Slashdot Mirror
OEMs Jump Onto Transmeta Bandwagon
Scooter writes "News.com is reporting that Diamond Multimedia has announced a Web-Pad product based on Transmeta's 3120 processor. The report also mentions that NEC, and possibly a dozen other companies are investigating similar possibilities. It's nice to see things taking shape for Crusoe so quickly. " For more details on the chip itself, check out our recent story.S3 has also announced development work that will be done with Transmeta. They are working on a "Linux-based Internet appliance".
This is going to be a great stock lifter for all these companies in this market. It is a great product, but even more so, to the big technologists it will be another bandwagon. + They'll be able to get Linux into their marketing more easily which again will lift their stock. There are myriad companies out there trying to work linux into their advertising and this will go the same way. Talking of which it appears we have just sold our coporate soul (what we had left anyway) to Freeserve...
Working for the (other) man
Now, let's just hope the Palm OS will run on one of these... Unless something like LinCE picks the thing up. PalmOS now supports colour, and has a huge software library.
Now that there's a low-power CPU available, I'd love to see a Palm or Visor with a Crusoe CPU, a colour screen, MP3 playback and recording, WAP, etc... The beast still running on 2 AAA batteries that I'd change once a month.
I think it'd be quite cool. What d'ya all think?
Max
-- It's always darker before it goes pitch black.
A Web pad -- an appliance for browsing the web... Okay, I can see the market for those. But then: price from $500 to $1000??? Battery life of 4-5 hours only???
Sorry guys, you have to do better than that. A
web pad should cost below $200 and run for the whole day (~12 hours) to be viable (IMHO and YMMV of course).
Kaa
Kaa
Kaa's Law: In any sufficiently large group of people most are idiots.
I'm sure this is not the first of such moves we will see. However I would also think that such moves are not just driven by the quality of the hardware. I think many companies may be aware of the attention the Transmeta's 3120 processor got from the involvement of Linus (I tend to refer to him as Mr. T since I can't spell). If the loyalty shown to Linux can be carried over to processor sales, or hardware that is using those processors, that's big bucks. A fact that was not lost on Transmeta I'm sure (not to say he didn't deserve the job). No doubt we'll see more such announcements.
"Run a piece of Linus in your PC!"
--
--
E2 IN2 IE?
I wouldn't be surprised to hear a lot more companies start offering Crusoe-based systems over the next few days or weeks. I also expect to start seeing major chip manufacturers slash prices, to avoid having their markets vanish.
(What's the point in maintaining multiple product lines, with different processors, if the Crusoe will do the job of all of them? It's much cheaper to use a standard one-size-fits-all package, and tailor it with software patches.)
Depending on the performance of the Crusoe in the field, expect the Itanium (whenever it actually gets released) to sell for only a fraction of what Intel usually try to fleece off it's customers for new releases. If they don't, it's dead. It's going to be slower than a P3, anyway, and companies are going to start asking if it's not better to just upgrade to the Crusoe instead.
I imagine the Alpha (horribly overpriced as it is) to take a knock from this, too. The ARM and the Sparc should be OK, as they're both RISC (so will be faster) and they're not -too- outrageously priced.
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
... when they said they said that they weren't announcing OEM deals to focus attention onto the chip itself. Its going to take more than Diamond to make Crusoe work but its a good start. And its good to see Diamond innovating again after the disaster that is RioPMP500.
h tml
Interesting article talking about the impact on other chip/OS companies.
http://www.uk-invest.com/homepage/breakingnews.
Please stop using the Crusoe logo to represent Transmeta. The Transmeta logo is a downward pointing blue triangle with a wavy line crossing it.
While I have to admit that Transmeta does use the Crusoe logo on every page of its website, and only rarely uses the Transmeta logo, they only use the Crusoe logo with the Crusoe name.
Work for Change & GET PAID!
