting libraries(d3d & others).

Re:Games Games Games

[eam]

This isn't about running windows on a different processor. It is about running applications for windows in a different operating system with different hardware.

MS might care about that.

Re:Games Games Games

[kubrick]

Graphics cards in Powerbooks aren't that hot (I should know, I own one) and iD aren't sure whether they should even release a *native* Doom 3 for Mac due to the low standard of cards as shipped on all but the most expensive of machines.

I have my doubts about this "universal emulator" too, but let's not set the bar too high.

Re:Games Games Games

[Moraelin]

20% performance is actually in the "too good to be true" range IMHO. I certainly wouldn't even begin to mind it.

I would be interested, though, if they implicitly meant on "nearly identical hardware". No, I don't just mean speed.

But for example the PC and the Mac are nearly identical architectures (if with different CPUs), and with identical graphics chips. The PC and the Playstation 2 are not. Even when emulating the old Playstation, you run into the problem that Sony's graphics chips aren't actually even vaguely similar to the direction that PC graphics chips took.

You can most certainly still emulate that on PC style hardware, but the work involved is non-trivial enough to cost more than 20% speed. It doesn't have to be something slow per se, just cost more than just piping the vertexes to the GPU via DMA. Even something as trivial as having to reverse the endianness of the vertex coordinates, can put a dent on emulation speed, and hardware can easily be even more different than that.

Basically, don't get me wrong, I still want a Mac which can play all PC games, so that would certainly be enough for me.

But I just doubt the claim that it can emulate _any_ hardware on _any_ other hardware at 80% CPU efficiency and nearly 100% graphics efficiency. On some hardware combinations that's just simply physically impossible.

Re:Games Games Games

[Durandal64]

Graphics cards in Powerbooks aren't that hot (I should know, I own one) and iD aren't sure whether they should even release a *native* Doom 3 for Mac due to the low standard of cards as shipped on all but the most expensive of machines.

You must be on crack. id most certainly are porting Doom 3 to Mac OS X. It was first demoed on a Mac, after all.

Re:Games Games Games

[Anonymous Coward]

Hey, I read it as italics. My first HTML editor was notepad, so simple slashcode doesn't have to work for me to "see" the result. Kind of like Neo, but not quite.

Re:Games Games Games

[Anonymous Coward]

What are you talking about? HTML does not use square brackets.

Re:Games Games Games

[Minna Kirai]

It is about running applications for windows in a different operating system with different hardware.

Wrong. This doesn't let you run Windows apps on OS X. It provides CPU compatibility only. OS compatibility is up to you- meaning the Mac user will need to install a copy of Win XP before running those games.

In the short term, Microsoft won't mind that, because they still get a sale. But long term, it will hurt them, as the increased flexibility allows users to migrate to Mac without needing to replace all their apps at once.

Re:Games Games Games

[kubrick]

From an iD rep at QuakeCon 2004:

A Mac "gamer" asked about the port to OS X. Apparently there is no current time for the release of a port. The game runs, but there is a lot of optimization, and currently they feel the Mac platform can not yet offer the same experience as the PC. Activision will not publish the Mac version of Doom 3. There is no publisher set currently.

although there's a slightly more optimistic version later on:

The answer: Tim Willits, lead designer. He's also the one handling the Linux port. He said the Mac port was playable, but was still in need of optimizations, especially for the lower-end systems. He said he was working directly with Apple to address the issue.

What was most intriguing of all was that they stated that the Mac version would NOT be published by Activision as previously assumed. Instead id would have to shop around for a new publisher. The bright side of this is that they MUST be putting out a retail version, otherwise they wouldn't be needing a publisher at all.

from here.

It's the first statement which got the most oxygen at the time, understandably. But just because they have a current port doesn't mean it makes commercial sense to release it for a large company. The Linux version is in the same boat with regards to a commercial release...

Re:Games Games Games

[Anonymous Coward]

Buggered up the spacing, sorry... if only /. would do the same thing when posting it does when it previews. (Looks different again when replying to it, wassupwidat?)

- kubrick

Re:Games Games Games

[Elwood P Dowd]

Supposedly they map the emulated APIs to host system APIs. They imply that you can run Wintel apps on OS X without Windows or an Intel processor.

In other words, they are either big fat liars, or they have silently employed more software engineers than Apple & Microsoft's OS teams combined. And they're stealing WINE.

Re:Games Games Games

[Minna Kirai]

Supposedly they map the emulated APIs to host system APIs.

Yep.

They imply that you can run Wintel apps on OS X without Windows or an Intel processor.

Nope. Quoting from their announcement: "Operating System Mapper. Dynamite supports operating system mapping between any two Unix/Linux-like operating systems, as well as mapping between mainframe and any Unix/Linux-like operating systems."

Re:Games Games Games

[Trejkaz]

I don't know how it's possible, no... $400 for a word processor seems a little steep already. :-)

Re:Games Games Games

[ChairmanMeow]

This is /. not ./ and what occurred above is what happens when you have a forum in one tab, and /. in another...

Re:Games Games Games

[Durandal64]

Hollensted made it quite clear that an OS X port is in the works and will be shipping.

Mac and Linux: Unfortunately I don't have dates for either of these. However, Linux binaries will be available very soon after the PC game hits store shelves. There are no plans for boxed Linux games. More remains to be done for the OSX version of DOOM 3 and that will take some time. We won't release the OSX version until it's just as polished as the PC version. The date for OSX DOOM 3 remains "when it's done", but I can confirm that it's definitely coming.

Or Timothee Besset's quote that ...

"Since gold, the Mac port was definitely the big thing; it benefits both the Linux and the Mac version, but Mac had to come first."

Re:Games Games Games

[kubrick]

We won't release the OSX version until it's just as polished as the PC version.

Waiting for the installed hardware base to catch up, maybe? :)

Re:Games Games Games

[Durandal64]

Well polish of the software really has nothing to do with the installed hardware base. They're probably optimizing the shit out of it though to get some sort of playability on older machines.

Re:Games Games Games

[kubrick]

Older machines? Apple are currently selling machines with video cards that will struggle with Doom 3, no matter the amount of optimisation. Older machines will struggle even more.

I really like Apple's machines, but I think they should offer more (and higher-powered) video card options than they currently do. Otherwise, what was the point of changing to industry standard components in the first place?

Re:Games Games Games

[kubrick]

(I'm referring here to machines that can't have their video cards swapped out, of course.)

P4 Hack anyone?

[XMyth]

Hehe.....

no performance hit?

[Anonymous Coward]

Yeah right, that's not possible, just translating to another set of instructions takes some of the cpu's resources... that alone would debunk the claim of no performance hit...

Re:no performance hit?

[AKAImBatman]

that's not possible, just translating to another set of instructions takes some of the cpu's resources

According to TFA, this is a pre-compiler/translator, not an emulator. i.e. The entire program is recompiled for another platform using only the binary data as the source. This is theoretically possible and has been attempted many times, but such compilers often trip over levels of indirection that programmers add.

For example, a programmer might place the video address in a variable, then reference that for screen paints. Such a trick would be impossible to detect at compile time, and would only be properly handled by a true emulator.

Re:no performance hit?

[sketerpot]

How about, "no performance hit on the order of what you get with CPU emulation? CPUs are generally fast enough that you can accept a relatively small performance hit on all but the most demanding programs.

Re:no performance hit?

[Firehawke]

Unless it's a static or dynamic recompilation technique-- it could translate before execution, dumping a new binary which it executes. You'd have a much longer start time, obviously, but it'd run at the full speed possible. Assuming, of course, that your recompilation techniques are 100% perfect.

Doubtful, but possible.

Re:No performance hit?

[Tyndmyr]

In such a case, developers everywhere will be immensely grateful. However, I still think its vaporware.

Re:no performance hit?

[beh]

Isn't this something that DEC tried in the early Alpha days? I vaguely seem to remember that Vobis, a German computer reseller, was including a tool claiming to do this with every Alpha they sold.

Though I have no clue how well this tool worked...

Re:No performance hit?

[photon317]

Actually, I could conceive a brilliant software engineer coming up with a universal translation mechanism that turns x86 assembler into functionally equivalent PowerPC assembler, or vice-versa, or to other platforms. I believe IBM had been funding research in thsi area for quite some years now.

What sets off the BS detector for me is the APIs. They consistently state that they can do this for any OS. You have to do API translation, and you'd have to do that per OS, and it's a staggering volume of work to get all the APIs translated (think Wine project, just trying to do windows->linux api on a single shared hardware platform). When my linux binary calls any given kernel, C library, or even other common library (readline?, pthreads?, opengl?, etc..), those calls all have to be translated to equivalent MacOS or Windows API calls.

Re:no performance hit?

[wertarbyte]

According to TFA, this is a pre-compiler/translator, not an emulator. i.e. The entire program is recompiled for another platform using only the binary data as the source. This is theoretically possible and has been attempted many times, but such compilers often trip over levels of indirection that programmers add.

So it's more or less a cross-decompiler with a compiler attached. This could perhaps convert from one CPU to another, but what about system libraries? This part is probably the most difficult, considering how much effort is put into the wine project.

Re:no performance hit?

[khrtt]

For example, a programmer might place the video address in a variable, then reference that for screen paints.

I don't see how that could be a problem. There is a serious problem with self-modifying code and code generated on-the-fly. Basically, every time anything crosses the data/code boundary, it would have to be translated. A simple translator would only handle the data/code boundary that the program crosses when it's first loaded into memory from an executible file.

Then, there are thunks (in Win16/Win32 programming; small pieces of code generated dynamically to help call Win16 libraries from Win32 code and vice-versa). Then there are compressed executables. Encrypted executables for copy-protected games. Dynamically generated code for inner loops for high-performance computations, maybe. Code generated by applications that compile on-the-fly, like your average LISP system., or JVM. A binary translator would not be able to handle any of those, unless you try to detect a jump to data somehow, which may be prohibitively expensive on a PowerPC. Is it?

Re:No performance hit?

[Gadzinka]

HP did some time ago some research into JIT compilers and they managed to run PA-RISC programs with JIT on PA-RISC processor faster than natively. The answer is loop unrolling, cache locality, linear execution etc.

Dig in Ars Technica for more info, they covered the topic in detail couple of years ago.

Robert

Re:No performance hit?

[Q2Serpent]

No - only the kernel system calls need to be translated. Since all of the other libraries (libc, opengl) sit on the kernel, they can just be copied to the new environment like the program was.

Of course, this won't help in the speed department...

Re:no performance hit?

[LifesABeach]

Considerig the nature of machine instructions, 'Cross Compiling' seems pretty straight forward.

I'm thinking that a call to an Operating System routine, lets say to access a Hard Drive, does not have the same address as would be found on another Operating System.

Re:No performance hit?

[mrseigen]

This link? Seems pretty interesting. Ars is full of all kinds of fun stuff.

Re:no performance hit?

[tazan]

I don't see a problem, they just need to put a sticker on the box with minimum system requirements in very small print. I think 30 ghz would be about right.

Re:no performance hit?

[fitten]

There were several. One was the translation of VAX code to Alpha to run on Alpha-VMS. The other was FX!32 that translated x86 Windows binaries to Alpha binaries.

FX!32 worked most of the time but it rarely gave the expected speedup we expected.

Re:no performance hit?

[pipacs]

This could perhaps convert from one CPU to another, but what about system libraries?

No need to deal with system libraries if you are converting the whole system: CPU translation + virtualization of some common hardware, a la VMware.

Re:No performance hit?

[julesh]

What sets off the BS detector for me is the APIs. They consistently state that they can do this for any OS.

Actually, it seems to be primarily other people saying this. Read their site; they only claim to be able to do it for UNIX-like systems.

Re:no performance hit?

[wertarbyte]

JVMWare? strange...

Re:No performance hit?

[back_pages]

Actually, I could conceive a brilliant software engineer coming up with a universal translation mechanism that turns x86 assembler into functionally equivalent PowerPC assembler, or vice-versa, or to other platforms. I believe IBM had been funding research in thsi area for quite some years now.

And there are more than a few published but not issued patents for exactly this. And when you're talking about x86 to PowerPC, you probably aren't insane. When you're talking x86 to and from some crazy ass systolic array processor, then you're out of your gourd.

Just keep an eye on those published-yet-unissued PGPubs. Some people would have you believe they can translate machine code for systolic array processors into machine code for your PDA.

In theory, it could be done. Just map your architecture back to a Turing machine and map a Turing machine to the new architecture. In practice, you're out of your gourd. Try performing a single floating point division on a Turing machine and see how many steps it takes... now try doing a 1,000,000 x 1,000,000 matrix inversion.

So many people in this field talk about Mac to/from PC conversions like they've solved every problem in the universe. All things considered, those are very similar architectures compared to some of the more exotic chips from the last 30 years.

Re:No performance hit?

[back_pages]

few published but not issued patents

Eh, just to be perfectly clear, I'm talking about published patent applications, not patents. You can search for them here: http://patft.uspto.gov/netahtml/search-adv.htm

Re:No performance hit?

[Anonymous Coward]

Actually, I could conceive a brilliant software engineer coming up with a universal translation mechanism that turns x86 assembler into functionally equivalent PowerPC assembler, or vice-versa, or to other platforms. I believe IBM had been funding research in thsi area for quite some years now.

You still need some emulation, at some level, because it's always possible for the program to be self-referential and self-modifying. While "good" code shouldn't do that, plenty of copy protection code (Ugh) and older software contains such nastiness. The problem is that when you directly translate the code, it changes the binary representation, and also makes self modifying code break. Re-translation is needed for every modification, which is essentially emulation, at best, but much slower for those sections. It's not unworkable, but it would introduce odd delays and latencies in the execution of the program that might break the code.

Re:No performance hit?

[photon317]

It's not that simple, which is why other projects like Wine haven't taken that approach. Ultimately some software ends up touching things in a hardware-specific way that the kernel does not abstract for you (it merely provides a pass-through interfaces for accessing some hardware in these cases), in which case the library doing so has to be translated as well.

Re:No performance hit?

[photon317]

You're right of course, but I bet if you did as a block-level jit thing and kept the original asm code around in memory too, you could work it all out. Sounds like something Transmeta probably had to work through in their translating microcode I'm sure.

Sod you ID

[Anonymous Coward]

Keep your Linux client binaries!

Any program?

[vistic]

Wouldn't you still need a bunch of supporting files and APIs to run a Mac program on Windows, vice versa, and for other operating systems? Programs make specific calls to the operating system, like windowing toolkits... this emulator must be huge to ensure everything works and they must have done massive successful reverse engineering of closed source files in the Windows architecture.

Re:Any program?

[BobTheAtheist]

Run the whole OS... The device drivers will just work by magic

Re:Any program?

[little_blaine]

Well said. I can see a program that does on-the-fly translation of assembly code, but the first time you try to access a windows .dll on a mac, or a linux .so on windows (for example), or make any kind of system call on a foreign platform, you will hit problems.

Now here's an interesting thought: MacOS X on x86. Or windows on PowerPC.

Re:Any program?

[skraps]

Let's see.. I want to word this carefully. After careful review of the claims, I have determined:

VAPROWARE!

VAPROWARE! VAPROWARE!

Re:Any program?

[GileadGreene]

Did you mean VAPORWARE? Perhaps you should have taken some more time on your careful wording... :)

Re:Any program?

[skraps]

Ooops. :-)

Re:Any program?

[BobTheAtheist]

He just translated it to another language on the fly without most people even noticing. You must be a power user.

Re:Any program?

[vistic]

I can run Windows on PowerPC... using Microsoft (ex-Connectix) VirtualPC. However, I think that emulating PowerPC architecture on x86 doesn't work as well as x86 on PowerPC... I guess because "complex" CISC x86 assembly instructions can be reduced to RISC PowerPC instructions... whereas PowerPC instructions (I assume) exist in x86, however the x86 is not optimized to run that way... x86 probably assumes you use the more complex instructions available like MMX stuff.

Re:Any program?

[russint]

Well, it's not like nobody's seen the windows source code. So its certainly not impossible.

Re:Any program?

[Anonymous Coward]

Check out the tech report (http://www.transitive.com/technology.htm). It only does "mapping between any two Unix/Linux-like operating systems, as well as mapping between mainframe and any Unix/Linux-like operating systems." It sounds like someone exagerated something somewhere along the line...

Re:Any program?

[strider44]

nope vapourware :P

Re:Any program?

[CableModemSniper]

I gots two words for ya: "register starvation"

Re:Any program?

[Anonymous Coward]

Now here's an interesting thought: MacOS X on x86.

PearPC Community Page
PearPC Sourceforge Page

Re:Any program?

[logic hack]

Or windows on PowerPC.

I see the Mac ejecting the defiled hard disk, sprouting a pair of robotic legs, and running home to Steve crying.

Re:Any program?

[damiam]

It doesn't work well but it's still possible.

Re:Any program?

[gt623]

i think i remember reading that the xbox 2 development kits use powermacs with windows nt ported to powerpc. and mac os x was ported to x86 on a collaborative project with intel until jobs pulled the plug.

Re:Any program?

[Durandal64]

I really wish this RISC/CISC myth would just die. Just about every desktop processor nowadays is a RISC/CISC hybrid, including the Pentium 4 and PowerPC 970.

Re:Any program?

[claytongulick]

From the website:

QuickTransit for x86. Allows application binaries compiled for a MIPS®, POWER(TM), PowerPC(TM) or mainframe processor to run on an x86-based computer. Application binaries compiled for other processors will be supported soon. Operating system call mapping from any Unix/Linux-like operating system or any mainframe operating system to any Unix/Linux-like operating system is supported.

I don't see any mention of Windows...

Re:Any program?

[vistic]

How is a Pentium 4 RISC?

Re:Any program?

[b1scuit]

"Or windows on PowerPC."

That kind of thinking brought us the v6 Mustang and 'fitness water'. Just because you can...

Re:Any program?

[Anonymous Coward]

Don't worry, we wish you would just die too...

anyway CISC vs. RISC may not be what it used to, but the intel type chips are still far less efficient and have a messier system with far more addressing modes, etc. to support, so it is still "complicated" vs. "simplified".

Re:Any program?

[snuf23]

The Pentium 4 (and other Intel processors) breaks down the larger x86 instructions into small RISC like components known as micro-ops which are then processed by the cpu.
There is some information here:
PC Processor Microarchitecture

Re:Any program?

[Chris Burke]

Every x86 processor since the Pentium Pro (possibly the AMD K5, not sure, but regardless...) has had x86 decoders that translate the x86 CISC instructions into much simpler RISC-style micro-ops that the rest of the machine deals with. The typical case is that the decoders can directly translate simple instructions (e.g. add eax, [rsp + 10] -> a load micro-op followed by an add micro-op), and refer to a microcode ROM to get the sequence of RISC ops for more complicated instructions like FXRSTOR or INT.

In the case of the Pentium 4, the x86 decoders only operate on instruction data coming from the L2. The trace cache on the P4 caches RISCy micro-ops, which is the main benefit of the trace cache on the P4 -- skipping the relatively lengthy decode of x86 instructions.

So from a micro-architecture standpoint there is very little difference between CISC and RISC, since you just translate CISC->RISC. x86 still manages to be a big PITA because of weird things like shifts by zero which don't effect the flags.

CISC vs RISC is basically a done deal, with RISC "winning" along with it essentially not mattering anymore.

Re:Any program?

[Fnord]

Regardless of the irrelevance of RISC vs. CISC, it is easier to emulate an x86 on PowerPC, but only because PowerPC has more general purpose registers.

Re:Any program?

[drinkypoo]

Yes, including the K5, AMD's first RISC processor. their 386 is pretty much the same as intel's. Their 486, which I understand is not actually the same, nonetheless is the same sort of architecture. It's not until the pentium raised the bar for x86-compatible processors to include stuff like superscalar processing that x86 processor design diverged considerably, and the K5 was supposed to combat the P54C (original Pentium.) Then the K6 came out to defeat it, then the K6/2 was aimed at the Pentium MMX, and finally the K6/3 was supposed to do battle with the Pentium 2. Actually, the K6/3 is a good enough processor to do that (though I doubt it has the fp math performance of a fast P2) in that it had high clock rates and pretty good 3D acceleration (3dnow!) but it was too late and no one took them seriously, so no one bought K6/3s (except a lot of them have gone into laptops) and AMD had to whip out the athlon to avoid sliding into the sea and being purchased by some other company. Which they did, and now they're a giant.

Re:Any program?

[Krach42]

Well, the big problem with modern CISC->RISC chips (to call them what they really are) is that you spend a lot of time figuring out what to do, rather than actually doing anything useful.

So, you have this HUGE chunck of silicon wasting energy, and making heat, just to figure out what it ought to be doing in the first place.

This is why I hate CISC right now. Not because it isn't fast (it is), but because it's just totally ill-thoughtout in every way. It's so ad-hoc, and almost random that I feel I'd be much happier using an architecture that was designed from a modern standpoint, such that it can spend as much silicon as possible doing what it should be doing, rather than figuring out what it ought to be doing.

Re:Any program?

[Durandal64]

Oh fucking blow me. CISC v. RISC is about two architectures whose distinction is no longer relevant. If you want to talk about "complex" vs "simplified," that's a separate issue.

Re:Any program?

[vistic]

I bought a K6-III 450... it was the first ever PC I built. I did a lot of research and settled on an EPoX board (MVP3G-M). It was a wonderful system for the time.

ooooooh, yawn!

[mrpuffypants]

In demonstrations to press and analysts, the company has shown a graphically demanding game -- a Linux version of Quake III -- running on an Apple PowerBook.

I'm sorry, but that's just not impressing me. Not to mention that there's already a native Mac OS 9/X port of Quake III, but it's not even the most system-dependent code that I can think of.

When I can run Office 2003 natively inside Linux then we can talk.

Re:ooooooh, yawn!

[swillden]

When I can run Office 2003 natively inside Linux then we can talk.

Define 'natively'. Because Crossover Office can run Office 2003 on Linux just fine, today.

Re:ooooooh, yawn!

[isolation]

If you can get it to install it will run but I dont know how well. At this time we dont support Office 2003 but we are working on it.

Re:ooooooh, yawn!

[Otter]

I'm not sure if this was his point but -- Wine is certainly at the "then we can talk" stage. A universal Wine would definitely be of interest, although you probably wouldn't want to run a hospital on it just yet.

Re:ooooooh, yawn!

[gbjbaanb]

it will never run 'natively'. What you'll be able to do is run Windows 'natively' (in a window no doubt), and run office2003 on that.

The big trick is getting windows to run on your linux OS first - there is more to running windows applications than just translating/emuating/whatever the binary as they all use OS features.

Just like in windows when I run the Gimp I need the GTK framework, in linux, when I run Office 2003 I need a bunch of COM, IE, and WinAPI calls.

Re:ooooooh, yawn!

[Anonymous Coward]

When they can get OpenOffice to run natively on Linux I might considering switching to Linux!

Re:ooooooh, yawn!

[Anonymous Coward]

He meant natively, as in a native port, not in some crappy API emulator.

Re:ooooooh, yawn!

[pknoll]

Define 'natively'. Because Crossover Office can run Office 2003 on Linux just fine, today.

I'd define "native" as "not needing Crossover Office."

For example, MS Office 2003 runs natively on Mac OS X, if you buy the Mac version. Admittedly, what CXO gives you is close to native performance, but it's still another software package that's allowing it to work, which is non-native, as far as I'm concerned.

Re:ooooooh, yawn!

[julesh]

When I can run Office 2003 natively inside Linux then we can talk.

You're posting on the wrong article. This software is only useful for running UNIX-ish applications, according to the linked web site. The article author must have got hold of the wrong end of the stick.

Re:ooooooh, yawn!

[Anonymous Coward]

Yes, and a lot of work went into making that possible. If they did anything remotely as good as Crossover Wine, I'd be very impressed.

There's thousands and thousands of API calls you need to emulate, which isn't even documented on the Windows side. It a lot of years work to do that.

Re:ooooooh, yawn!

[swillden]

I'd define "native" as "not needing Crossover Office."

But needing Transitive's software is okay? That's why I raised the question of the definition of "native", because it seems like under my normal definition, it would exclude Transitive's stuff, too, and under most any reasonable definition that includes Transitive's stuff, Wine would be acceptable as well.

Re:ooooooh, yawn!

[pknoll]

That's why I raised the question of the definition of "native", because it seems like under my normal definition, it would exclude Transitive's stuff, too

Yup, it would exclude Transitive, too. Sorry, I didn't mean to imply it wouldn't. I don't like stretching the definition of native too far, which to me means it stops at making direct system calls. If anything is needed between the app and the system which doesn't ship/install with the app or the system, it's not native.

For the record, I think this includes Wine, Cedega/Transgaming (same thing, really), CXO (ditto), and even *BSD's Linux ABI.

Of course, we're getting into semantics, but we are in fact discussing the definition of the word. =)

Either it's native, or it's not. This isn't.

Any processor?

[Anonymous Coward]

In Soviet Russia, software application binary runs you!

Today's Poll

[eSims]

Transitive Software:
1. Vaporware
2. Miracle
4. Coyboy Neal

Personally, I vote it's just Coyboy Neal at it again.

Re:Today's Poll

[shfted!]

What happened to option 3? Did it evaporate?

Re:Today's Poll

[ProfessionalCookie]

3. Profit?

Sorry.

hoax

[Anonymous Coward]

i call hoax. rob enderle is apparently involved.

Re:hoax

[Anonymous Coward]

Oh it isn't a hoax, this emulator is completely real. Unfortunately the emulator itself only runs in Dos 3.2

I'll be impressed if...

[Anonymous Coward]

I can run MechWarrior 4 on my Redhat 9 Dell insperon. Do that, and I will buy myself a copy today!

Like java's HotSpot?

[tunah]

They're talking about recompiling sections of critical code, like java's HotSpot. It'll be interesting to see how fast it ends up - the startup time is a pain in java, but it's pretty decent after that. I can't find a source for the "no performance hit" bit. It looks real, and quite impressive, but not exactly what the summary indicates ;-)

Re:Like java's HotSpot?

[Anonymous Coward]

"One of the key breakthroughs is performance," Wiederhold said. "You can't tell the difference between a translated application and a native application."

Re:Like java's HotSpot?

[Anonymous Coward]

Somebody wrote a while ago that JVM1.5 doesn't suffer from startup costs:
http://slashdot.org/comments.pl?sid=98264& cid=8393 524

You mean like classic Mac OS's mixed-mode?

[tepples]

Instead, it's probably something in between that performs as advertised, but mitigating factors (300MHz CPU?) result in it not being everything everyone expected.

I remember a product for the Mac called SoftWindows that broke out to native code when accessing hardware, producing somewhat of a speedup. Some versions of Classic Mac OS did the same thing during the transition from 68LC040 to PowerPC instructions.

Re:You mean like classic Mac OS's mixed-mode?

[Eight 01]

The Amiga did it first! Google the saga of Jim Drew and the Emplant board. The same types of claims were made. It could emulate the Sega Genesis, the PC and the Mac at full speed using some revalutionary Amiga-Fu.

Of course it was all B.S. then too.

Re:You mean like classic Mac OS's mixed-mode?

[Gax]

Erm, the Genesis emulator was never released and the Mac emulator was horribly buggy and was done better in software 5 years later. The PC emulator was also terribly slow IIRC. Geek points for effort, but no cigar (to mix a metaphor)

Misleading

[The Grey Clone]

The blurb posted is misleading. The Wired Article states "with almost no performance hit."

Probably vapor ware

[Anonymous Coward]

The CPU translation is nothing new, however, what about all the operating system calls? The APIs and what not? Their software would have to make Unix and Windows system calls, like Wine.

Welcome

[h00manist]

hope it works /me performs rain dance

Re:Welcome

[mattyrobinson69]

/me joins in rain dancing

no, really.

what do I think?

[Anonymous Coward]

extraordinary claims require extraordinary proof.

Taos

[mirko]

A few years ago, while I was still primarily using my Acorn ARM-based RiscPC, I remember being in contact with TAOS people, they were making an heterogeneous processor operating system on which they claimed they emulated a virtual processor on which the whole environment would run, regardless of the hardware.
So, this idea reminds me of this project...
It could still be possible, we've got Java classes instantiated and running on many architectures, after all...

Re:Taos

[Kardamon]

Somehow OT, but this reminds me of the PC and MS-DOS emulator for ARM/Archimedes back in 1987 (IIRC, it may have been 1988). My friends did not believe the Archimedes could emulate a PC before they had seen it run WordPerfect 5.

It runs on magic

[funkdid]

I'm calling bullsh!t on this one. Processor differences being what they are I don't see this working. All other attempts at this have been so-so, to make a product like this "whithout a processing hit" sounds like this was written in some new programming language, perhaps one based on Magic.

Re:It runs on magic

[no reason to be here]

sounds like this was written in some new programming language, perhaps one based on Magic.

So how much mana would I need tapped to run Photoshop in Linux?

Re:It runs on magic

[strider44]

yep, the BS needle is moving to "high"

Re:It runs on magic

[untaken_name]

WUBRG, of course.

Re:It runs on magic

[Masami Eiri]

3 blue, 2 red, 5 green, 1 white, and 3 colorless. Oh, and it has an upkeep cost of 2 black/turn.

Re:It runs on magic

[Kehvarl]

none at all, but you have to write the compiler yourself, puting the lines on seperate pages to keep it from crawling away. and the first time you attempt to model the universe, you'll probably turn into a 6 foot tall bug. Then you'll have to go steal a supercomputer that the russians went to a lot of misdirection to acquire, and while you're doing this, a pair of kids will be building giant robots to defeat anything you can throw at them.

Re:It runs on magic

[blueg3]

None. You can't tap mana, you can only tap lands (or use other effects, some of which may require tapping) in order to put mana in your mana pool.

Re:It runs on magic

[Anonymous Coward]

So how much mana would I need tapped to run Photoshop in Linux?

Hmm, good question. You might need a shot of KEI to get some apps running well. And a couple FT items... heh.

Vaporware

[g0bshiTe]

Does this mean I can finally get WINE to work under Windows?

Re:Vaporware

[Anonymous Coward]

No, but you might be able to get it to run under Linux ;)

Re:Vaporware

[u01iz]

Only if you want to run CoLinux.

Re:Vaporware

[Anonymous Coward]

Odds are not bad that you can run Wine under Windows, at least with cygwin. On the WineHQ website is a Project registered on that subject.

Just think of the possibilities: Wine on Win on Wine on Win on ... and in case of QuickTransit you can't tell the difference!

Re:Vaporware

[stoborrobots]

what if the superinterpreter can reduce 4 target machine instructions per program instruction - then you get a 75% speedup, more than enough to compensate for the interpreter's computation time...

Be careful when making rash claims about something's impossibility. Note: I don't think it could do this, but hey, if you're going to make an assumption, any assumption will do...

Re:Vaporware

[JohnPerkins]

Does anyone know what the record is for nested emulation- a computer emulating program a, which is emulating program b, which is emulating program c, etc.?

Re:Vaporware

[dasmegabyte]

Sure. Also Cygwin for Linux and Pear for Mac OS. It'll be a grand time, self emulation for all!

Re:Vaporware

[julesh]

I've run xnest inside xnest (repeat seven further times). OK, so its the same program, but does it count?

Re:Vaporware

[Anonymous Coward]

$bash
$bash
$bash
$bash
$bash
$bash ...
etc.

Re:Vaporware

[JohnPerkins]

I guess that would make two different contests then- one for any nested emulation, including duplicates if they are in fact running inside of each other. The other for the greatest number of not-same nested emulation (emulate a inside of b inside of a inside of b, etc..).

No performance hit?

[theluckyleper]

If there's no performance hit, there must not be true "emulation" going on... it would be impossible to emulate another OS and architecture without a few extra cycles!

The only way I can imagine this happening is if the software reads your executable and then does a one-time translation into a native executable. That way the native executable wouldn't be emulating anything, it would be the real deal. But... the complexity of such software would be staggering.

Here's hoping it works!

VAPOREWARE

[Anonymous Coward]

VAPOREWARE. Just like how a transmeta chip was supposed to run x86 code as fast as a dedicated x86 chip

Vaporware, Miracle, ORrrr...

[Sgt_Jake]

Gag order, lawsuit, violation of the DMCA and/or Patriot Act and theft of 'Intellectual Property'.

And SCO will sue. Think about it.

Arch or Library/API ?

[Gherald]

"Emulating" architectures and "Emulating" native OS libraries/APIs are very different things.

Is Transitive claiming to do BOT universally!? If so I am very skeptical, because even doing 1 of the 2 would be impressive.

Write once, run everywhere

[slaad]

Pfft. That's already been done.

Easy refutation

[Yartrebo]

Just take Alan Turing's original Turing Machine. It can be proven that certain algorithms, like a binary search, will take an algorithmically longer time on a Turing's machine than on your standard x86 processor.

Binary search is logarithmic time on a normal processor, but it is at least quadratic time on Turing's machine.

Therefore, I have found a counterexample to their claim.

PS: Turing's machine used an infinite tape and that tape could only be moved 1 space per cycle. Most of the time spent in the binary search will be moving the tape around.

Re:Easy refutation

[gUmbi]

It can be proven that certain algorithms, like a binary search, will take an algorithmically longer time on a Turing's machine than on your standard x86 processor.

Okay, it's Monday morning and my brain isn't floating in coffee yet...but what the hell is does 'algorithmically longer' mean??

Re:Easy refutation

[Shea_Butter]

That's faulty logic. First of all it doesn't say "no performance hit" it says "VIRTUALLY no performance hit." Not only that, but it's concerned with the software translation aspect, and some sort of idiotic storage system really isn't the fault of that software now is it? That's like saying their claim isn't true because it won't work on a computer with no RAM.

Re:Easy refutation

[gedhrel]

Your refutation is flawed; taking a step back, binary search is constant time on a normal processor, because the addressable space is finite. If you actually attempt to produce a binary search algorithm on your "normal processor" then you're going to have a hell of a hard time making its performance scale to an arbitrarily chosen number of elements.

Re:Easy refutation

[attam]

i think he means "asymptotically longer"... which means that as the size of the search space, n, goes to infinity, the time it takes to run the search approaches lg(n) for x86 and n^2 for turing's machine

Re:Easy refutation

[jjoyce]

When you speak of the computational complexity of an algorithm, you must define what constitutes a computational step. The total cost of the algorithm is then based on these steps. For example, when people say that binary search runs in O(log n) time, they mean that it takes a logarithmic number of steps (where n is the number of possible values to search through) and the computational step is a comparison of two values. What you are doing is saying that the computational step is the movement of tape, which changes the model. That does not prove that binary search's running time is no longer a logarithmic number of comparisons.

Re:Easy refutation

[Yartrebo]

The tape can be infinite, but at any time only a finite amount is used.

Also, the claim was for that this works for all processors, as if they had developed a fundamental breakthrough in comptuting. When their is that much hot air in the balloon, it doesn't take much to pop it.

Re:Easy refutation

[Yartrebo]

That is what Turing's Machine does on every step. It reads the value at the read/write head, decides what to do based on the value read and the state of the state machine. After deciding, it writes a new value on the tape at the same location, goes to a new state, and moves the tape left 1 or right 1. Turing's Machine isn't very efficient, but it is a Turing-powerful machine, making it a computer by definition.

Re:Easy refutation

[Yartrebo]

A machine with no memory is not a computer.

Also, the storage system might be idiotic, but it is a bona-fide Turing-powerful machine, making it a computer. The claim was that it runs with no substatial hit to performance on any machine. That's a very big claim, and easy to shoot down.

Re:Easy refutation

[Shea_Butter]

It was an analogy for asking something to do the impossible. Dismissing it because it doesn't work on one obscure piece of technology misses the larger picture of what it does promise to do.

Tortoises all the way down.

[Libertarian_Geek]

Sounds like an emulator equivalent of a perpetual motion machine. I can't say if this is real or vapor for certain, but it sure sets off my BS alarm.

Re:Tortoises all the way down.

[Cylix]

Danger Will Robinson! Danger!

Robby seems to think these claims are a bit outrageous.

Just think of how much work would be involved in something like this. Maybe it's a compiler and their own widget set. Emulate APIs? Not quite an emulator really. Perhaps its a java emulator? Who knows what twist is really there, but we know from past experience this is tricky stuff.

Boisterous claims are often given by boisterous men... neither of which have any solid value.

Re:Tortoises all the way down.

[gfxguy]

Surely you can't read into "no performance penalties" not to mean "on like hardware".

Re:Tortoises all the way down.

[Chess_the_cat]

I think you're the only one who has interpreted their claims that way. I don't think Wired would bother writing up a company that claimed it could run Office on a C64.

Re:Tortoises all the way down.

[TurtlesAllTheWayDown]

Er. I resen^Hmble that :/

Re:Tortoises all the way down.

[AstroDrabb]

I am sure they didn't mean it could run your application at the same speed on _slower_ hardware. Anyone with more then 3 brain cells could figure that out. Basically they are claiming on similar hardware you should not notice a speed loss.

Re:Tortoises all the way down.

[Anonymous Coward]

Wow, right over your head.

Re:Tortoises all the way down.

[Anonymous Coward]

> Surely you can't read into "no performance penalties" not to mean "on like
> hardware".

That's exactly what the guy you're replying to is suggesting. Where does the software house claim than they can enable a 300mhz pc to have more clock cycles in software? Is anyone claiming that? Bizarre. I guess that's Slashdot for you.

Re:Tortoises all the way down.

[Blitzenn]

Even with that, the feat is not possible. If one processor (let's say a CISC processor) has an instruction for a function in the target code, and the second processor (let make this one a RISC processor), does not, It will HAVE to take more clock cycles to accomplish the same task. To make open claims that there is no performance hit is simply ridiculous. The facts of how different processors handle the same or similar instructions make that impossible. I'll bet you they claim it can suck a golfball through a garden hose too.

Re:Tortoises all the way down.

[Anonymous Coward]

Any anyone with more than 4 brain cells could tell you that on "similar hardware", you will still need to translate all the OS API calls of an application built for one OS to a native API, all of the hardware specific calls to native hardware (no simple task) and possibly all of the assembly code for one CPU arch to another. All of this combined (or even parts of this) will cause a significant performance hit. Remind me not to hire any "Senior Programmers" with less than 4 brain cells. My guess is when they say and app on any platform, they mean any windows app on a x86 linux platform; something like VMWare or a re-packaged Wine.

Re:Tortoises all the way down.

[gfxguy]

Look, I don't believe the claims myself, but to deride the company by asking if you think a program running on a 3Gh Pentium will run as fast on "my 600Mhz laptop." Of course, that is not what the company is implying.

Re:Tortoises all the way down.

[Dashing Leech]

Even with that, the feat is not possible.

Although I also believe the "no performance hit" seems impossible, your example doesn't quite fit. You describe different processors. The apparent claim is that on the exact same hardware, you could run an application under Windows or under Linux and get the same performance. I'm not sure you could ever prove this capability is impossible, but it seems to be. Somehow the OS instructions need to be translated which must take some additional clock cycles. But it is at least theoretically possible for some other aspect to be sped up (e.g., Linux does something faster than Windows would) so the net speed is roughly the same. Having that work both ways (Linux app on Windows vs Windows app on Linux) seems a little bit of a stretch though.