I don't want a "webpad" with a pen interface. I want a wireless, diskless laptop running as an X terminal to my PC. Think about it:
-Runs on Crusoe so it is quiet, cool, light
-no disk so it is even quiter, cooler and lighter AND there's no need to sync with the mothership AND it is more robust (HD's are delicate and often fail)
Obviously this is fairly useless once I leave the house (until wireless Internet gets popular), but so what? I'm sick of being stuck in one location while I'm browsing/programming/reading-docs/looking-at-porn
Furthermore, with some intelligent design you could even get rid of the keyboard. Put 8 buttons on the back of an LCD screen (4 on left and 4 on right). Your fingers would rest on these keys as you hold the pad. Pressing them in "chords" causes characters to appear on the screen--just like typing only using more than one finger at a time. Even using only 2 fingers at a time you get 256 different characters.
BTW, if this device is ever patented, the above is "prior art"....
---
This comment powered by Mozilla!
Linux MAPI Server!
http://www.openone.com/software/MailOne/
(Exchange Migration HOWTO coming soon)
http://www2adm.transmeta.com/crusoe/download/pdf/T M3120_DataSheet_1-18-00.pdf
and
http://www2adm.transmeta.com/crusoe/download/pdf/T M5400_DataSheet_1-18-00.pdf
and the chips are impressive. But the data sheets are very incomplete (the TM3120 data sheet is only 6 pages !!). No pinout, no electrical specifications, no programming info, no absolute maximum ratings, etc. They need an additional 1000 pages of documentation before you actually can design anything with it.
RFC1925
Transmeta is not a publicly traded company, and so is not required to disclose ownership information to the public. It is known that Paul Allen's Vulcan Group has money in it; other venture capital groups were mentioned in the presentation yesterday; presumably the employees have gotten offered some equity whether as stock or as options on stock.
The stock could certainly head up from zero, but presumably the venture capital folk would want to have some return on their investment, and so it would take a fairly high valuation for selling out to prove worthwhile.
If you're not part of the solution, you're part of the precipitate.
"It's nice to see things taking shape for Crusoe so quickly"...
It's been five or more years, hasn't it?
...j
I wonder if Dell will using the Transmeta chip? While I don't pretend to understand the details of chip design, it seems like OEMs would be missing out on a speedy, cheap, low wattage chip if they ignored Crusoe. Also, we know Dell has been chomping at the bit to find other products to help them maintain their growth and profit margins.
On the other hand, Dell has always enjoyed "Best Customer" status from Intel when their supplies are running low while disloyal companies like Gateway get the shaft.
If you were Michael Dell what would you do? Would you enter a high growth area and risk your PC business or stay with Intel all the way?
That which does not kill me only makes me whinier
If Diamond indeed ships out a Webpad running Mobile Linux (their page lacks any press releases at this point), I think they surely deserve some sort of Hypocracy Award, if not an annual award than a 'Lifetime Acceivement' award. It would be truely ironic for a company to ship a webpad that only runs linux, after years and years of outright denying specs for things as trivial as the Rio uploading code (claiming it was "valuable intellectual property") to things like specs/help on video card drivers (based on S3 chips, the big help was needed because Diamond deviated in such fscked-up ways from the reference design).
;-)
Yes, Diamond deserves a nice trophy anyways, but if they really plan to ship an entire computer now (mobile or not) running the linux operating sytem when they have yet to support a single project under linux.. well, let me just say it is too early in the morning for me to fully imagine what the figurine at the top of the trophy should look like, and what it will be doing
I was REALLY excited about the Crusoe chip when speculation first arose that it would be able to run executables from multiple hardware platforms. Then came the speculation that Crusoe would be aimed at the mobile market (I lose 50% interest right here).
Now that Transmeta has made their announcement, I have no real interest left in it. They've done a nice job of diverting attention to it's power consumption, and that's about it. It's x86 compatible...so what? It's theoretically compatible with lot's of stuff, so pony up. I wouldn't buy one of these JUST because it's not Intel (I don't use PDA stuff, so that's not a big drawing point for me). My MAIN interest was in being able to run applications written for Apple computers, or Sun workstations. And then you have the benchmarks...nice fluff work there. Combining their low power consumption (see, here it is again) with performance doesn't even come CLOSE to giving a fair representation of how it stacks up against a comparably clocked Intel or AMD chip.