Re:Tortoises all the way down.

[Blitzenn]

What are you trying to say? That a horse is not a horse if you look at it from the ass end? That's silly! If the code was written for a CISC processor and you run it on a RISC processor, it HAS TO take extra clock cycles to tranlate the complex instructions back to reduced instructions. It cannot happen magically and software actually takes clock cycles to run so the emulation has to provide a performance hit in itself. It's simply self delusional to accept anyones claim in this area as truthful. It is a claim that you can get more energy out than you put in. It cannot happen. You cannot make hamburger from steak without putting energy into the crank to turn the grinder. Yes it's still all beef, but it takes extra effort to change it from one packaging to the next. (Silly Goose!, or is it Silly Goose Meat!)

Re:Tortoises all the way down.

[irc.goatse.cx troll]

You could somehow 'compile' the application for your native machine, as in, run through all possible code and map it to the native syscalls/machine code.

Of course, this wouldnt be 1:1, but you'd be able to run it much more efficiently than doing it dynamicly. Kind of like how wine can use .so replacements for windows .dlls, only on a higher level.

Re:Tortoises all the way down.

[Steve Cowan]

That's just ridiculous! GeoWrite's all the word processing I could ever need.

Re:Tortoises all the way down.

[LiquidCoooled]

I have a sneaky suspision that this software relies on caching the local translated version of every known executable file.

The engine will likely allow template driven translations, and x86 linux to x86 windows could be byte for byte identical.

So it takes a moment to scan a file. We all run virus checkers in Windows every day, they scan our files and we dont notice.

THIS is the kind of comparisons it could feasibly say is 100%. Clicking on the Doom3 windows exe in linux loads up the preconverted translated file.

Now, different processor configs do work differently, and from the sounds of things they can't have done much work there. I think shying away from it by using Windows->Linux as their "test" operating system purely because they are the simplest.

This is like the Macintosh emulators that became available for the amiga. Both machines ran on 680x0 processors, and emulating a same period apple mac was a lot more usable than emulating an 8086.

Re:Tortoises all the way down.

[Anonymous Coward]

Surely you can't read into "no performance penalties" not to mean "on like hardware".

Surely you couldn't have not written that without not using more negatives.

Re:Tortoises all the way down.

[Blitzenn]

I agree with your template idea. Not for all executables, but for all instruction sets. They would have to have somekind of a table of translations to load equivelent sets of instructions for the CPUs to successfully translate the code.

For the Amiga, although it is quantifiably a CISC processor. It's age as compared to other processors almost makes it appear as a reduced instruction set processor. That would make it run Applish code more efficiently than '86ish code. Instructions just match up better with less translation needed between them.

Re:Tortoises all the way down.

[LiquidCoooled]

My recollections of ShapeShifter at the time were that it was such a revolutionary idea simply because it didn't try to emulate the system, but merely used the cpu.

Everything happened as expected, the 680x0 code ran exactly as intended on the mac, all that was needed was a replacement exception handler for all the custom bits ("Ooops, your code tried to access memory 0x00123460, instead of throwing a Guru, its meant to be a macs' printer PageFeed or whatever it was meant to be").

Running an x86 emulation caused massive problems due to running a full cpu emulation and requiring multiple instructions per byte of PC code.

I doubt it.

[Anonymous Coward]

Will it run software written for my Wang 2200? Will it run VMS? No. Please do not make such broad claims if you can't even begin to back them up.

She wanted a reason.....

[SomeoneGotMyNick]

allows any software application binary to run on any processor/operating system

My wife said, "Give me a good reason why you need to keep those things! There's not enough room in the closet to put my shoes."

Now I can use this as an excuse to hold on to my Commodore 64 stuff.

Re:She wanted a reason.....

[savagedome]

If you start giving her good reasons to keep Commodore 64 stuff, you might have to give her a good reason to keep YOU eventually ;)

Just a thought.

Re:She wanted a reason.....

[ZoolTheNinja]

No, your wife is right. email me and I'll send my mailing address so you can ship it to me.

Re:She wanted a reason.....

[Anonymous Coward]

My wife said, "Give me a good reason why you need to keep those things! There's not enough room in the closet to put my shoes."

At this point I would have thrown your wife's shoes out. And your wife, too. Hell, I don't even know her!

Now I can use this as an excuse to hold on to my Commodore 64 stuff.

I think I like you. You can move in, if you want.

Re:She wanted a reason.....

[hellfire]

I dunno... my Apple IIc stuff takes up about 4 boxes. Do I want a woman who needs that much room to store her shoes?

Re:She wanted a reason.....

[AbbyNormal]

Desktop support? (On her brand new Commodore 64 desktop?)

Re:She wanted a reason.....

[euxneks]

As opposed to her fucking shoes? At least the commodore is relevant in history.

Re:She wanted a reason.....

[Anonymous Coward]

Besides, if your woman needs fucking shoes then perhaps you aren't doing your job properly.

Re:She wanted a reason.....

[drinkypoo]

Maybe she's just short, and they're some really tall platforms.

Re:She wanted a reason.....

[Anonymous Coward]

Maybe she's just short, and they're some really tall platforms.

Oh, I wouldn't know about that because I'm a Baptist.

wow, that takes some nerve

[John_Allen_Mohammed]

to name your new company the "tranvestite corporation." I guess it makes sense considering the product they're selling but, sheesh

Were cross-platform ports shown?

[Cus]

a Linux version of Quake III -- running on an Apple PowerBook
...and...
Windows laptop running the Gimp image editor for Linux

Funny how those applications are already available for those platforms, hmmm? I'd like to have heard about something being shown that isn't already available natively.

Re:Were cross-platform ports shown?

[Jamori]

Something else you might notice in their provided screenshot http://www.wired.com/news/images/0,2334,64914-1449 5,00.html is that if you look very closely, their mac seems to be running KDE. While there probably is a theme for OSX to make it look like KDE, why would they do that while demoing that their software lets you run a linux binary on a Mac? Looks to me like they just installed some linux distro with ppc support on the powerbook and loaded up quake. It sounds great .. but I call BS.

Re:Were cross-platform ports shown?

[mattyrobinson69]

This actually says to me that they might be telling some truth. i think this because gimp was ported to windows, therefore, its code base must be portable and not contain any (or not much) linux specific stuff, and quake was definately written to be portable, so again, no to much os specific code.

obviously if im right, it means most of what tehy are saying is lies, but at least they have a product.

although i do believe this is coincidence and the whole site is complete bollocks.

Re:Were cross-platform ports shown?

[lullabud]

This is the first thing I thought while reading it... By running software that already runs on all 3 main desktop platforms they haven't demonstrated anything useful about their product. Why would they do that? The only reason I can think of is so they could say "here is how it runs natively, and here is how it runs through our architecture-translating process." However, since they didn't say anything specific about speed comparisons, only that "95 percent won't notice," I seriously doubt that comparisons were the reason. If they really want to clear the air of the scent of BS they ought to at least release a video, not just a photo of the title screen of Quake III running KDE on a powerbook. 95 percent of powerbook users would notice KDE running on instead of their beloved Aqua.

what about

[xmple]

what about api calls which are native to windows, or even directx applications, I don't think that they will be able to "emulate" all this without a performance hit (or at all for that matter).

Or are they just talking about the processors and their native instructions?

If it is real...

[Talonius]

...they'll soon be sued away by the likes of Microsoft and Apple, both of whom have an established interest in maintaining the status quo.

It sounds like a virtual machine they've created for each host operating system and "virtualized" operating system. While possible - see WINE and the lately not heard from David project - this would require quite a bit of work. Hell, trying to emulate Linux in this way would be a hoot. Which window manager do you want to emulate today?

I think it is mostly vapor. Enderle, the famed SCO analyst, has his hands in it and I immediately distrust anything he works with and endorses.

(I just found out that my sister's ex-boyfriend's brother is one of the major financers of the Phantom. How's that for being close to slime?)

Re:If it is real...

[Anonymous Coward]

1) Find relation to sucker.
2) ...
3) Profit!

Sheesh - this is a good thing: he's stupid slime, and you're not even related to him!

Re:If it is real...

[Ohreally_factor]

I think it is mostly vapor. Enderle, the famed SCO analyst, has his hands in it and I immediately distrust anything he works with and endorses.

I don't know whether the FUD department at MS realizes this yet, but they've got a powerful weapon in Enderle. Whenever an exciting new tech comes along that might threaten MS, all they have to do is have Enderle comment positively, and then the new technology will have zero credibility.

Looks like their web site running on this...

[Anonymous Coward]

...software. It's already down.

Riiiight

[Gyorg_Lavode]

"What do you think, vaporware or miracle?"

Is that a rhetorical question?

Kinda ironic isn't it...

[arock99]

That they claim their software is fast but yet their web site is reaaaaaaaaaaly slow

Re:Kinda ironic isn't it...

[ZoolTheNinja]

Well of course it's slow! They are emulating IIS on WinNT, all running on a Commodore 64!

Great Success Story!

[Anonymous Coward]

Given the illusive nature of their product, it is especially incredible that they were able to get five of the world's largest computer OEM's on board so early.

Now you can run any software, anywhere, with no speed hit (relative to a 4.77 MHz PC XT or a C= 64 or a 512K Mac) on hardware from these everyday major name brand OEM's:

Billy-Bo's Bayou's only Computer OEM
Wang Tu Short Compuder OEM of China
DR CLEMENT OKUN NIGERIA BUSINESS COMPUTER MANUFACTURERS
San Rio Hello Kitty Laptop Division
TransitivePC & Electronics

Act now, because supplies of this software are very limited, and once this run is completed, no more can be made (because their damn drunk coders crashed a pickup truck into their RAID array)!

Game piracy?

[tepples]

From the article:

"We try and avoid the word," said Wiederhold. "When people think of emulators they think of things that are very slow."

When some people think of "emulators", they also tend to think of the arcade and console game ROM piracy scene that goes along with it. The company may have dodged more bullets than it's letting on to us by not using that word.

Xbox 2

[Loadmaster]

The article says that this software will allow Xbox 2 to play Xbox one games even though Xbox 2 uses a PPC instead of Intel. Correct me if I'm wrong, but isn't the main problem with backward compatiblity for the Xbox the Nvidia graphics chip? How do they plan on doing their translation of code without violating Nvidia IP? Seems to me like their being too simplistic.

Re:Xbox 2

[kidgenius]

I think another issue here is that in order to get Xbox 2 games to play on an Xbox, they will need their software on an Xbox. Now, how do you get software onto an Xbox. Either you do it w/ Microsoft's blessing, or you put unsigned code on it by way of hacking apart the old xbox and inserting a mod chip. As you can imagine, Microsoft wants people to buy Xbox 2. It wouldn't be too happy about people buying the games to use on their old systems. Now that we know Microsoft won't give their blessing to this company to run their code on the Xbox, I'll let you decide which of the two options is left.

I'd imagine that Microsoft would be none to pleased about option 2 either.

Re:Xbox 2

[mrgreen4242]

The OP (and presumable th FA) are talking about play XBox1 games on the Xbox2. Not the otherway around, as you seem to be talking about.

Re:Xbox 2

[kidgenius]

My bad. I have been reading since TFA, "will allow the next-generation Xbox (which will have a Mac-like PowerPC chip) to run ON first-generation Xbox software (which was written for an Intel chip)."

well

[TheAxeMaster]

Sounds like complete BS, but on the other hand if it even works reasonably well it would be a step up from anything we've got now.

Say it took 10% of your resources to run this thing, and the rest could be devoted to the program, that's only a big deal in the most absolutely recent high load games and programs. The rest of the midrange stuff where we normally work would run great. So it is possible that this could be a great thing, some will just have to try it and report back.

Impossible

[marcovje]

E.g. assume that the application is a game. The exact timing must then also be duplicated, a quite difficult process. (see e.g. the various Mac, Amiga emulators and Dosbox)

Did anyone note...

[cavac]

...that they seemed to run only processor-native code. Even Linux-Quake: Linux IS ported to the Mac ;-)

And the example of the XBox: Xbox is essentially a PC anyway.

This looks more like the technique the WINE project is using: Run a program on it's native hardware platform on another OS by making library- and systemcall-wrappers.

If that is indeed the case, "translating the code page-wise" can be translated to "re-linking dynamically loadable code page-wise".

Just my 2 cents

Re:Did anyone note...

[cavac]

What i did forget to mention: If they show me a Windows DOOM3 running on an Apple (at whatever framerate) THEN i start believing.

Anyway, isn't it a little late for /. to post dot-com jokes on the frontpage.[1]

[1] No, SCO doesn't count. That would be called "tragedy".

Re:Did anyone note...

[Spoing]

"Even Linux-Quake: Linux IS ported to the Mac ;-)"

Quake III has been ported to the Mac. Quake (not III) has been ported to Linux on Mac hardware. I don't think Quake III for Linux has been ported to Linux on Mac hardware. (Could be wrong!)

Because of that, running an x86 Quake III (Linux or Windows) on OSX or MacOS on Mac hardware is a real trick. I'd like to see it.

Re:Did anyone note...

[vadim_t]

Um, Linux runs on the Mac because it's written in portable C. You compile it and get a Mac binary, which won't run on a x86 box.

Same goes for programs, a x86 Linux binary won't run on a Mac even under Linux.

Re:Did anyone note...

[LousyPhreak]

...that they seemed to run only processor-native code. Even Linux-Quake: Linux IS ported to the Mac ;-)

you got a little flaw in your example... quake and linux might be ported to the mac but still, the linux version of quake3 is afaik x86 which is quite a bit different to a ppc, so either they cheated (gimp is available for windows and quake3 for mac) or they do more than just slap on some wrappers

Re:Did anyone note...

[Glock27]

...that they seemed to run only processor-native code. Even Linux-Quake: Linux IS ported to the Mac

No. From the website:

Dynamite allows software that has been compiled for one processor/operating system to be run on another processor/operating system without any source code or binary changes. To do this, Dynamite provides a hardware virtualization technology that consists of four key components. First, an integration "FUSE" allows Dynamite to be easily integrated into the target system. Second, a dynamic binary translator tackles the challenge of moving from one instruction set architecture to another. Third, an operating system mapper translates operating system calls from the source system to the target system in situations where the source and target operating systems are different. Finally, a graphics subsystem mapper translates graphics system calls from the source to the target system in situations where the source and target graphics systems are different.

Hello, I am in write-only mode

[Mauvaisours]

QuickTransit fully supports accelerated 3-D graphics and about 80 percent computational performance on the main processor. It requires no user intervention: It kicks in automatically when a non-native application is launched.

"80% computational performance" == "almost no performance hit", of course. And I will start dancing on the moon tomorrow with "almost" no equipement.

There company website server.

[Anonymous Coward]

Is probably running on a 64bit compiled version of apache for xinix run under contiki OS on a spectrum ZX that also doubles as there primary VMS server.

iow, It's down !.

Hints in the adv...I mean, article.

[Howzer]

Unlike most other emulators, QuickTransit translates blocks of code rather than a line at a time. In addition, it identifies and stores the most commonly executed code.

This seems to me to be saying that the emulator's got to be HUGE, rather in the fashion of the fake-o compression software that comes up every couple of years promising 1,000:1 compression. And then you find out that at each end is a store of "templates" and the compression application is literally a couple hundred GB in size. Maybe in this case a lot of standard libraries have already been ported?

And since that line is the most informative in the article, I'd be keen to hear other opinions on how this could work...

Software choices....

[kidgenius]

Turley said he watched a Windows laptop running the Gimp image editor for Linux.

In demonstrations to press and analysts, the company has shown a graphically demanding game -- a Linux version of Quake III -- running on an Apple PowerBook

Does this company realize that proper existing ports of each of those particular pieces of software exist in some kind of native form for those architectures? I've used GIMP in Windows w/ no problems. Also, as mentioned previosuly, Quake III already exists for the Mac as well. What good are they doing by using software that already exists in ports? I want to see a copy of some DirectX game running on a Mac/Linux w/o a performance hit. This company so far has not proven anything by using the two comparisons cited in the article.

Fine print

[archeopterix]

It runs everything (1) on everything (2) without performance hit (3).

(1) Uhm, err, the current version only runs Pacman, which required some modifications to the binary
(2) only on Windows XP, but we're working on the Win 98 version.
(3) The technology used allows for theoretical performance equal or even exceeding the native hardware. This will work in next version, "FlyingPig 6.0".

Re:Fine print

[dasmegabyte]

Nobody with any sense at all is working on the Win 98 version of anything.

Supporting deprecated and crummy operating systems is asking for a support nightmare. You'd doubtless have to charge more than the cost of WinXP for it.

Any on ANY ??

[Ancient_Hacker]

Any on ANY?? I think I still have a CDC Cyber 172 system dump tape under the raised floor.... So if they're right I can run NOS 1.3 and PLATO on an old Apple ][ ?

Apple to move to Opteron?

[9uG]

Maybe this is Apple's chance to jump ship from PPC/IBM/Motorola and use the Opteron? There was a rumor they were developing an intel versin of OSX next to the PPC version, so why not move to AMD64 and ditch PPC? This emulator could be just what the doctor ordered.

lead to gold

[bodrell]

What do you think, vaporware or miracle?

Is this a survey? Then I vote vaporware. Is digital alchemy kinda like Wicca?

Shifty Business

[jorenko]

The "pretty darn impressive" examples given are GIMP under Windows and Quake 3 on Mac. Both of those have completely native versions available. I smell something not quite honest about these demos.

On their "technology overview" page...

[Kippesoep]

there is mention only of unices. Operating System Mapper. Dynamite supports operating system mapping between any two Unix/Linux-like operating systems, as well as mapping between mainframe and any Unix/Linux-like operating systems. Don't see "Windows" mentioned in there. I assume it would be a lot easier to run a Linux version of Quake 3 on BSD-based Mac OS X than to convert stuff to/from a rather more different OS such as Windows.

Tivo?

[cat_jesus]

So I can emulate my Tivo on a PC that I build myself?

You have to be very careful with claims about being able to emulate ANY hardware, or being able to run ANY code.

But if they can, I'll use it.

Now I'll go RTFA and see how stupid I am.

Re:Tivo?

[multipartmixed]

> You have to be very careful with claims about being able to emulate ANY
> hardware, or being able to run ANY code.

Personally, I was hoping to download it and run HPC apps on my Commodore 64.

My guess..

[k98sven]

The summary should almost be modded flamebait for making such an obviously impossible statement like that.

So what's really up here? TFA says they demonstrated running a Linux Quake III on a OS X powerbook.
(And they quote Rob Enderle praising this technology.. this is the guy who thinks SCO will win, which speaks loads for his credibility.)

Now, I haven't seen the source for Quake III, but I'm pretty certain it uses OpenGL, which the Mac has. OS X is also POSIX-compliant. So, most of the API calls done by Quake can already be done natively on OS X.

So what I guess they're doing here is translating API calls (like Wine) while emulating the processor core (like a real emulator).

That isn't anything new. For instance, I've written similar code for an Atari emulator, which can emulate an Atari hard-disc filesystem as a local directory through translating OS calls.

(Note: And that was far from the first time it'd been done either.)

Antitrust; shunning the Mac users

[tepples]

Windows will team up with hardware vendors (Dell, HP, IBM) and give them incentives to make this NOT work

Antitrust. Microsoft has been convicted before in multiple jurisdictions and will be convicted again. A repeat offender should draw a real penalty.

Office prices will skyrocket (don't know how that's possible, but it will to prevent this) UNLESS you have a valid Windows license

Do you mean Microsoft will raise the price of Microsoft Office for Mac OS X? In that case, watch Mac users flee from Microsoft MBU products to competing products such as products built on top of OpenOffice.org.

Re:Antitrust; shunning the Mac users

[gfxguy]

Antitrust. Microsoft has been convicted before in multiple jurisdictions and will be convicted again. A repeat offender should draw a real penalty.

That's an interesting word..."should".

As Cletus the slackjaw yokel told more "Shoulda but didna."

Had me until "no performance hit"

[visionsofmcskill]

Bottom line, a porcessor essentialy comes down to several basic comparisons and read/write add/subtract operations.

so it is technicaly feasible that if you map out a fair amount of the pipelines of most of the popular chip sets, you could technicaly have a command chain to allow binaries the same calls through a sudo-emulation layer of the software.

fundamentaly possible, and even do-able.... but without a performance hit? no way. Each processor is geared towards a particular way of solving a physcial and mathmatical set of problems... some processors are designed for massive loads of database driven calculations (XEONs)... some for multimedia (G5)... some for science (PPC, Sparc?)... some for power savings (ARM)....

depedning on which archetecture your using, the performance will be greatly hindered if your trying to do something designed for a radicaly different chip. Such as trying to run some expansive G5 optimizied photoshop plug on a ARM chip.

"no performance hit" = total bullshit

Re:Had me until "no performance hit"

[Stalus]

The problem here is that the 'no performance hit' is a bad quotation by the summary author. From the article:

[...] with almost no performance hit. [...] QuickTransit fully supports accelerated 3-D graphics and about 80 percent computational performance on the main processor [...]

Later on, a consultant, after seeing gimp running on windows, says "There was no performance hit", but based on further explanation only means that he did not notice any UI lag from the emulation. That doesn't mean that there wasn't a performance hit, it just means that the UI is approximately as responsive as the normal thing.

Reminds me of...

[Astadar]

that IBM "Universal Business Adapter" commercial. The one for which to interface with system "xyz" you need an ADDITIONAL adapter.

So, all you need to run:
- a Windows app on Linux is our software and Wine
- a Linux app on Windows is our software and cygwin
- ...

Clearly vaporware

[hopethishelps]

without any performance hit. That would allow any program written for Windows to run on Linux or Mac, and vice-versa,... What do you think, vaporware or miracle?

This is vaporware. What they're claiming - "without any performance hit" - is impossible. Accomplishing the rest of what they claim is not impossible, but it's very difficult, and since the "without any performance hit" claim establishes conclusively that these people are bullshitters, I don't believe they can even come close to doing it.

Re:Clearly vaporware

[tiocsti]

I'm not so sure, they could be using some binary translation technology combined with api shims to achieve this. For binary translation technology, they could maybe base their work off of uqbt, and provide api shims a la wine.

I'd be surprised if it were real, but it's not one of those things that seems impossible to me.

>>
without any performance hit. That would allow any program written for Windows to run on Linux or Mac, and vice-versa,... What do you think, vaporware or miracle?
This is vaporware. What they're claiming - "without any performance hit" - is impossible. Accomplishing the rest of what they claim is not impossible, but it's very difficult, and since the "without any performance hit" claim establishes conclusively that these people are bullshitters, I don't believe they can even come close to doing it.

Re:Clearly vaporware

[mofochickamo]

The Slashdot post didn't communicate this point clearly. Here is the text from the Wired article they are referring to:

"There was no performance hit," he said. "I was expecting a lag, some symptom that things were not as they should be, but that was not the case. There was no hand-eye delay. It seemed completely normal. It responded really quickly."

The performance hit seems to be the perceived performance hit, and not a measured one. Also, Transitive's site doesn't claim there is no performance hit.

Re:Clearly vaporware

[Anonymous Coward]

they could maybe...provide api shims a la wine.

Wine's api shims are executable code. So they take some amount of time to execute. So there's a performance hit right there. And they're claiming no performance hit. So that's not a solution.

Never mind the huge amount of time and effort that the Wine project has put into doing just one emulation - Win32 on x86 Linux - and still, AFAIK, not got it 100.0% right. Perfect emulation, even with a performance hit, is a hugely difficult task. They're saying that, unlike Wine, they have 100.0% perfect emulation; that unlike Wine, they can emulate anything on anything - Windows on Linux, OS/390 on Windows, OS/2 on Mac OSX. Nobody in the history of computer science has ever done this, but it is theoretically possible. Luckily, they then make a ludicrous claim - "without any performance hit" - which is good, because it saves us the bother of investigating their credibility on the rest of their claims.

Re:Clearly vaporware

[drinkypoo]

There are cases in which JIT-translated code runs faster than the original code. I'm not saying that's happening here, I'm just saying that it's not strictly impossible. It does seem extremely implausible that it would happen in all cases no matter how smart the programmer was, however; it seems like it would pretty much depend on a combination of the author of the JIT translator being smarter than the author of the compiler optimizations that come into play in a given routine, and the hardware having at least the same capabilities as the original host hardware.

What was it they said...

[Anonymous Coward]

about Marijuana getting stronger?

Legal status (pretty OT)

[Maffy]

Does anyone what the legal status of running this operation over commercial software would be?

The reason you need a licence to use software is because your CPU makes a copy of the program (in RAM) and this would otherwise violate the programmer's copyright. I believe that the licensing terms are generally pretty strict, e.g. one copy, to RAM only. Therefore, I'm not sure you'd be permitted to take a copy of their program, mangle it and dump it back out to disk.

Does anyone know of any reason why this would be permitted, or how people intend to get round this problem?

I appear to have been reading too much groklaw.

Re:Legal status (pretty OT)

[Anonymous Coward]

The reason you need a licence to use software is because your CPU makes a copy of the program (in RAM) and this would otherwise violate the programmer's copyright.

No, it doesn't.

Re:Legal status (pretty OT)

[illumina us]

Needing a license to use the software because your CPU makes a copy of the program in the RAM makes no sense at all. Especially since most software cannot even fit entirely into RAM. If I am not mistaken licenses cleary state that you are licensed to install this program on one machine at a time only. If you choose to install it on another machine you must remove the copy from the aforementioned machine. It says nothing about the program being processed in the computer's memory. The license is for you to install it onto your harddrive and execute it for your purposes. Most licenses also allow you to make backups of your software for your own personal use.

The reason you need a license for the software is because that software is copyrighted and in order to use or display copyrighted material you need to have a license for it. Music and video that you purchase at the store has an implied EULA type agreement too. Just because you don't see it doesn't mean you are not bound by it.

Re:Legal status (pretty OT)

[Maffy]

I believe there is a loophole in that law, as described in Wikipedia's software licence article.

http://en.wikipedia.org/wiki/Software_licence#Copy right

Re:Legal status (pretty OT)

[Anonymous Coward]

That's a pretty weak loophole. All I need is a sales receipt to show that I am the "owner of a copy." A perpetual rental agreement--paid in full up front with sales tax included, I might add--is legal nonsense.

Besides that, suppose we accept that 117 doesn't apply. How do I legally run setup.exe and see the EULA in the first place?

Re:Legal status (pretty OT)

[FooAtWFU]

The reason you need a licence to use software is because your CPU makes a copy of the program (in RAM) and this would otherwise violate the programmer's copyright

Now that's an interesting take on it. Who says this? Where can I read more about this aspect of it? It's certainly a new argument as far as I'm concerned- has it ever been mentioned in a courtroom setting? Where did you come up with this?

Re:Legal status (pretty OT)

[Fnkmaster]

Agreed. Said loophole is not just weak, it is ridiculous. If I walked into a store and bought a piece of software or provided my credit card online to buy it, then I am the owner of a copy of it, unless I explicitly knew I was renting it and signed a rental contract of some sort up front. Have you ever "rented" anything without a rental contract (even when you rent videos at Blockbuster, you need an account and signed contract first)? You can't buy something then discover after the fact that the buyer wasn't ever really selling it, they were just selling you a box and "perpetually renting" you the contents, which they reserve the right to deny you your property rights (resale, etc.) on in the future.

As always EULAs have always been pretty much bogus - you can refuse to accept them, or click on "Accept" while your fingers are crossed, but you can still feel free to use the software as you will within the constraints of copyright law.

Re:Legal status (pretty OT)

[Maffy]

The reason you need a license for the software is because that software is copyrighted and in order to use or display copyrighted material you need to have a license for it.

This isn't true.

Copyright only prevents you from copying something. If the software publisher makes the copy and gives it to you, you can read that copy (e.g. play music from a CD) without requiring any other licence from them.

Computer software is in the interesting situation of requiring making a copy before use. As a previous poster pointed out, there is an additional section of copyright law to try to get round this.

Re:Legal status (pretty OT)

[Dr. Evil]

IMNSHO, it's BS... as I'm sure you were implying :-)

You've always been able to copy stuff to your heart's content... what you're not allowed to do is copy and distribute... among other things.

Re:Legal status (pretty OT)

[Blastrogath]

Actually, this is worse than you suspect. Most modern licences explicitly forbid reverse engineering their code. This sounds like it'd be likely to be caught by those licence terms.

Re:Legal status (pretty OT)

[iamlucky13]

Even if it did violate the letter of the copyright law, I would argue that it does not violate the intent of protecting the originator's interests regarding the program. I can't see any reason why programmers wouldn't want users to be able to run their software on any platform. You could compile and market one version, but sell it to Win+Mac+Linux+misc OS users.

Oh yeah, I cast a vote for vaporware...I would wager it would fail to make a perfect compilation a significant portion of the time.

Re:Legal status (pretty OT)

[servognome]

The reason you need a licence to use software is because your CPU makes a copy of the program (in RAM) and this would otherwise violate the programmer's copyright
The CPU does not make a copy of the program into RAM, RAM is where the used data (not necessarily the entire program) resides.
I think this falls under fair use:
117. Limitations on exclusive rights: Computer programs
(a) Making of Additional Copy or Adaptation by Owner of Copy. -- Notwithstanding the provisions of section 106, it is not an infringement for the owner of a copy of a computer program to make or authorize the making of another copy or adaptation of that computer program provided:
(1) that such a new copy or adaptation is created as an essential step in the utilization of the computer program in conjunction with a machine and that it is used in no other manner.
IANAL, but I think this would qualify as an adaption of a computer program for use in conjunction with a machine.

Re:Legal status (pretty OT)

[Chris Burke]

Is that even a loophole? Would a court recognize a distinction between the "owner" of a copy and someone legally authorized by the copyright holder to "posses" a copy? 117 provides an exemption to the copyright; it seems to me the only way you could be charged with copyright infringement for performing a sec. 117 copy is if you were not starting from a legal copy.

Security Implications?

[niiler]

If it's not vaporware, does this mean that us Linux users will be able to benefit from the plethora of windows viruses too? Of course this is probably a stretch at best.

Things that make you go hmmmmm....

Re:Security Implications?

[kellererik]

Well said, especially the wording with no user intervention makes me nervous.

my 2 cents

Transparent with little overhead...

[DrXym]

... is impossible. Unless of course their emulator / JIT compiler starts in 0 seconds flat and recompiles / emulates the native code in 0 seconds with 0 byte memory footprint and 0 disk activity with 0% affect on performance.

If in fact, their software doesn't all of the above things the performace will take a hit. And not just a small hit either since it's not just the app that has to be translated at runtime but everything running underneath it - the C runtime, any system libs & APIs, X windows / Aqua / GDI etc. If that's the case, you might as well stick with an emulator and be done with it, or use something like WINE where it is appropriate.

Comparisons to WINE also begs another question. What's the point of running your Mac OS X application (e.g. Photoshop) on Linux if it requires you also have a large chunk of OS X to do it? Is a Linux user meant to buy an OS X licence in order to run this one app on their machine? Is an XP user meant to have yank half a Linux dist from somewhere to support the GIMP through X Windows?

It sounds like a recipe for licence hell and lawsuits to me.

Not impossible but...

[gilesjuk]

I'm sure it will require a lot of grunt.

With enough CPU power you can emulate things full speed. But it's the quirks and bugs of the silicon which you are emulating that causes problems.

Not to mention the emulation code probably contains bugs too. Can be hard to achieve 100% compatibility.

Will it run on my Xbox?

[samberdoo]

I always want to run Lotus Notes after I play Halo

not impossible, but we'll see.

[pb]

The marketing jargon goes a bit over the top, but it isn't impossible to translate code for one ISA to an intermediate form, optimize it, and then generate code for another ISA. I don't know that it's revolutionary either. Note that LLVM takes a similar approach, and has a very simple intermediate form. I hear someone on their team is working on a PPC front-end, and as for language front-ends, Java and C# is in the works.

Getting back to Transitives, in July 2001, they claimed to already be doing x86->MIPS translation, which bodes well for x86->PPC. However, doing things efficiently the other way around is tougher. And of course you need to support or translate a ton of the native OS API calls etc. It'll be interesting to see for Windows on Linux (for example) if they require a copy of Windows to run the binaries.

Re:not impossible, but we'll see.

[magic]

You don't need to translate OS calls if you license the other operating system and run it behind the scenes. OS2 provided flawless Win16 emulation by this method.

-m

Qemu

[ssewell]

http://fabrice.bellard.free.fr/qemu/ Qemu needs work, but already has several different host-target CPU combinations complete (in the ARM x86 and PPC areas anyway).

http://pearpc.sourceforge.net/ PearPC fills in the gaps on PPC emulation where Qemu is lacking. (I run OS X 10.2 on my PC at work.)

Re:Qemu

[vinsci]

It's worth pointing out that Qemu is about 65 times faster than Bochs and is actually usable for running that last windows app on your Linux system, for example, so that you can finally drop the windows partition forever, without having to get a VmWare license. You can run just about any operating system under Qemu, including Windows XP, Windows 2000, Windows ME/98/95 and Windows 3.1, not forgetting every Free OS on the planet, historical operating systems etc... FreeDOS runs fine, of course, as does OS/2 Merlin, I hear.

It's excellent for trying out your ISO boot disks as well, just start it with "qemu -cdrom path/to/image.iso -boot d" and it boots from your ISO image file.

Did I mention that it's Free as in freedom and beer, GPL:d (parts LGPL:d and still others MIT license) software?

Thanks, Fabrice Bellard and the rest of the team for this great work!

Cached site

[Anonymous Coward]

Coral Cached link to Transitive site: http://www.transitive.com.nyud.net:8090/

Vaporware or Mirical?

[uberlinuxguy]

Considering when I click their link in the post and get a "Document Contains No Data" message, I'm going to lean towards vaporware.

Time will tell

[Un0r1g1nal]

Guess in a few days we will know if it is all just vaporware or not.

If it's for real then we have a lot of fun times to look forward to, no more excuses to not move companies to a Unix based OS, can then start weeding out the crap programs and replacing them with decent open source ones over time till don't need this program either.

Does sound too good to be true though, and when something sounds to good to be true ... it usually is.

is it true?

[Goeland86]

I have to say that to me it doesn't sound like it does what they want it to. I mean, why would they have a version for x86, powerpc, et-al if it's supposed to run on multiple OSes? Shouldn't it subdivide the categories with windows, macOSX, linux-2.4, linux-2.6, *BSD, BeOS...? Also, just a little quote from the article about the "almost no performance loss" : "QuickTransit fully supports accelerated 3-D graphics and about 80 percent computational performance on the main processor. " So in fact when computing, those translated programs are 20% slower than a natively written or source-ported program in terms of fpu. Now, I don't know about you guys, but 20% to me sounds ALOT! Especially for games. Doesn't sound like it's an impressive converter to me. Winex/Cedega does better for me in some cases. Granted, it might be helpful for medical applications, but gamers are hardcore, they'll want all the juice they can squeeze out of their boxes, so no way they'll voluntarily cut off 20% of their gaming speed, even if the graphics follow. Of course it SOUNDED great, but when you look under the hood it's impressive if it does work, granted, but it's not as impressive as they're trying to make it sound.

Now this is funny...

[illumina us]

In demonstrations to press and analysts, the company has shown a graphically demanding game -- a Linux version of Quake III -- running on an Apple PowerBook.

Because the past 2 generations of PowerBooks haven't been running on a BSD kernel which is very similar to *nix and runs most *nix applications natively anyway? Moreover, Quake III uses OpenGL as a renderer which makes ports a lot easier than games using DirectX.

From the website...

[cagem0nkey]

...in the certain event that it gets slashdotted.
"Dynamite allows software that has been compiled for one processor/operating system to be run on another processor/operating system without any source code or binary changes. To do this, Dynamite provides a hardware virtualization technology that consists of four key components. First, an integration "FUSE" allows Dynamite to be easily integrated into the target system. Second, a dynamic binary translator tackles the challenge of moving from one instruction set architecture to another. Third, an operating system mapper translates operating system calls from the source system to the target system in situations where the source and target operating systems are different. Finally, a graphics subsystem mapper translates graphics system calls from the source to the target system in situations where the source and target graphics systems are different.

Integration FUSE. Dynamite is integrated with the operating system of the target system and runs like any other application. Unlike static translators that translate an application's binary once before run time, Dynamite translates instructions and operating system calls while an application is running. When a foreign application is started, the operating system recognizes that the application needs translation and automatically starts Dynamite. Depending on the requirements for the integration, Dynamite can be configured with a wide range of options, including the ability to build "bridges" between translated code and code running native on the target platform. This feature has been used, for example, to allow translated applications to call a native accelerated graphics library for the graphics chipset in the target platform, delivering higher quality and speed than other solutions. Integration options also provide for different optimization schedules and the ability to reconfigure the use of the translation code cache to better match performance to customer needs.

Dynamic Binary Translation. The dynamic binary translator in Dynamite is a breakthrough technology that uses a modular architecture consisting of three key components. The front-end decoder reads blocks of instructions from the foreign application's binary and decodes them into an intermediate representation. The intermediate representation allows Dynamite to understand the higher-level semantics and intent of the code.

The optimizing kernel reads the intermediate representation and optimizes the code. At first, simple optimizations are performed. In most applications, however, a 90/10 rules holds where 10% of the code is executed 90% of the time. The optimizing kernel looks for blocks of code that are executed often, spends increasing amounts of time improving the optimization of this code, and then stores this optimized code in memory. Each time a frequently used block of code needs to be executed, the highly optimized code stored in memory is used instead of optimizing that block of code again. Because the blocks of code that are executed change frequently, the optimizing kernel flushes old optimized blocks and generates new ones. The optimizing kernel produces superior code optimization compared to static binary translators or compilers. It optimizes code based on how an individual user is using that application and does not need to optimize code for the general case.

The back-end code generator outputs code for the target processor. Different RISC, CISC and VLIW back-ends provide different ways of performing register allocation, instruction selection and operand addressing modes to take advantage of the distinct features available in a particular target processor.

Operating System Mapper. Dynamite supports operating system mapping between any two Unix/Linux-like operating systems, as well as mapping between mainframe and any Unix/Linux-like operating systems. Where similar operating system calls exist between the source and destination operating system, Dynamite maps calls between the two.

huh? 24hrs early

[SubtleNuance]

If The Transitive software is supposed to be released today

Why not ask "What do you think, vaporware or miracle?"

tomorrow?

Why speculate when you can... i dunno, wait 24hrs and inform yourself...

Ha

[ScrewMaster]

I'd say it would be a miracle if it isn't vaporware.

I've also got a bridge to sell you ... cheap.

Any software on any processor/OS?

[Anonymous Coward]

So can one run Doom3 on Commodore 64?

Reminds me of DEC alpha x86 translation

[Anonymous Coward]

When reading that part of the article, I do recall that DEC once touted an alpha "feature" where pages of memory could be marked as foreign, and when faulted, could be ran through a translator that turned x86 code into alpha code, and then cached the result on disk for future loads. This allowed the Alpha port of NT to run some (maybe many) x86 native apps, with a bit of initial overhead, at near native speed. Of course it also required hardware support in the Alpha MMU to accomplish this. \

Did anyone note...