And of course, the price on those web "appliances" (aren't you sick of that term?) is still too high for what amounts to a big kids toy. You're not going to get any real work done without a keyboard (in most cases) so basically what you are paying for is a $500 - $1000 Rolodex with some added functionality. (And yes, I know, virtual keyboard or whatever they called it. Did you ever try to do any touch typing on the old Atari 400 membrane keyboard? If so, you know my objection here)
I really think notebooks are going to be where Crusoe blooms, if for no other reason than to not melt your lap while your working with it. But is $329 for the 700MHz really any kind of a bargain?
...not that is it reasonably low-powered. The interesting thing is that Transmeta is actively trying to prevent people from writing in its native instruction set, thus creating historic compatibility problems. If successful, Transmeta will be able to change chip architecture, instruction set, etc. etc. without breaking any existing applications.
This is a big thing -- consider that PIIIs and Athlons still have to be able to pretend they are 8088 processors. If you believes that does not put a huge cramp in their style, think again. If Crusoe manages to be free from this limitation, it could evolve much faster than the usual CPUs. That could be a decisive advantage several years down the road.
Kaa
Kaa
Kaa's Law: In any sufficiently large group of people most are idiots.
I absolutely love the idea of Webpads. In fact I'm posting this from my Sharp Triapd connected via wireless ethernet. I would never buy a new device from Diamond though. Since the technology is new there is a greater than average chance that you'll have to send your device in for RMA work. As anyone who has dealt with Diamond's RMA department can tell you... there is a good chance you'll never see your device again after you send it in for warranty work. You'll be out-of-pocket for the cost of the device and have nothing to show for it.
d
www.HearMySoulSpeak.com
Ummm...sorry, but just because Transmeta finally decided to share their toy with us publicly doesn't mean that it's the first time any of these companies is glimpsing it. I would suspect that those deals have been ready to go for a long time, and companies like Diamond have merely been sitting under NDA waiting until the Big Day. On the subject of quickly, how long has Transmeta been around? In theory, the marketroids could have cut the deals before work on the chip even began.
www.HearMySoulSpeak.com
How dare you... linking w/o attribution. I've sworn his articles off. Now I feel dirty.
I just had a friend write to me and ask about a company called Seligman Communications & Information. The reason he asked was because he thought this company owned Transmeta??????
Well that is not a company, it is a mutual fund. How on Earth did he hear this, especially considering he works for IBM?
Excuse me, but why were they supposed to spend time and money developing drivers for Linux? To support the small market that would develop them themselves anyway, at no cost to Diamond? I don't think so. Not a smart business decision.
Now using Linux on a webpad... Well, it's free. Much cheaper than Win98. That's a smart business decision. Maybe they are hypocritical from your viewpoint. But from the viewpoint of their investors, they're quite consistent. They always act in the way that will gain them the most money. That's how public businesses are REQUIRED to operated.
...the similarity of the look between the WebPad and the PADD's from the sci-fi series...
"Captain's Log, Star........"
(A ramble, I know, but I'm on early shift, and it's only 20 mins to go home )...
...and all the related plug-ins. Every time a story is posted on Slashdot that link to a .mov file or a RealAudio file, people bitch because the appropriate software isn't available on Linux. It's surely not available on StrongArm! But it is availabe on Windows. So to really get a quality web browsing portable at this time, it has to be X86 compatible. Which leaves StrongArm in the cold.
Arrrrrhhhhhhggggg....
I can't disagree that there will be many detractors to Transmeta, and it will be a bumpy, uphill ride. But I wouldn't really take any opinion of Jessie Berst as an indicator of what these problems might be. Heck... I wouldn't take anything that blowhard says seriously. Jessie Berst is nothing but a load of sensationalistic crap.
You were thinking of the, "Head up ass" reward?
:-P
Cheers,
Ben
My usual seat in the cluetrain is at A HREF="http://pub4.ezboard.com/biwethey.ht
In the press conference yesterday TransMeta said they have had customers lined up and ready to introduce products for quite some time. My guess is, they've been working with them for at least a year, under a VERY tight NDA, so that TransMeta could pop up at the conference with plenty of goodies to show off - and in the afterglow, those partners could say "here we are! we're on the trolley!"
Subliminal ads in the Linux kernel. If "Buy Transmeta(tm)" showed up for a few microsecs every now and then, don't you think their sales would increase?
They should force Torvalds to do it.