[tgd]

That the Xbox 2 is not the same processor as XBox 1?

Virii?

[attam]

from TFA:
It requires no user intervention: It kicks in automatically when a non-native application is launched.

does this mean my mac running this software is now susceptible to windows worms and virii w/ me totally unawares?

Sweet!

[Anonymous Coward]

Now, instead of writing a 100% accurate decompiler (which, afaik, is impossible), I just need to design a processor/OS to use INTERPRETED C (or LISP or whatever), and use this software to translate my binary to that platform.

Oh wait. Did I just reduce an uncomputable problem to what this software claims to do? Ah well, I'm sure Turing was wrong!

slashdotted - vaporized

[tsager]

See, the domain already vanished...

I bet that webserver was emulated. No perfomance loss..ha!

No Windows support?

[Elessar]

The website says "...supports operating system mapping between any two Unix/Linux-like operating systems, as well as mapping between mainframe and any Unix/Linux-like operating systems...". This would suggest that there is no mapping between windows and unix...

Look at this! It must be true!

[Anonymous Coward]

Did you read the article? It says:

Analyst Rob Enderle of the Enderle Group said Transitive benefits from the fact that most modern machines are fast enough to emulate each other without much affecting performance.

Wow! We all know how reliable Rob Enderle's analysis is. It must be true!

Admittedly, I don't quite see the logic of his statement. After all, if an app takes twice as long to run when emulated on processor X as it does when running natively on processor Y, it'll still take twice as long when you speed up both X and Y in the same ratio. Maybe he means that you can emulate an Apple ][ dating from 1981 on a 3-GHz Pentium, and get the same performance as the Apple ][ would have given in 1981.

emulator or compiler?

[khrtt]

Has anyone tried translating binary code between Mac and IA32? Should be much faster than emulation, and the output could be cached to disk, maybe. Generate a binary for the target architecture the first time the IA32 binary is loaded, linked with the API converter library, and cache it to disk. Given that the PowerPC has more registers than an IA32 machine, and a more flexible architecture, the translated code could be made as efficient as the original, as no registers would have to be emulated in memory.

Than, all you need to do is to write the API converter library. That could very consuming of man-hours, and difficult to get the bugs out of, but it has been done before many times over, with variable success.

Maybe that's what they do. If not, this post will be prior art for anyone trying to patent this shit:-)

OT: where is that from?

[Bill, Shooter of Bul]

I remeber the quote "Tortoises all the way Down" as being from a women that believed that the earth was supported on th backs of tortises or something like that. Some one asked what the tortises were standing on and she replies "Tortoises all the way down". Where Do I remeber that from? Was in in one of Richard Feyneman's books, or do I remember it from history class ( it was a native american belief that the earth was actually a giant tortoises shell)? Or maybe it was from both.

Re:OT: where is that from?

[D-Cypell]

It was definately mentioned in one of Hawking's books. I cant remember which though, possibly 'Breif History...'.

Re:OT: where is that from?

[Hard_Code]

Various cultures have had this view, including hindus, who I believe insert a level of elephants either on or below the turtle.

Re:OT: where is that from?

[Matthaeus]

From the forward to Hawking's "A Brief History of Time."

Re:OT: where is that from?

[Anonymous Coward]

"Foreword." The word is "foreword."

You fucking imbecile.

Re:OT: where is that from?

[Anonymous Coward]

In another current thread, concidentally enough, is this link to a succinct retelling of that story: http://www.apatheticagnostic.com/articles/meds/med 09/med173.html

Re:OT: where is that from?

[VendettaMF]

Also, I seem to recall, in Pratchett (plus some less funny guys)s Science of the Discworld...

Re:OT: where is that from?

[AndroidCat]

Many religions and cultures believe that it's BS all the way down.

Re:OT: where is that from?

[stankulp]

Nobel Prize-winning physicist Stephen Hawking starts his book A Brief History Of Time with an anecdote about a scientist giving a public lecture on the nature of the earth, the solar system, and the galaxy. After his talk, an elderly woman rises from her chair and says "What you have told us is rubbish. The world is really a flat plate supported on the back of a giant tortoise." "Ah. And what is the tortoise standing on?" asks the scientist. "You're very clever, young man, very clever," retorts the old lady. "But it's turtles all the way down."

http://www.imdiversity.com/villages/native/arts_ cu lture_media/ture_turtles.asp

Re:OT: where is that from?

[plover]

It is attributed to an unnamed woman attending a lecture by Sir Arthur Eddington in the first half of the 20th century.

Google found this page for me.

Re:OT: where is that from?

[IamNotWitchboy]

Everything you say it's true, except the part about the Nobel Prize.

Professor Hawking has never won the Nobel Prize. You might want to confirm that up in the several bios available on the net.

Re:OT: where is that from?

[Anonymous Coward]

Indeed. As soon as someone can figure out how to experimentally confirm something like Hawking radiation from black holes, he would definitely be in line to get one. Until then, the Nobel committee will probably hold out on him.

Re:OT: where is that from?

[Anonymous Coward]

"Saying you "Could care less" implies that you do indeed care."

Only to those without a sense of irony and/or sarcasm, you fucking imbecile.

Re:OT: where is that from?

[Anonymous Coward]

Not only are you a fuckwit who doesn't understand the simple phrase "Could not care less" you're also demonstrating your total lack of understanding of irony and sarcasm and your inability to decide wether your statement is inclusive or exclusive with the wishy-washy use of "and/or"!

You Sir, are the biggest failure of the day. YOU FUCKING FAIL IT FOUR TIMES!

Re:OT: where is that from?

[Anonymous Coward]

Its also the opening to chapter 1 of Prometheus Rising

Re:OT: where is that from?

[The_Wilschon]

There's nothing wrong with infinite series...

Sum from i = 0 to infinity of Turtle_i = Earth

Re:OT: where is that from?

[VendettaMF]

Generally people worth speaking with will be able to use the phrase correctly, rather than parroting whatever halfwitted mangling of english they last picked up from Eminem (Though Shania Twain is also guilty of this particular offense).

Face it. "Could care less" is an idiocy, matched only by "Could of been", created by illiterate trash talking halfwits and defended only by anonymously posting trolls (also generally halfwits).

Re:OT: where is that from?

[Anonymous Coward]

Didn't I just explain to you how "could care less" is a sarcastic remark that makes perfect sense? I won't defend those who parrot it mindlessly and intend no sarcasm by it, but you're mistaken that the phrase was "created by illiterate trash talking halfwits." Rather, it was coined by those with more highly developed linguistic abilities than yours.

Re:OT: where is that from?

[VendettaMF]

You may have attempted to but it was not very convincingly argued I'm afraid.

Standing against you you have three main points (and a host of minor ones).

Firstly, there's the sheer stupidity of the way your corruption of a usefull phrase sounds, even if you do manage to proclaim it sarcastically.

Secondly there's the general use of the corrupted version by folks who (to put it kindly) are a little too low on the evolutionary ladder and/or (to use a favourite of yours) lacking in general education and intelligence to have mastered any form of sarcasm, never mind irony (though I still don't see how/why you dragged irony into this).

Finally there's the real killer of the whole debate. You're posting as an anonymous coward, hence rendering any input you might have void of any validity it might possibly posess (Not so much of an issue in this case its true).

Grow some balls (or ovaries. Your choice.) If its not worth putting your "Karma" on the line for then it's not worth saying.

Re:OT: where is that from?

[Anonymous Coward]

The "corruption" you allege sounds perfectly natural unless one has an underdeveloped sense of verbal irony (including sarcasm). Please be assured that a dulled ironic intuition is nothing to be ashamed of. Contrary to what you allege me to believe, it doesn't necessarily reflect an inferior education or intellect. There are more terrible interpersonal deficiencies by far, and your inadequacy in this particular regard can probably be remedied by a social life that extends beyond the other similarly disadvantaged individuals with whom you've been hitherto spending your time.

As to your second point, I already said I agree the phrase is commonly used without awareness of its sarcastic intent. An unfortunate state of affairs, to be sure, but one which in no way indicates that the phrase was lacking in sarcastic intent in its invention, as you say you believe.

Finally, I'm not sure what point you were trying to make with your third-argument non sequitur, but I'm not particularly inclined to sit here and puzzle it out, as I have an engagement to make. By all means reply if you feel compelled.

Realtime, most probably not.

[catwh0re]

The basic reason why you can't run each platforms software on other platforms at full speed via emulation, is usually due to the architecture of the chipset at the heart of each platform.
For example it's very slow to emulate a PowerPC chip on an x86 platform. This is due to the x86 platform simply having less available registers, hence the software must emulate these, which is a processor intensive operation as it's no longer an on-chip operation.

The best PPC emulation on x86, I have seen so far would be the PearPC software, whose author is sadly recently deceased. However even with that, just simply booting Mac OS X was a multiple hour long task.

I saw a beta test!

[Atryn]

Confirmed! I saw a beta test of this product. It was used during the filming of Independence Day. They successfully used the program to upload and execute a virus from a Mac to a never-before-seen Alien computer system. It was even able to display graphics without having prior experience with the displays in question!

Re:I saw a beta test!

[halivar]

Confirmed! I saw a beta test of this product. It was used during the filming of Independence Day.

Interesting. That means that this miracle software was written in Pascal (watch the movie closely). I always knew there was a darn good reason not to move CS 101 students to C. There it is.

Re:I saw a beta test!

[Anonymous Coward]

Actually I can confirm this too, each time I saw them do this it ran sucessfully. I must have seen this 4 or 5 times each time working flawlessy as the time before.

Re:I saw a beta test!

[Anonymous Coward]

I beg to differ. In Independence Day, they DID have experience with the alien computer, as they had captured one of the alien ships...the one they used to upload the virus.

Re:I saw a beta test!

[Atryn]

they DID have experience with the alien computer, as they had captured one of the alien ships...

Well, assuming the captured ship had full access to the same computer system, you are right. But they still had (IIRC) less than 24 hours experience with it as it only regained power when the aliens arrived back.

Re:I saw a beta test!

[Anonymous Coward]

It wasn't really a virus that Jeff Goldblum loaded up in Independence Day. He knew that Mac OS 7.5.3 was so buggy that it would crash anything it came into contact with.

Vaporware or miracle?

[Anita Coney]

It simply doesn't matter. If the program works, Microsoft will simply buy it and bury it. If it doesn't work, then who cares!?

I want SAMBA for Windows

[MarcQuadra]

You laugh, but if I could get SAMBA for Windows I could slowly convert my workplace to OSS while making my job easier, and the enduser experience better.

SAMBA, in my experience, serves Windows clients better than Microsoft's servers do.

Re:I want SAMBA for Windows

[drinkypoo]

One thing I've noticed using samba is that I get prompted for a password about a bazillion times quicker when connecting to a share. Also, transfer rates seem to be higher, but that could just be Linux responsible for that rather than Windows SMB, and it probably is. (Networking, filesystem, etc.)

Re:I want SAMBA for Windows

[MarcQuadra]

Well the password thing is probably just because your SAMBA config is pulling passwords from /etc/smbpasswd, while the windows box might have to query an active directory, which is an order-of-magnitude more complex.

I've got an OS X SAMBA box at work authenticating users against our Windows 2000 Active Directory, and it's a bit slower than authenticating against local password files.

But I've noticed overall that SAMBA seems to handle SMB better than Windows. It DOES seem faster, has cool options like case-conversion-without-sensitivity, *NIX permissions to *NIX CIFS clients, and hard-settable no-bullshit browse master settings.

I have a lot of OS X boxes that keep their home folders on SMB, and it's a LOT better when I host their folders on the SAMBA box than on the Windows box. When I put users on the Windows box they get filename errors when downloading long filenames, and firefox won't work. When I put them on SAMBA they notice a speed boost and everything starts working right again.

Article is vapor-news

[Anonymous Coward]

Since TFA is a worthless content-devoid POS, and since the transitive website is /.ed, here is a useful link on HOW they claim to do it. It sounds plausible, at least. http://66.102.7.104/search?q=cache:KjTa-qAM7LQJ:ww w.transitive.com/technology.htm+site:www.transitiv e.com+-qwerty&hl=en

Re:Article is vapor-news

[Firehawke]

Entirely plausible. They'd probably end up using some sort of caching of binary translations to prevent there from being the same startup delay every time you start up your application. Applications where you have an identical interface (eg OpenGL) would see no performance hit on the graphics side.

Emulation programmers have been playing with dynarec cores for years now, but compatibility at a low level tends to suffer. Some software is expecting scanline-perfection because they're talking right to the hardware and are tuned to that performance. You'll notice the ABI/API compatibility only talks about various Unix variants and mainframes. So, dynarec PROBABLY won't have to worry about that so much in this case.

Kinda disappointing. I'd like to see what wrangling they could do with MacOS and Windows. Those would be much harder to pull off.

Remember FX!32 ?

[Anonymous Coward]

DEC had several compilers that translate from executable to executable format. This trick was
done compiling Vax images on Alpha, and I believe it also was done for Solaris Sparc images on Alpha and
x86 images on Alpha, and has now been done at least for alpha images on IA64 (Itanium). Performance achieved there could be as high as 80% where the self-tuning aspects of the technology were used. (That is, every time you ran the image, more stuff got identified as code and compiled.) At least for Windows, though, it would be necessary to have emulations for all the NT functions, which led to a rat race trying to chase all the new DLL functions down, which would change frequently, with the apparent aim of keeping anyone from building an emulation of everything. The FX!32 product did pretty well even so and allowed many x86 binaries to be run on Alpha. Seems to me that the claimed performance may be a tad optimistic (80%) unless many tuning runs are done, but still quite usable. But I would be surprised to see more than a small subset of OSs supported, much as it would be pleasant to be able to run Alpha VMS on an Athlon 64 this way.

Answer from Transitive's Website

[mofochickamo]

Transitive explains the architechure of their system here. Basically, to support APIs on different operating systems they have what is called an Operating System Mapper. They don't claim that it maps Mac to Windows or Linux to Windows. Basically, it maps two like systems together (like Solaris to AIX or HPUX to Linux). If there is no straightforward mapping then the customer defines the map.

After reading this, the term Universal Emulator doesn't seem to apply. Here is the text from Transitive's Website:

Operating System Mapper. Dynamite supports operating system mapping between any two Unix/Linux-like operating systems, as well as mapping between mainframe and any Unix/Linux-like operating systems. Where similar operating system calls exist between the source and destination operating system, Dynamite maps calls between the two. Where an equivalent operating system call doesn't exist in the target environment, Dynamite maps to similar calls per the customer's guidance. Dynamite also monitors certain system calls, for example thread scheduling and memory mapping calls, to ensure that it can reproduce the complete behaviour of the program it is executing.

Re:Answer from Transitive's Website

[cyngus]

This appears to be the big 'gotcha'. They mention they're going after the mainframe and high-end server market, so this makes sense that they can translate to/from mainframe and Unix/Linux.

The problem I see here is that its not that hard to rewrite the app in this situation anyway, particularly between Unix and Linux. Now, if you told me I could run a Windows app on Solaris, now you're talking. So they've compartmentalized the key problem (system call emulation/translation), but haven't really solved it.

Re:Answer from Transitive's Website

[tiocsti]

The more I read about it, the more it sounds like a proprietary uqbt implementation. UQBT has many of the same limitations (particularly only handling like os - like os...

http://www.itee.uq.edu.au/~cristina/uqbt.html

>>
Transitive explains the architechure of their system here. Basically, to support APIs on different operating systems they have what is called an Operating System Mapper. They don't claim that it maps Mac to Windows or Linux to Windows. Basically, it maps two like systems together (like Solaris to AIX or HPUX to Linux). If there is no straightforward mapping then the customer defines the map.

Lawsuit!

[Anonymous Coward]

SCO owns all IP!

Freeport Express?

[argent]

Digital, before Compaqtion, was getting pretty impressive results from FPX. They were statically recompiling VAX and Sparc executables to run under OSF1, and using a combination of recompilation and emulation for Windows x86 code to run on NT Alpha.

It still wasn't fast enough for Alpha NT to take off even before Microsoft pulled out, but it was still pretty impressive.

A good recompiler would be enough to make this work, and even keep it from being terminally painful.

That's true.

[pb]

However, you're still effectively paying for a copy of the API. I wonder how much MS would charge a competitor to VirtualPC...

Binary translation is beliveable

[nurb432]

Sure, having an emulated CPU that runs simple binaries is easy..

But running any *program* anywhere? Nah.. too many variables such as libraries, hardware differences, OS issues, etc..

Unless they are writing virtual machines for each platform, ( or stealing others ) i dont buy it.. Nor do i buy the performance statement.. emulation doesnt work that way...

( and no i wasnt able to read the article.. cant get wired from here today it seems

Re:Binary translation is beliveable

[gerardrj]

Seriously... do they expect us to believe that GTA will run on a PDP11?

Not "no performance hit"

[Anonymous Coward]

The article says that they expect 80 percent computational performance, which is not out of the question, especially for native-on-native like Gimp running under Windows.

I wouldn't be surprised...

[Anonymous Coward]

...if is a product of the raelians.
http://www.rael.org/

tortoises Re:OT: where is that from?

[samjam]

I read that in "Surely you're joking Mr Feynman"

"You think you're clever don't you? It's tortoises all the way down"

Sam

JUST IN TIME!

[Eric_Cartman_South_P]

Duke Nukem Forever can FINALLY run on my Powerbook. w0ot!

Ok, here's a question...

[Weaselmancer]

Exactly what makes this any different than, say...qemu?

Enderle? Media Whore.

[sparkhead]

"Analyst Rob Enderle of the Enderle Group said Transitive benefits from the fact that most modern machines are fast enough to emulate each other without much affecting performance."

All I needed to know. This guy will say anything and if he appears in your press release (yeah, it's an "article" but certainly the material is in a press release), chances are you're straining for credible commentary.

Emulated Web Server??

[Jumpin' Jon]

To prove it's not vapourware, I think they're running a C64, emulating NT4, hosting IIS for their Web server.

Complete Text of Transitive's Architecure

[mofochickamo]

Dynamite allows software that has been compiled for one processor/operating system to be run on another processor/operating system without any source code or binary changes. To do this, Dynamite provides a hardware virtualization technology that consists of four key components. First, an integration FUSE allows Dynamite to be easily integrated into the target system. Second, a dynamic binary translator tackles the challenge of moving from one instruction set architecture to another. Third, an operating system mapper translates operating system calls from the source system to the target system in situations where the source and target operating systems are different. Finally, a graphics subsystem mapper translates graphics system calls from the source to the target system in situations where the source and target graphics systems are different.

Integration FUSE. Dynamite is integrated with the operating system of the target system and runs like any other application. Unlike static translators that translate an applications binary once before run time, Dynamite translates instructions and operating system calls while an application is running. When a foreign application is started, the operating system recognizes that the application needs translation and automatically starts Dynamite. Depending on the requirements for the integration, Dynamite can be configured with a wide range of options, including the ability to build bridges between translated code and code running native on the target platform. This feature has been used, for example, to allow translated applications to call a native accelerated graphics library for the graphics chipset in the target platform, delivering higher quality and speed than other solutions. Integration options also provide for different optimization schedules and the ability to reconfigure the use of the translation code cache to better match performance to customer needs.

Dynamic Binary Translation. The dynamic binary translator in Dynamite is a breakthrough technology that uses a modular architecture consisting of three key components. The front-end decoder reads blocks of instructions from the foreign applications binary and decodes them into an intermediate representation. The intermediate representation allows Dynamite to understand the higher-level semantics and intent of the code.

The optimizing kernel reads the intermediate representation and optimizes the code. At first, simple optimizations are performed. In most applications, however, a 90/10 rules holds where 10% of the code is executed 90% of the time. The optimizing kernel looks for blocks of code that are executed often, spends increasing amounts of time improving the optimization of this code, and then stores this optimized code in memory. Each time a frequently used block of code needs to be executed, the highly optimized code stored in memory is used instead of optimizing that block of code again. Because the blocks of code that are executed change frequently, the optimizing kernel flushes old optimized blocks and generates new ones. The optimizing kernel produces superior code optimization compared to static binary translators or compilers. It optimizes code based on how an individual user is using that application and does not need to optimize code for the general case.

The back-end code generator outputs code for the target processor

like DNF

[WormholeFiend]

This thing will only be fun when I have it installed on my PC...

Until then, since I'm not in the high-risk investment market, I won't be bothered to speculate or hype this thing further.

Current game console makers are sure to be pretty pissed off.

Well, maybe except the guys who are making the Phantom console.

Bangalore

[Anonymous Coward]

You just need a fast internet connection... it transmits the entire code to a cadre of Indian programmers who quickly perform the translation and then it's sent back to your computer.

Genius!

Re:Bangalore

[Anonymous Coward]

So, is that Out Source Coding?

Dynamo

[daveho]

The idea of machine code to machine code translation is not new. For example, Dynamo appeared in 1999, although in its case the source and target instruction sets were the same. Regardless of whether or not transitive is for real (I'm guessing it is), it's clearly just a matter of time before this technology appears in a practical form.

Marketing Speak vs Legal Speak

[buckhead_buddy]

My guess is that they are selling something which legally fulfills their legal claims. Of course what we're seeing are their marketing claims which sound like they were written by the good witch of Oz.

A disassembler that pulls apart certain binaries, applies a transformation matrix that the user defines to compensate for the difference in processors, and then re-assembles a binary sounds like something that could be done in very limited circumstances while laying the success or failure of the transformation (along with the legal liability) on the customer's shoulders. What about their performance claims? The legalese may claim the speed is of the transformation process and not the speed of the re-assembled transformed binary.

Just one example of why you might want to see the legal speak rather than the market speak.

All of ye nay-sayers

[aldousd666]

Had better be pretty sure about that before you say it's all bunk. After all, they didn't say it has NO performance hit, just not a substantial one -- and they did even concede that power users may notice it. The one question I have is -- what about other stuff other than the hardware, say the windows Registry and DirectX and all that bullshit that goes along with emulating a windows host -- or the Resource and Data forks in Mac OS files -- how about HFS or Ext2 file systems? NTFS? how are these issues resulved by 'code translation?' How about installation ? You can run programs that are in MSI format in any emulator I've seen... Is there a fake System32 folder on a linux host running this emulator?

I think not.

[geordie_loz]

About a minute out by my reconing.

The sound of the other shoe dropping...

[pjt48108]

And in a related item, Microsoft has now "acquired" Transitive...

Oh, I'm kidding of course (but would it shock anyone?)...

Ahh, but can it emulate...

[gosand]

I really want to know if it can emulate the Phantom game console.

Let the script kiddies rejoice

[Anonymous Coward]

So dos that mean I have to now fear the latest Windows virus on my MacOSX box? Sounds like another reason to hate Microsoft.

Come on guys

[cheezit]

The interesting part of this story is not the product---snake oil at best---it's the wild-eyed enthusiasm of Slashdot posters who know nothing about the product and yet start blathering about how great it's gonna be to run Windows app X on Linux. Always helps when the post misuses some technical term like "instruction set" or "assembly language."

Fool me once, shame on you; fool me twice, shame on me; fool me a zillion times, I'm on slashdot!

Guys, think about it---if apps can run on any platform, that means that all the OS components that the app is dependent on must be emulated or must travel with it. Sound like a possible licensing problem for commercial OS's? And if the app interacts with hardware AT ALL then they have to create an intermediate layer to allow for access to all device drivers. But wait---the OS components have to talk to that layer too! Uh-oh. Across "any" platform, with full functionality. Suuuure.

Now I know that CS professors are smart---I understand some of them were even required to write a bit of code to get their doctorate---but let's come back down to earth shall we?

Re:Come on guys

[Erwos]

"Now I know that CS professors are smart"

The most fascinating thing that no one's commented on: this guy isn't even a prof, he's a lecturer.

-Erwos

Vaporware

[crmartin]

Consider: if the superinterpreter can reduce the instruction stream of the program to one target machine instruction per program instruction, it would be able to execute as quickly as the program would have when compiled to the target. But since the interpreter has to do some amount of computation to get there, it will necesarily be slower over all.

And, of course, if the interpreter executes epsilon more than one target instruction per program instruction, it'll necessarily be slower.

If follows, therefore, that there must be some performance cost.

Total user transparency?

[isolationism]

[QuickTransit] requires no user intervention: It kicks in automatically when a non-native application is launched.

Congratulations, all of your Solaris and Linux boxes have been infected with MyDoom.

Finally!

[Viking Coder]

Now I can finally emulate my 3.6 GHz machine with a BFGTech GeForce 6800 GT OC on my $0.85 calculator that I bought at the gas station! Thanks, Transitive Corporation!

</sarcasm>

The marketing guys shouldn't use big words like "without any performance hit" that they don't understand. Other than that, it sounds interesting.

Maybe they should just call it...

[Trolling4Dollars]

..."write once, crash and burn everywhere" technology? ;P

remember the DEC re-compilers?

[Anonymous Coward]

Remember when the Alphas first came out? DEC had a re-compiler that took your old VAX VMS binaries and translated them to run under Alpha VMS. Same OS, different machine architecture. It actually worked quite well. They also had a re-compiler that took Sparc-SunOS4 binaries (I think, it wasn't Solaris 2.X was it?) and generated Alpha-DEC-OSF/1 binaries. I don't think they actually sold that, but it made the rounds in universities. It was good enough to run WordPerfect for SunOS on an Alpha.

Project david has a friend

[DMJC-L]

oooh.. wow.. what's next.. toothfairy os 5.6 santafactory 1.0? really people, this stuff is bullshit and we all know it.

nothing new, just new implementation

[Pr0xY]

this is not as revolutionary as they make it out to be at all. First of all the concept of "converting blocks of code instead of lines" is known as JIT compiling by the JAVA people and "dynamic recompilation" by you everyday emulator. Also they state that this is only successly multi-platform emulator...uh MAME anyone "Multi-Arcade Machine Emulator".

My thoughts is that they have combines several known techniques to come up with a nice solution. My guess is that they implement some intermediate API emulation layer (similar to wine) and do some form of dynamic recompilation of the code to the native arch (at the same time, translating the system API calls to ones which can be trappped byt the wine like wrapper). After a few secs most of the code should be running native.

Not impressed....yet

proxy

What do I think?

[iii_rjm]

It does not matter what I think. It either works or it doesn't.

Not an emulator

[joshv]

From the end of the Wired article:

"The company is keen to avoid the term "emulator," instead calling its technology "hardware virtualization.""

I suspect that they are emulating a hardware environment and running an entire OS in a virtual machine. Sound familiar? VMWare accomplished this several years ago, and they did in on multiple platforms. Transitive might might have some new performance tricks up its sleeves, but I'm betting they have a similar approach to the problem.

As others in the discussion pointed out, you can't just run the Linux version of the Quake III executable on OS/X in an emulator, because the executable makes calls to OS routines that simply aren't there. So you either provide native libraries that reimplement the foreign OS routines on the new platform, or you run the entire OS in the emulator. It sounds like the latter is what was used for the Quake III demo mentionned in the article.

VMWare all over again.

Re:Not an emulator

[edittard]

"The company is keen to avoid the term "emulator," instead calling its technology "hardware virtualization.""

I'm keen to avoid the term "bullshit", instead calling the article "bovine excrement".

I had a recompiler on my C64

[Anonymous Coward]

I had a recompiler on my Commodore 64 that would convert BASIC to machine code. My program would run maybe twice as fast (which is a far cry from what it should run had it been written in pure assembly/ML to begin with). It also would bloat up the program. 5kb in BASIC would bloat up to over 70 kb when recompiled.

Call me when there's a demo.

[Etone]

They can make up any sort of fancy claims they like, but until I can run a demo copy and try it for myself, this is not news. It looks like currently the only people who can do this are the ones who shell out for the whole license -- something that most of us won't do sight-unseen.

-E-

Hold on everyone .. Rob Enderle is involved

[talexb]

• Analyst Rob Enderle of the Enderle Group said Transitive benefits from the fact that most modern machines are fast enough to emulate each other without much affecting performance.

We've seen this gentleman before .. in the SCO takes on the world saga.

It seems they've convinced a few people .. we'll see if it's the real deal within the month I guess.

Um...

[Theaetetus]

Does this company realize that proper existing ports of each of those particular pieces of software exist in some kind of native form for those architectures? I've used GIMP in Windows w/ no problems. Also, as mentioned previosuly, Quake III already exists for the Mac as well. What good are they doing by using software that already exists in ports?

... so that they can show the native version side-by-side with the translated version and show that there is no noticable hit in performance.
/obvious

Re:Um...

[Anonymous Coward]

No this isn't informative or obvious. What's the point in showing something that already works, working, but running another OS' version? What would make sense, say, would be two similarly spec-ed machines, one running Linux and the Linux port of GIMP, and then another running Windows and the Linux port of GIMP.

Re:Um...

[Theaetetus]

What's the point in showing something that already works, working, but running another OS' version? What would make sense, say, would be two similarly spec-ed machines, one running Linux and the Linux port of GIMP, and then another running Windows and the Linux port of GIMP.

When you're demoing a product, you want to show it in multiple different lights (all of them hopefully good). So first, you show that there's no performance hit, by running a non-native version of an app alongside a native version of the same app. Then, as a second step, you start showing things that do not have native version - such as Garageband on Wintel, or Cool Edit Pro on Mac, etc. Thing is, if you just started with step 2 and showed Cool Edit Pro on a Mac, how does anyone watching know how well it's working? Sure, it seems snappy, but maybe it's really taking a huge performance hit and could be a ton faster.

None

[siskbc]

...As long as your character is a Gimp. Sort of like a cross between a dwarf and a troll, but the Gimp wears a leather suit and a ball-gag as default equipment.

SOES

[Flyboy Connor]

This sounds remarkably like the affair around the Sloot Digital Coding System (link is in Dutch, while there are plenty Dutch links on this issue, it seems this affair has stayed in our little country).

Sloot had created a system which allowed any movie to be encoded in 1 kilobyte of data, using a group of five algorithms that were stored on a hard disk of less than 400 Mb. He demonstrated this system on several occasions, but didn't allow anyone to open his machine. Roel Pieper, one of Holland's most notorious IT professors (who basically became notorious because he had been a sportsman in the US, not because he knew a lot about IT) invested quite heavily (we're talking millions here) in Sloot's company. When Sloot died, taking his secret with him to the grave, they opened his magic box and found a big hard disk inside on which many movies were stored - using their regular space.

Of course, any first-year IT student realises that it is mathematically impossible to store any movie in just 1K of data, since it puts an upper limit to the number of movies that can exist, while it is obvious that the number of movies is infinite (although you wouldn't say so if you examine Hollywood's output).

The same seems to be happening with this universal translator. Of course it is mathematically impossible to create such a thing. The creators hype it up. They get loads of money. And then, I predict we will never hear from this again.

Some people get rich, others lose wads of money, the world keeps turning.

not entire program, but "blocks" of code

[Creepy]

That's not how I read it - the article says it recompiles blocks of code on the fly and caches the most used ones. The difference between this and most other emulators is that most other emulators compile line-by-line (or so the article says).

I noticed that they demo'd OpenGL programs for 3D, of which a graphical equivalent exists on all systems (and many of the functions in hardware) - what happens when you run DirectX, where this is not necessarily true? I suspect the only way would be to run Windows (barring card support - I'm just using this as an example) in emulation. What kind of emulated hardware support if running Windows on MacOS or Linux does such an emulator have?

I think

[kc0re]

This may be a too good to be true story. Like many other posters, I have my doubts regarding this one. But if it did work, that would be beautiful.

That'll be some feat!

[His name cannot be s]

That'll be some feat!

I can't even run my Windows binaries full speed on Windows, how the fuck are they going to put MacOS binaries full speed on Windows?

Tell me that!

feh!

It pegged by BS meter before I got to the page. I'm pretty sure that it's going to be like the magical compression that can compress anything down to 256 bytes.

Details, details, details...

[chill]

The graphic on Transitive's website shows only Unix/Linux operating systems. One of their steps, Operating System Mapping, says "QuickTransit supports operating system mapping between any two Unix/Linux-like operating systems, as well as mapping between mainframe and any Unix/Linux-like operating systems." Doesn't sound like Windows to me.

All of this is supposedly done on the fly, and not beforehand.

Quake and The Gimp wouldn't be my choices to show off flexibility. Quake is OpenGL on Linux and OSX, so there isn't any graphics magic going on. And the ability of BSD-based systems to run Linux binaries is old hat. The Gimp isn't exactly taxing on a CPU as far as user responsiveness goes.

Extraordinary claims require extraordinary proof. Press conferences for journalists aren't a conducive forum for proving anything. They are a good place to baffle 'em with bullshit, though.

-Charles

It doesn't say "No Performance Hit"

[nathan s]

It says specifically "QuickTransit fully supports accelerated 3-D graphics and about 80 percent computational performance on the main processor" - that's 80% performance. Later, it says "The power user might notice the difference, but the other 95 percent won't notice.""

This means that there IS a performance hit, just that they don't expect the average user to notice (probably rightly so).

Note to moderators:

[Anonymous Coward]

This is intended to be "funny", not "informative".
If you don't "get it", please do not moderate.

Re:Note to moderators:

[tabrnaker]

maybe it's just not funny?

Ports

[42sd]

Don't those applications that they were running( Quake3, Gimp) have native ports.
I'm not saying its vaporware, but I could see someone trying to show off the software using the native ports instead of the "hardware virtualization"

It would obviously be more impressive/credible if they tried it on programs sans native ports.

Re:Ports

[dirty]

I agree. Show Garage Band, Final Cut Pro, Motion, or Soundtrack running on a PC, and I'd be the first in line to buy this.

Re:Ports

[canadacow]

Download PearPC and my sound emulation patch on the sourceforge site and eat your heart out. http://www.sourceforge.net/projects/pearpc/

Re:Ports

[Gax]

Damn! I hate these GPL clauses. I had to watch my mouth drop to the floor to install Wine last week.

EXCELLENT!!!!

[Medievalist]

any software application binary to run on any processor/operating system

Wow! I can run all my old PDP-11/24 binaries on the Nixdorf mainframe now! I need this software!!!!

How will the software run?

[AstroDrabb]

Just because it can work cross-processor doesn't mean it works cross-platform. For example, how will it allow an MS Windows app to run on Linux or Mac OS X? The hardware part can be done. But what about all the libraries? What happens to the win32 calls under Linux or Mac OS X? What about Linux GTK+/KDE/X or Mac OS X with Quartz/etc running on M? Is this company trying to say that they ported _all_ these API's to _every_ possible platform? I doubt it. It is just like VMWare where you need to install a whole other OS? It really just sounds like marketing hype to me.

Left out

[Anonymous Coward]

3. ???

alternative webserver

[Anonymous Coward]

/.'d - http://transitives.com should also work.

newhardware vs old hardware speed

[updatelee]

Weidel said in most cases, QuickTransit allows translated applications to run faster on new hardware than it did on the original platform, thanks to the speed of today's machines compared with those made a decade ago.

I would hope so ... a decade ago, 1994, we would have been using 486dx66's.

They keep saying emulators are slow, but Ive used a few emu before and none Ive seen running on my intel 3.2HT run at the speed of a 486DX66.

pretty much every emu on the market is gona be running faster on new hardware then on decade old hardware, I dont think its even an issue worth pointing out. if thats what they mean by no performance hit... this is gona be some useless crap.

Re:no performance hit? - correction

[jandersen]

Sorry, it should have read 'The performance is no hit'

Really?

[lokedhs]

How? Considering the fact that the C64 only had 64 kB of RAM.

it MIGHT work....

[Malor]

This isn't completely impossible, but they would have to be exceedingly smart, very probably inhumanly smart, to be able to run all binaries from all systems at good speed on any other. But there are some shortcuts they could potentially have taken that would bring it into the realm of 'very difficult, but possible.'

What most emulators do is to translate, instruction by instruction, the code stream for another processor into the instruction set of the current one. This takes, generally, anywhere from 10 to 20 native instructions per foreign one; in some cases, a lot more. Very clean architectures like the 68000 and PowerPC are easier to emulate, because their instruction sets are fairly uniform. The translation segment can be very fast, just a couple of table lookups and jumps. The x86 instruction set, on the other hand, is extremely snarly and nasty, with a whole slew of different instructions at different sizes. There are 8 bit, 16 bit, and 32-bit instructions, and several processor modes (8086, 80286, 80386). this makes the simple translation of an X86 instruction a potentially very complex process.

This emulator, on the other hand, claims to not be doing this. They claim that they are pre-translating the binaries to native code and running them that way. This would be enormously difficult, in essence extracting the 'sense of' a program from the compiled binary. This hasn't been done much, and there's a reason for it; it's hard to tell what side effects instructions are going to have ahead of time. The only way you can be CERTAIN of the state of an emulated PC is by emulating every step to get to that point. Just inspecting the code, and writing your own program to do the same thing, won't always give you a running program.

Now, they could be simplifying their X86 translation code a WHOLE lot by not emulating 8086 and 286 modes. Very little code, now, is 8- or 16-bit, so if they only support 32-bit code, their problem would be vastly simplified.

They may also be able to simplify their problem by sticking with 'ordinary' programs, ones that stay in userspace and just use system calls for their I/O. I believe that precompling a program that twiddles hardware bits directly would be exceptionally difficult. Transmeta's approach strikes me as the only likely one in this case; emulate it instruction by instruction, watch how it works, and then write yourself shortcuts to make it run faster next time.

But if they stick with 32-bit only, userspace programs, they could possibly get things running fairly quickly. Quake is a good demo for them. It's using OpenGL and is fairly predictable; the inputs and outputs are easily understood and quantified. (network, user input, results of opengl library calls). They're not getting any weird side effects from Quake hitting the hardware directly.

Upshot: it MAY work, but it's likely only going to work well for a subset of programs. It will likely never run operating systems. But, given those limits, it may actually run fairly well.

They mention using this tech to run XBOX1 stuff on XBOX2; if what I'm guessing is correct, that's likely to fail a great deal of the time. It's possible that XBox games may not hit the hardware directly; if they go through library calls for everything, they might be translatable in this way. But from what little I know of console programming, there probably is no class of program MORE likely to be twiddling the hardware. The recompilation technique strikes me as HIGHLY unlikely to work well in such cases.

I suppose they might be able to store special-purpose precompiled binaries on Live.. when you install Game Y for Xbox1, it could potentially go get a program binary from Live that would work. Might be better than nothing, but it would require a lot of engineering effort by Microsoft.

I should also note that I'm NOT claiming this is real. It could very easily be a scam. I'm just trying to point out that if they change the rules a little, they may be able to ship a product that works, without needing brains the size of Jupiter.

RTFA?!

[ElDuderino44137]

"That would allow any program written for Windows to run on Linux or Mac, and vice-versa ..."

Did the person who posted the article read it?

As far as I can tell ... It speaks nothing of running your Wondows software under your Linux OS. It says that you could run your Windows OS on any "chip".

I think the real problem is the confusion as to what the word "platform" means. It doesn't just refer to Windows vs. Linux vs. MacOS. It could refer to one compiler vs. another compiler. It could refer to one chip vs. another (as in this case). For example: the PowerPC chip vs. the Intel chip.

The article specificly talks about the xBox issue this software could solve:

"For example, Wiederhold said QuickTransit will allow the next-generation Xbox (which will have a Mac-like PowerPC chip) to run first-generation Xbox software (which was written for an Intel chip)."

Cheers,
--The Dude

Slashdotted

[evanh23]

Too bad their webserver wasnt emulating a beowolf cluster of linux servers!

Another pipe dream

[Anonymous Coward]

And my can plant my ass on any seat.

Wait what's that sticking out of my asshole.

HP Dynamo come to mind

[MarcoPon]

It remaind me of an HP project named Dynamo. It allowed to translate and existing binary at runtime, eveng gaining some performance in the process. The demo/test was demonstrated on a PA8000 (IIRC). So it got PA8K ASM in, digested it (!), and come out with different PA8K ASM. Every translated block (to a branch point) was cached and reused if possible. All with minimum overhead.

See here: '99 Paper

Bye!

Virtual machine technology

[tod_miller]

Having seen the benefits of Java, and marvelled at OS emulation with VMWare (Should be OS ware) I think this is the next evolved step.

I have no idea how it works above the level of basic code execution (libraries etc) so I think this has no real 'now' uses. (unless running windows on a Sunfire is your thing)

This still doesn't 'do' any emulation that can magic up any system library that is no longer present (becuase it is the different system)

Will this however give rise to a whole bunch of cross compiled libraries that kill off the OS/Architecture debate by saying, write once, run on this.

Will it run mame :-) Can this mean we can beowulf a SNES, A2600, Gameboy, dreamcast, Atari ST and Coleco?

Who knows.

Wow

[Anonymous Coward]

This almost sounds like my Amiga 3000 from 12 years ago, with a Bridgeboard and Emplant I could run Amiga, PC and Mac software at the same time. Granted, it 'cheated' by using a 386 processor for running the PC software, but it shared video, HD, keyboard and mouse.

all i know is...

[enrico_suave]

That one time... (but not in band camp)...

I tried installing slackware inside os/2 warp from a dos shell inside windows sessions... and I saw god!

(and nearly destroyed the space time contium, thank god for tachion rays!).

=)

e.

Looks like another Itanium performance enhancer...

[dtjohnson]

After looking at their website, it looks like the main thrust is just another attempt to improve the dismal x86 performance on Itanium. The rest is probably hype and fluff.

READ THE ARTICE

[Conor Turton]

Nowhere does it say there isn't a performance hit. It merely says that because of the speed of current hardware, only real power users are likely to notice any difference.

"QuickTransit fully supports accelerated 3-D graphics and about 80 percent computational performance on the main processor."

Probably standard OEM rates

[davidwr]

Back when Connectix owned VirtualPC, you could buy VirtualPC for Mac with an OEM-labeled copy of MS-Windows.

My guess is Connectix paid the same as if they were a hardware company - probably a rate based on the # of units shipped per month or some such.

Now that MS owns VirtualPC, if they tried to charge more than they traditionally charged, they might be in trouble on anti-trust violations in some parts of the world.

Still, published APIs whose function is not covered by broad patents can be rewritten, a la WINE.

Commercial Proprietary

[orasio]

Aside from the fact that that is an interesting way to look at things (if what you say were true, maybe a program recompiled by a licensee would no longer have copyright issues), you are taking an incorrect assumption.

Commercial software doesn't have to be proprietary. I use non-proprietary commercial software, and get paid to produce it exclusively, too, and I am not alone in the universe. Commercial does not mean proprietary.

Quick someone rat em out to RMS!!!!

[i_r_sensitive]

Is it just me, or does anyone see the potential for collision with the GPL?

IANAL, but, the GPL is based on a simple fact: For any given binary source code exists. This software has the potential to permanently alter that equation. What exactly is the status of source built by reverse engineering a translated binary? I think there is a potential issue there for the GPL, and even some poorly written Closed Source Licenses... (though these are rare nowadays...)

I wouldn't care to bet one way or the other, but maybe a pre-emptive rewording of certain portions of the GPL is in order...

run anything?

[Anonymous Coward]

"allows any software application binary to run on any processor/operating system"

Lets see the c64 crowd get doom3 to run!

Warning! The above post may contain sarcasim, and no I did not rtfa.

Their website ...

[NiteStar]

... is probably IIS emulated on Linux.

How does it integrate?

[mydigitalself]

Surely this is a blend of vapour ware and some truth. Games, for instance, could perhaps work - but what about applications that hook into operating system fundamentals?

The first example that springs to mind is Microsoft Word. Say you ran Word on this UberEmulator and you clicked File-Save. What would happen? Would you get a Win32 Network Neighborhood/My Documents type dialog? If so, it would probably blow up with some unhandled C++ exception either before it managed to display or upon user input. Or say the Word document had a embedded object in it from Excel or Vision and you double clicked on the object - then what would happen? Without OLE/DDE present (provided by the O/S) then things wouldn't work either.

Re:How does it integrate?

[man_ls]

This operates at a lower level than that.

It's obvious to me that your application won't run if it doesn't have its dependancies, regardless of where system calls, opcodes, and instruction sets are translated: if Word doesn't have the comdlg32.dll library, Common Visual Controls, it won't run.

That'd be like trying to a Linux app with dynamic bindings to glibc....without glibc being on the system.

No performance hit?

[Weirdofreak]

Pfft. Download two versions. Run the Linux one. Use it to run the Windows one. Use that to run the Linux one. Etc. If you get fifty or so and it's -still- not lagging, then I'll believe them.

[NELSON] Ha, ha! [/NELSON]

[Anonymous Coward]

Not only did you get -1, Troll; you also get -1, Offtopic in reply to your -1, Troll. Double-whammied like a punk-ass bitch.

The Demo was Rigged

[Ironsides]

Note that they only did Intel to Intel or Intel to MAC emulation. When they can allow a Mac program to run effieciently on an Intel chip, I will be impressed. From what I understand, the main problem running Mac software on Intel is that Macs have many many more registers.

Re:The Demo was Rigged

[itsNothing]

Oh, so you're pointing out that they solved the easy part of the problem, ehh?

Re:The Demo was Rigged

[Ironsides]

More like they've solved something that has been solved before. Repeatedly. I have seen windows progs run on macs. And linux on windows and vice versa. I have yet to see efficient running of mac on Intel/AMD chips.

Universal? -- No

[1u3hr]

in the small print you find something to the effect of "for this product to function as described, you must meet the following requirements....".

Like this, I think:

Operating System Mapper. Dynamite supports operating system mapping between any two Unix/Linux-like operating systems, as well as mapping between mainframe and any Unix/Linux-like operating systems.

So nothing about Windows, and Mac only as OSX, I think. No mention in the Wired article of it being limited to *ix-like OSs. Which idiot calls this "universal"?

As the IBM advert would say:

[Anonymous Coward]

Does it work in Europe.

(You need an adaptor)

Not as cool as it sounds...

[Phixxr]

QuickTransit supports operating system mapping between any two Unix/Linux-like operating systems, as well as mapping between mainframe and any Unix/Linux-like operating systems.

So, if you've got a program on your P4/Redhat box, you can run it on a Mac/OSX... yay... Seems like a pretty big nothing to me...

Both apps are ported

[Vilim]

Interesting, Jim Turley (whoever that is) said he watched Gimp run on windows and that they had Quake 3 runing on a Mac. These two applications are both ported to thier respective platforms, meaning this demonstration is open to hoax. Run iPhoto on windows or Nero on a mac and I would take it more seriously

Re:Both apps are ported

[sugarmotor]

Or accessing a Winmodem through Linux.

Stephan

Vapux

[Anonymous Coward]

..theres a good anme for the OS.
Microsoft assimilator - the software formerly known as Transitive Corporation

But will it allow

[ajs318]

My question is, will it allow you to run any damn thing you like -- even stuff that came in from an overflowing buffer -- on a processor with native hardware NX support?

Rob Enderle

[Goo.cc]

This quote raises a red flag for me: "Analyst Rob Enderle of the Enderle Group said Transitive benefits from the fact that most modern machines are fast enough to emulate each other without much affecting performance."

Rob Enderle. Why would anybody believe this man is beyond me. For those who do not know, this is the same man who recently gave a speech at SCOForum entitled "Free Software and the Fools Who Use It"?

I also have to wonder how it happens that Rob Enderle would be interviewed about this software, unless he is somehow employed by Transitive. If so, then this Wired article is really more of an advertisement than journalism.

the most important question to ask

[Anonymous Coward]

Can it run "Duke Nukem Forever" from its native code in the ether?

No? How about the unreleased final sequels to "SwordQuest" on the Atari 2600?

Well...

[Anonymous Coward]

Did they just invent Java programming?

Hynes Convention Center

[rezac]

From Transitive's own press release, they indicate that they will be exhibiting at the Hynes Convention Center in Boston on September 20-23 at the Software Developers Conference. http://www.transitive.com/news.htm

The Hynes Convention Center is hosting the SDC on September 20-23 http://www.mccahome.com/hynes/default.aspx

And the SDC lists Transitive as an exhibitor at the conference. http://www.cmpevents.com/SDe4/a.asp?option=H and http://www.cmpevents.com/SDe4/a.asp?option=H&V=13& MgEid=1455&X=1

Interesting? Probably not a hoax.

zac

What's really interesting

[Bruha]

Is the fact that they were quoted as saying:

"For example, Wiederhold said QuickTransit will allow the next-generation Xbox (which will have a Mac-like PowerPC chip) to run first-generation Xbox software (which was written for an Intel chip)."

So they already have a deal in place for this? Doubtfull.

But, but, but, but... !!!

[voxel]

This wasn't possible historically primarily because of "hacking" that occured among programmers.

Being "one of them". I used to throw in bits of assembler and hacks to hide things here or there. "Peek and Poke" (inp/outp) to ports in various places to get the hardware to do what I wanted it to do. Hide some memory on a floppy disk controller then use it later for something.

This kind of activity makes it very very difficult to do what these guys claim they did.

BUT... In the modern programming world, look at an assembler dump from MSVC in a Win32 EXE and you'll see no funny business anymore. No hiding memory, or peek/poke/inp/outp anymore. Just mostly straight forward code with calls to the OS and thats it!

A C compiler generally doesn't take C code, then convert it straight to (YOUR_CPU_HERE) e.g. x86 op codes and dump out a binary.. no... What they generally do is convert C code into an intermediate language. Bear with me as I don't know the proper name for this, but its almost like an assembler language (very low in nature and very simple codes) that is independant of architecture. Then this intermediate code has two options, it can go to the assembler for your CPU or it can go to the optimizer, which will optimize THIS intermediate code then go to the assembler (which can further optimize for the current CPU).

What I am trying to get at, is if you can take a Win32 binary, or a Mac binary, and convert it BACK into this intermediate platform/cpu independant intermediate code, then you can feed it back into the optimizer/assembler again for a different CPU.

There of course is alot of magic around this, but I beleive this is the general idea.

If you can accomplish this, then you have just wrote a translator that can translate one CPU's opcodes into another CPU's opcodes and reoptimize the opcodes for the target platform. The end result would be VERY VERY FAST translated binarys for the target CPU/OS that would perform almost as good, or even better than the original binary on the original CPU/platform (this is if you could say CPU-Apple = CPU-Orange performance wise).

Anyways... I am rambling, but hope they pulled it off and its not vaporware as it would be really cool!

- Voxel

Misleading article...

[jbarr]

From the /. article:

"...That would allow any program written for Windows to run on Linux or Mac, and vice-versa..."

This is a bit misleading, because it's really a "one flavor of *nix on another flavor of *nix" system. Playing a Linux version of Quake III on an OSX Mac doesn't seem like rocket science because of OSX's native BSD roots.

I'll wager that if I took something like "Quicken" or "Microsoft Office Professional" for Windows and tried to run it on a Mac running QuickTransit that it certainly wouldn't work. I doubt if iMovie would run on a QuickTransit-enabled PC. THAT, my friends is the "computer-alchemy" goal. Of course, I would LOVE to be proven wrong on this!

Now, if they are talking about "any program written for Windows [that adheres to QuickTransit Requirements] to run on Linux..." then they may be accurate, but again, this really isn't "universal emulation".

Re:Misleading article...

[swordgeek]

If you keep reading into their website, you're quite right.

However, the very first sentence on their page is:

"Our software allows any softwware application binary to run on any processor/operating system."

So /. is no more misleading than the company itself.

No performance hit

[rjamestaylor]

• "allows any software application binary to run on any processor/operating system" without any performance hit.

It does take quite an acid hit, though...

(Speedy processors would not stop their software from running; reality would, but if it ran on a 300MHz processor, it'd do just fine on a 3GHz)

Not Windows

[boatboy]

Unless I'm reading things wrong, this doesn't help run windows binaries on linux. From the site:Operating system call mapping from any Unix/Linux-like operating system or any mainframe operating system to any Unix/Linux-like operating system is supported.

You could do Linux -> Windows w/ cygwin, Linux -> OSX, or Unix -> Unix, but not Windows -> anything. This makes much more sense, but it still smacks of vaporware.

Vaporware or Miracle?

[Irvu]

"The Transitive software is supposed to be released today. What do you think, vaporware or miracle?"

Ask me tomorrow.

Not even theoretical limits look so good.

[cryptomancer]

Lessee, the Universal Turing Machine required exponential slowdown to emulate absolutely anything. Emulating CISC on RISC is going to take a linear slowdown (expanding some operation into the X fundamental operations). Heck, if this emulator even looks at the program it's emulating, it has to read program P of size N; it can't run itself in constant time C and then run the program it's emulating as if it were native. Unless the miracle is that in constant time C this emulator transforms your system into one native for the program.

I'd like to test this program by emulating a program written for a Cray. Love to have my p.o.s. top-of-the-line-for-1998 transformed into a Cray. :)

I don't buy it...

[GhodMode]

So...

The fine folks over at CodeWeavers are a bunch of fools...

Transgaming has been wasting their time, and ours, with Cedega/WineX...

The group of open-source programmers that have put hours of time, effort, and expertise into WinE have not even been paying attention...

VMWare is a shoddy piece of software and a waste of time and hard disk space...

Projects like BOCHS, DosBox, and DosEmu really never had any purpose or market ...

All of the developers behind all of these projects must be really stupid. It sounds so simple: "Instead of working on every chunk of code, QuickTransit translates a sentence, or a paragraph, at a time."

I don't believe it for a second. It's not because of any great expertise of my own, but because I know there's great expertise behind many of the projects/products I've mentioned above.

- - - - - -
Total: $.02

-- GhodMode

response

[mortation]

Vapor ware I bet. Sounds to good to be true!

Suggests an interesting idea

[currivan]

I wonder if it would be any easier to make a binary-to-binary translator on the same architecture? The idea would be to translate legacy i386/i586 binaries to take advantage of the latest CPU extensions.

Any complicated self modifying code could be left the same if the program could at least spot it reliably. It might even be possible to translate to 64-bit at the cost of a few "x &= 0xFFF..." instructions around shift operations.

Sounds like BS, but maybe not... think microcode..

[divisionbyzero]

First, FWIW, the PowerPC version would only allow MacOSX to play Linux/x86 games, not Windows games, which is kind of strange. If they were going to lie, why not go all the way?

Second, the concept isn't totally outrageous. It sounds analogous to what was done with x86 chips. There is an x86 front end but a RISC-like core that works because of grouping instructions. With the right kind of abstraction it might work, but I'm not holding my breath.

For those too lazy to visit the site

[Anonymous Coward]

Transitive's Breakthrough Hardware Virtualization Products Eliminate the Need for Software Porting

"QuickTransit (TM)" Software Allows Applications to Run Transparently on Multiple Hardware Platforms with No Source Code or Binary Changes

Los Gatos, Calif. - September 13, 2004 - Transitive Corporation, the leading provider of software that enables transportability of applications across multiple processor and operating system pairs, today launched its QuickTransit(TM) product line, a family of products that allows software applications compiled for one processor and operating system to run on another processor and operating system without any source code or binary changes. The company's breakthrough hardware virtualization technology is unique because it provides 100% functionality, transparent interactive and graphics performance, near-native computational performance, and allows virtually any processor/operating system pair to be supported.

The first products available in the QuickTransit(TM) product line are:

*
QuickTransit for Itanium®: with support for MIPS®, POWER(TM)/PowerPC®, x86, and mainframe binaries
*
QuickTransit for Opteron®: with support for MIPS, POWER/PowerPC, and mainframe binaries
*
QuickTransit for x86: with support for MIPS, POWER/PowerPC and mainframe binaries
*
QuickTransit for POWER/PowerPC: with support for MIPS, x86, and mainframe binaries

The QuickTransit product line allows computer OEMs to rapidly increase the number of user-written and ISV (Independent Software Vendor) applications available on their platforms; it enables ISVs and internal software development groups to eliminate porting costs; and it allows IT services companies to migrate their customers' legacy applications to modern platforms at a fraction of the existing cost.

"The reality is that it's often difficult and expensive to port software applications to hardware platforms," said Bob Wiederhold, President and CEO of Transitive Corporation. "QuickTransit breaks the burdensome history of hardware/software dependency and therefore provides immediate benefits to the computer industry. QuickTransit dramatically reduces the cost, risk and time-to-market to support new or multiple hardware platforms, and it makes significantly more software available on computer platforms."

How QuickTransit Works
QuickTransit utilizes a unique and patented modular architecture. It runs on top of the operating system, with no end user intervention. As a translated application runs, the QuickTransit "front-end decoder" reads in blocks of binary code and translates them into an intermediate representation (IR). An "optimization kernel" then optimizes the code represented in the IR, and a "back-end code generator" encodes the optimized blocks for the target processor and caches them. QuickTransit's high performance comes from exploiting the fact that only 10% of the code in a typical application is executed 90% of the time. So, the optimizing kernel looks for frequently executed blocks of code and aggressively optimizes them as they are identified. The QuickTransit architecture is modular, allowing front-end decoders and back-end code generators to be easily mixed and matched for the source and target environment.

QuickTransit products support applications written in any language including C, C++, Fortran, Cobol, Basic, Ada, Pascal, Modula, PL/1 and assembly language. QuickTransit products let software applications run on the target platform exactly as they run on the source platform, with 100% functionality. Graphics and interactive performance are transparent, and computational performance is 80% of what could be achieved with a native port, which is often higher performance than is available on the original platform. Today's Itanium, Xeon(TM) or POWER processors, for example, offer 10 times the computational performance of mid-1990's mainframes. Using QuickTransit software, today's processors could

I wouldn't trust Enderle's word

[asoap]

He probably doesn't notice the performance hit because his comp is a Ferrari. Also he's praizes the quality of products that SCO produces, and we all know the amaizing stuff SCO makes.

-asoap

I believe it

[SlashDread]

Look, marketing is king yeah: "No Performance Hit", Enderle: "Moderns cpu's are fast enough to emulate each other".

What they MEAN is, CPU speed hardly mathers today, for ANY app.. I mean, sure a few photshoppers might like better hardware, a few beancounters who worship complex excel might even like dual cpu's, but for the rest of us, its how fast an app loads, and how fast it prints..

And if one app rules yor life, you buy native speedy hardware anyway. For the REST of your software needs it doesnt matter.

I have been wondering why there isnt a VMWARE for different arches for example, well maybe now there is...

"/Dread"

QEMU does this too...

[Nagus]

QEMU does something similar... it uses a just-in-time-translation approach to emulate different CPUs (currently x86, ARM, SPARC and PowerPC).

For x86-on-x86 emulation, it's only about 4 to 6 times slower than the host CPU.

It can emulate a PC complete with network card, hard drives, graphics, etc. Best of all, it automatically provides a firewall and DHCP server on the emulated network connection, so an operating system running in the emulation can access the "outside" network painlessly.

Apart from emulating complete systems, it's also able to launch binaries compiled for another CPU. In that case it uses dynamic loader tricks to translate library calls to native calls.

I believe the Darwine project plans to use it in conjunction with Wine to run Windows applications on MacOS X.

It's really really cool stuff!

So WinXP could run on an 8 pin PIC?

[Anonymous Coward]

If I am reading right what the poster says it would
imply that I could run winxp on a microprocessor
with under 128 bytes of ram with no lost in
performance.
It sounds to me like the poster or the guys
promoting the "emulator software" have been
sniffing some funny stuff or have been in
management or marketing too long.
It should be reminded to those folks that the
laws of physics can't be as easily broken as
the speeding laws.

VM's

[Anonymous Coward]

Now the company site isn't slashdotted, I had a flick through their site - they mention VM's quite a bit, and this got me thinking:

Maybe It's an OS with a built-in VM, for booting your 'real' OS. If this runs on all platforms, you can install the OS onto it transparently. You then get to install the new OS's apps, then any foreign binary gets converted into specific instructions by the transparent host OS.

If the host has good graphics card and sound (no mention of sound at all on the site) drivers then it will work fine. Good idea.

Re:How is this NOT a dynamica recomplier?

[bussdriver]

Actually, I have been wondering for years why nobody as not attempted to decompile code into an inter-language and from that into a compiler or dynamic compiling runtime VM like java.

I thought GCC worked around this concept, some abstract inter-language it then generates specific cpu code from that??

Processors have a great deal in common in their designs. So I can see it as a realistic approach to the problem. Startup time could be a big deal, but if you could recompile most or all the code, why not do that and cache the results to disk for next execution? I bet about 10% of the code can't be translated, in which case, include an interpreter code block to handle it in the generated exe. So then you lose "about 20%" when ever those small untranslated parts execute.

Then the process of translation could be quite slow for the '1st run' but not matter a whole lot.

Clear this up ?

[polyp2000]

Im a little confused as to how this technology allows Windows applications to run on Linux. (assuming x86 Linux) Wouldnt you need to have the rest of the Windows operating system within linux in order for this to work? I can see how great this might be cool for something like Wine on PPC (allowing v.quick x86 emulation / functionality) but I dont see how it makes windows binaries magically compatible with linux running on the same processor.

Nick ...

Re:Clear this up ?

[Hognoxious]

Wouldnt you need to have the rest of the Windows operating system within linux in order for this to work?

Don't some emulators, like old versions of WINE, work like that?

I just ordered it

[EastCoaster]

I just ordered the "QuickTransit for x86" so I can play my XBOX games on my Mac. Even better, I can run an x86 distro of Linux on my Mac. I am so out of control someone help.

oh just what we wanted...

[snellgrove2]

Great, so we can get all those lovely windows viruses!

Transitive = Transmeta?

[shirai]

It's probably too late to be modded up but doesn't this remind everybody of another company that claimed a similar thing? I almost thought it was the same company.

Transmeta used to claim how they were building a chip that could emulate any other chip using what amounted to software. Ultimately, Transmeta is as we know it: Not a high performance chip company but rather fills in the low power consumption niche.

I remember all the hoo-ha about how if they can get x86 to run at competitive speeds how native code on the Transmeta chips should blow everyone away.

At the time, the claims were just as dubious as this and, ultimately, Transmeta never did become the "as fast as x86" people; however, they do provide a valuable product to companies that want low power consumption in their chips.

This may be what Transitive becomes. A useful company, filling a useful niche, but not equivalent to their original outlandish claims.

Re:Enderle? Media Whore.

[Anonymous Coward]

"ainalist" Rob Enderle is the fly in the ointment. Why Wired News would even touch a vowel of this "Microshaft paid for, SCO whore", (Google: Rob enderle) is beyond impune. The shill.

It works. Please read the article.

[master_p]

For those that are suspicious, it is obvious that this application is not a an emulator, but a translator. CPUs have their own programming languages: machine code. The software they have take this code, translate it to some intermediate form (a pseudo-assembly language) then translate it to the target CPU's machine code. This is quite possible. Add a little emulation/translation of hardware calls, and voila.

love these claims "Without a performance hit"

[dindi]

I mean these kind of apps are mostly for the technically enabled people ...

If you sell a virtual machine (or emulator) the first thing to *NOT DO* is to state it runs without a performance decrease.

It will eat mem, it will use your CPU, and it will access your drives *SIMULTANIOUSLY* with the host operating system.

You can say: minimal performance decrease, and 99 percent of the applications, but when someone claims these thing I just get itchy ...

I ran VmWare, Wine, and several others ... what emulators do is the run some appse (more or less), what VMs do is run all apps, with the burden of 2 oses on your machine, and most of the time having problems with spec hardware (even a simple USB cam) ...

on the other hand they might be cool just do not SAY 100% apps and no performance damage

A few corrections

[Guspaz]

Their software, QuickTransit, is already shipping, just not to end-users. They claim they're already shipping to OEMs, ISVs, etc.

Also, they don't claim no performance hit; on their page oriented for gaming OEMs, they claim 80% of what you'd get for a native-recompile.

Anyhow, it doesn't look like they're going to ship a consumer model, so somebody is just going to have to licence this and put it into a product. Then we can all evaluate how good it is.

BTW: Their press release from today about QuickTransit is linked to from here:

http://www.transitive.com/news.htm

Java

[McLoud]

Nah, the virus was written in java. 'Cause java runs in any plataform.

But, wait

[itsNothing]

If you could predict when the company folds and goes out of business, does that mean that you can compute the transitive closure?

want to impress me?

[Cynikal]

show me doom 3 running on my windows pc.

Dynamite: A framework for dynamic retargetable bin

[Anonymous Coward]

Alasdair Rawsthorne has written a few technical reports on the technology behind this which includes some discussion on performance. See:

Machine Adaptable Binary Translation

The paper's a few years old, but I would definitely say this is not vaporware and I would expect them to have worked out most of their performance limitations since that time. Will it run as seamless as people here seem to expect? That remains to be seen. In the meantime, instead of immediately crying foul, read up on the subject and come to an informed conclusion. It looks promising. Search around, there's more up to date info.

Does this mean....

[punxking]

it will run Duke Nukem Forever!?

EMUALLSYSTEMS

[gihara]

hahaaha... sounds like a guy i met a long time in the deepness of the net REIROM who was quite a personality in the Emuscene. He was Often talking about his EMUALLSYSTEMS capable of emulation all video games/computers. LOL++

os specific system calls?

[big daddy kane]

what happens when a binary uses libraries that aren't installed on the host computer or makes a system call unsupported by the host os? a hardware emulator can't compensate for that unless it's running another os on top of the host.

idea

[itzdandy]

what if you took execution cache to a new level. you have a basic processor emulator that can emulate the entire target processor, even if it does it very slowly. then, you have a cache monitor that watches every execution and writes the results to a file. slowly, as all executions are done, the cache gets full and you no longer actualy execute the code, but pull the previous execution out of cache. now, you also take this cache data and compile it and append it onto the original program(after making a backup) so that upon next execution, that cache data is used again allowing the machine to run fast :)

UNISYS Already 0WnZ Them!

[Anonymous Coward]

Check out patent 5,896,522 -- how the hell is this company going to register something similar?

Done a few years ago - Project Odin

[paugq]

Project Odin does a binary to binary dynamic translation from Win32 to OS/2. And it does quite well.

Odin has been circulating for some years now (it was called "w32os2" when it started). Extending Odin idea to other architectures is not trivial or easy, but it can be done.

Finally!!

[logicnazi]

So I am a little skeptical of the cross OS benefits but the cross processor emulation is very possible and long overdue.

Technology like this has been availible for some time. There was even an emulator for the alpha chips that would often run code faster than native code on the same processor.

The reason this seems difficult is not because the problem is particular hard but because the solutions we commonly see are so bad. If you want to run interpreted code at any reasonable speed you MUST maintain a cache of translated instruction blocks. Identify commonly used code chunks and translate them into blocks of native instructions and remember run-time branch information. I simply don't understand why things like JVMs and so forth don't do this today. Sure it takes a little effort but it can actually produce faster code than native compiles (and certainly better than a distributed binary which must be able to run on any of a family of processors).

I fully expect technology like this to ultimately predominate. Operating systems will primarily be compilers and optimizers designed to efficently keep caches of optimized native code. At some point even our web servers will be written in an interpreted language (although at run time they will still mostly be jumping between cached native code snippets). I just don't understand what took so long.

Spaceballs?

[standsolid]

Phantom Investor: "I am your sister's ex-boyfriend's brother!"

Talonius: "So what does that make us?"

Phantom Investor: "Absolutely nothing."

Yeah, but will it allow...

[BitwizeGHC]

... my PlayStation 2 games to run on the Phantom?

this was accomplished years ago...

[the-build-chicken]

here

It is _definitely_ BS

[andrewmmc]

Has anyone actually read the technology overview? It's complete rubbish. I especially like the way they claim code optimization is performed relevant to how the user is using the application. Sorry, but no.

Claim-by-claim analysis

[null etc.]

Okay, I doubt anyone will read this post, as it's late in the game, but here's some questions that came to mind when I analyzed the text from Transitive's web site. In all examples I will use Windows as the native operating system and Linux as the foreign application, unless otherwise noted.

Comment 1.

When a foreign application is started, the operating system recognizes that the application needs translation and automatically starts Dynamite.

How does the operating system recognize that the application needs translation? I see three possibilities.

The first possibility is that the native operating system lets Transitive execute all applications, and Transitive decides whether to translate an application, or let the native operating system run the application unaltered. Integrating this functionality (properly) into an operating system would be extremely diffult, let alone multiple operating systems.

The second possibility is that the native operating system attempts to execute all programs, and only invokes Transitive if an unexpected application type is encountered. Due to the way that Windows only executes files according to file associations (".exe" = DOS/Windows executable, ".com" = micro-executable, ".bat" = batch file), this seems very unlikely, as Windows wouldn't even know how to execute "gimp" because it lacks an extension such as "gimp.exe".

Even if one was to rename "gimp" to "gimp.exe", Windows would attempt to load the "Windows Program Header" for the file, which would be invalid because it's a Linux application. Windows would then generate an error. Of course, Transitive could overwrite certain error handlers within the operating system to catch this kind of error, and then analyze the file using a "magic number" command to determine which operating system and what file type was under scrutiny. But then Transitive would have to overwrite these certain error handlers within all operating systems, a very unlikely proposition.

The third possibility is preconfigured "virtualization sandboxes". Virtualization software like VMWare assumes that for a virtualized system, all files and executables within that virtualized system will be executed according to the rules of that virtualized system. Ergo, if you're running a virtualized Linux system, any executable will be executed as a Linux application. Does Transitive require one to preconfigure foreign applications as existing within a predefined virtualized system? Either way, there's alot more to this than simply the "operating system recognizes that the application needs translation."

Comment 2.

Depending on the requirements for the integration, Dynamite can be configured with a wide range of options, including the ability to build "bridges" between translated code and code running native on the target platform.

This sounds like a fairly intensive process, given the number of operating systems and APIs out there.

Comment 3.

This feature has been used, for example, to allow translated applications to call a native accelerated graphics library for the graphics chipset in the target platform, delivering higher quality and speed than other solutions.

Does this mean that Transitive will take a Windows DirectX application and translate it to Linux OpenGL? Or does Transitive have it's own API, which is compatible with all video card drivers out there? For either claim, that's pretty impressive; a technology worth being bought out by nVidia or ATI. Considering the breadth and depth of all graphics libraries and versions available for numerous operating systems, that's phenomenally impressive.

Comment 4.

The front-end decoder reads blocks of instructions from the foreign application's binary and decodes them into an intermediate representation. The intermediate representation allows Dynamite to understand the highe

cool!

[sl4shd0rk]

I'm going to find a SPARC 2 running SunOS and emulate a Windows XP session through an emulated OS/2 desktop while playing Doom in a DOS window. If it doesn't work, I'm going after them for false adverstising ;D

Vapor?

[fred3666]

Can I run QuickTransit(TM) on my Phantom gaming console? And if so, will it play Duke Nukem Forever?

Cross-Processor, but not cross-OS?

[mh101]

I just looked through their web site, and never once saw it specifically stated that you could run Windows apps on a Mac, Linux apps on Windows, etc.

They did state numerous times though that "Operating system call mapping from any Unix/Linux-like operating system or any mainframe operating system to any Unix/Linux-like operating system is supported." Plus in a section talking about game systems, it's worded to indicate Xbox games could run on Xbox2. Not those exact words, but is seams fairly obvious. What other game console was x86 but is now PowerPC? :)

Anyway, this would indicate to me that this only works provided the target and host OS is the same, or similar, i.e. OSX is Unix-based, so it could run the Linux Quake as described in the article. The only thing that throws me off is the article talking about running Gimp on Windows. Methinks they were really using a native Windows binary there.

Actually so

[Sycraft-fu]

You can emulate a PC on a Mac, as Virtual PC demonstrates. Donig it at a reasonable speed, however, it a whole different matter. If you think that a Mac will emulate a PC anywhere near to full speed, well, go and try it. Be prepared to be dissapointed.

You can emulate a PC at a speed reaonable enough to do productivity work like using an office app that doesn't happen to be out for MacOS. You cannot run modern videogames.

Along those lines

[Sycraft-fu]

CISC also "won" in that now there are almost no true RISC processors left, even those eith a RISCish ISA. The R in RISC is for Reduced, and that was the idea. Real RISC chips like old MIPS chips did two things:

1) Reduced the number of instructions. They had basic math ops for integer and maybe FP if the chip supported it, load/store, branches, a couple bitwise operators and that was IT. There were very few instructions.

2) Reduced instruction complexity. Ideally, an instuction did one thing and one thing only. No messing with multiple data units, no messing with flage, just add two numbers and stop. Each instruction was very simple.

Well, both of these have gone out the door. Modern "RISC" chips have loads of instructions that are very complex, in a large part because of SIMD units. SIMD units are a great boon to modren processors because much of what is intensive these days can be done faster with SIMD. Thing is, they are complex instructions, and you need a lot of them for a good unit. So you basically have to throw away some fundimental RISC principles if you want them.

As the parent noted, this is fine. The way you write code for a chip and the way its execution works are not 1-1 related as they used to be. Chips break up and reorder the commands you give them to whatever works best internally.

It does, however, mean that the fanboy crowing about RISC vs CISC is pretty damn stupid. I mean if you want to talk advanced ISAs Intel is really ahead right now with their EPIC (sorta) on the Itanium 2. It's a more "advanced" and "modern" ISA than RISC or CISC, but we all see the lot of good that does against the Opteron with it's "ancient" CISC ISA.

What really matters, and indeed all that matters, is how fast teh chip runs the apps you want it to run. If it's the fastest for the money, then it's probably the best choice.

Hah!

[tmjva]

Would never run MPE or MPE/iX!

Check out their Board of Directors...

[Nate4D]

You'll see that most of these guys can be found elsewhere on the Web. A fair number of them seem to have a reasonably solid reputation, and also seem to have backgrounds that fit their current roles. Obviously, this doesn't settle the question of whether QuickTransit is vaporware in the least, but you have to wonder: Most of these guys have at least decent reputations in their old circles, it seems. Why would they go and trash those by making ridiculous claims about complete vaporware? I don't know that it's all it's cracked up to be, but I'd guess that at least Transitive's Board of Directors has a lot of faith in it. Which, of course, means absolutely nothing... I'd also point out that when you poke around the Web, you'll see evidence that some of these folks have been working on exactly this problem for a while... Like, since the mid-90s, if I understand everything I saw correctly. Just something to chew on.

Turing Machines

[atrent]

If my (brain) memory serves... Turing Machines can emulate other Turing Machines (search for theorems on your favorite CS book), the only problems are (ram) memory and speed.

Commodore 64, Cray 2- both directions bad

[billstewart]

Go ahead an run your Cray applications on your Commodore 64 hardware, and call me when you're finished. The other direction is worse - try running the important Commodore 64 apps on a Cray 2, and your frog will get run over the millisecond he steps off the curb, and the Space Invaders will kick your ass.

Just because the ship date isn't vapor, that doesn't mean the features work as advertised.

Could be useful

[Gax]

Though it doesn't appear to be a universal solution, it may bolster the PPC Linux effort by allowing it to run software compiled for the x86 platform.

While doing some background checking I found an interview from 2003.

Re:I just got a copy...

[Anonymous Coward]

- - - If the sun is a star, why can't I see it at night?

You can, but you have to go to Australia. Unless you live in Australia, then you have to go to England, or something.

Woot sig reply (totally OT)

[Anonymous Coward]

If you could guarantee the secure transmission of a OTP, then why wouldn't you send the message that way.

Because you can securely transmit the OTP whenever you want. Then you can send the message later, when you have interesting information, even if the previously secure channel is no longer available.

Ideally, if you're going to actually use OTPs seriously, you give the recipient a big stack of them, keep a copy for yourself, and then go through them together in sequence.

This could allow you to correspond securely with someone for years (depending on the frequency of the messages and the number of OTPs you shared) if you can just hand off the OTPs at some opportune moment beforehand.

Re:Woot sig reply (totally OT)

[Krach42]

God, I'm so sick of discussing this that I'm changing my sig.

First of all, if you have a secure channel to transmit an OTP, then you could very well send the uncoded message along the same channel.

Send a stack of OTPs? Ok, what if someone intercepts those, and makes copies? What do you do if someone finds these OTPs, and is then able to later decode the message?

The fact of the matter is that even with an OTP, the size of the message has increased by a factor of 2, and you've guarenteed that having one full portion of the message cannot retrieve the rest of the message reliably.

So, rather than attempting to mathematically attack the available encrypted text, you physically/socially attack the security of the OTP. And guess which of these ways better protects security?

Entirely honestly, I would much rather entrust my data to an incredibly difficult mathematical problem, than another person. That's because I know that social engineering is almost always the weakest link in any security system.

The only place an OTP makes sense is when transmitting over a channel where you can know for certain if someone listens in. Such as a quatum communications channel, where an observer will automatically collapse the message. In this situation, you can guarentee that your message has not been intercepted, and you could safely transmit an OTP, followed by the actual message, and be certain that if anyone listened to the cyphertext, that they would not be able to unencrypt it, because you wouldn't have sent it if you had known that the OTP hadn't been sniffed.

But then you run into the problem that, you sent a message the exact same length of the message that you were intending to send, and you know it wasn't listened to. So, the question becomes again. If you know your message wasn't listened to, then why didn't you just send the unencrypt message? You'd have saved time.

It's honestly a sort of Catch 22. Your message is only safe if the OTP is safe, but if you transmitted the OTP securely, then you didn't need to transmit it at all, because your message would have been just as secure.

There remains, and ever will remain, no way to communicate a message with another party while guarenteeing security if someone can listen to every channel of communication that you have with the other person. Is the OTP the best option for giving you the best chances at security? Yes. Is it impossible to break? Not if they can intercept/retrieve the OTP.

So I ask you again. If you can send an OTP through a channel that you know to be secure, then why wouldn't you just transmit the message through that channel, since you know it to be secure.

Possible use

[Anonymous Coward]

Not sure if somebody said this already, but it would be pretty remarkable to be able to run x86 code on an Itanium system. Even with a performance ratio of 80% as promised on the company webpage for some undisclosed translation...

AFAIK all implementations so far have been mere fractions of these kinds of speeds, and Intel has certainly been trying.

Still I would be willing to bet actual numbers in this kind of emulation will be far from advertised and this even when accounting for the fact that you have a native winXP for both platforms with identical apis.

Universal emulators

[sglines]

Geesh - anyone ever heard of p-code, perl, python, ruby all use "emulators" that allow once write run many applications. Why do we keep rediscovering the wheel?