[joke -^ ]
~~~~~~~~~
auntfloyd
I don't know about you people, but I'd love to have one of these bitches... something about the combination between a wireless network connection and streaming porn just makes me want to spend $1000.
:-)
"Software is like sex- the best is for free"
-Linus Torvalds
Honestly, I thought Transmeta was going to announce their intentions to destroy the world by rereleasing "New Coke" in the form of a giant stream from outer space unless the monopoly known as Microsoft was brought to their knees and humbled. Thoroughly.
But I guess a webpad is cool too.
ALL HAIL BRAK!!!
According to this article, NEC is the only company mentioned as currently evaluating/investigating the chip.
Yeah, this may seem like nitpicking, but I find it a bit frustrating. I sometimes do not have the time to read through the referred articles, and instead just quickly scan the Slashdot headlines (and no, I do not post a reply about stories if I have not read the referred article.) If I had done that with this story, I would have been misinformed.
> I thought the main benefit of the Crusoe was that it could emulate all other processors.
Not the main benefit, no.
Crusoe does appear to isolate the virtual machine platform (currently an X86 emulation) from the hardware platform. This has a couple of interesting benefits. The first is that your underlying hardware can change at any time without breaking your software (this includes the OS and all applications.)
So while it is theoretically possible to emulate many chipsets, in most cases there's probably very little reason for Transmeta to expend the development effort required to do so. A possible exception is Java. Crusoe is imho exceptionally well suited to inline Java emulation (probably using the X86 memory management hardware to mark specific pages as Java bytecode). Interestingly enough, Transmeta acknowledged that Sun's work on dynamic compilation for Java parallelled a lot of their own development of the code morphing software.
So you won't see Transmeta's marketing strategy promote the Crusoe as a general purpose cpu emulator. They're very specific. They want a slice of the X86 market.
To me, Crusoe's main benefit is the ability to break away from the X86 architecture while maintaining compatibility with existing operating systems and applications.
Consequently, Transmeta can implement the actual computing engine using any architecture they wish. They do take a performance hit but it appears acceptable and performance seems likely to improve as development continues on their code morphing software.
Assuming performance is acceptable to the majority of OEM's, a couple of interesting possibilities suggest themselves.
One (raised by Transmeta at the briefing) is that bugs in their X86 implementation can be corrected by updating the code morphing software. This allows correction of (for example) floating point division flaws by downloading a cpu patch over the Net. Such updates can be generated by Transmeta extremely quickly.
(Of course, bugs in their underlying VLIW hardware will of course require a new silicon pass to fix properly, but there is the interesting possibility that they could release code updates to work around such flaws anyway.)
The second, is the potential for OEM's or other interested groups to lobby Transmeta for the creation of custom opcode extensions to the X86 chipset. Such extensions could then be distributed directly with the software that required them.
(I can see it now: Requirements - Pentium II 500 or Crusoe version 2.62 or higher)
> The power consumption is good, but so is the power consumption of a StrongArm.
Well the StrongArm has a somewhat lower software base than the current Intel dominated market. Much as I like the StrongArm, I think Crusoe will annihilate it.
> What benefits does the Crusoe have? Its not like we need to run Excel on a WebPad.
No, the WebPad runs mobile Linux. The idea is to have a web enabled device that is low power but can also leverage the Open Source behemoth that is Linux. Once the WebPad hits critical mass (or possbily even before), I can see the number of WebPad compatible applications exceeding the number of Windows CE compatible apps by several orders of magnitude.
The 5400 is designed as a replacement for Intel's notebook cpus. In addition to the base flexibility of the Crusoe processor, the dynamic power management allows the chip to run cool, saves battery life and eliminates the need for a fan in your notebook (which further extends the battery life).
Transmeta's marketing people estimate battery life of about 200% over a standard notebook pc. This is probably just wild speculation on their part, but it sounds like a fair estimation to me.
At the press conference somebody mentioned in passing that the web pad had a USB port that you could use to add (among, presumably, other things) a keyboard.
The pop-up virtual keyboard on the touch screen is probably sufficient for truly mobile use -- e.g. when you're holding the pad in one hand (kinda like a clipboard) and just want to one-finger type in a URL or a few simple commands. For serious typing I'd rather have a real keyboard and something to rest my hands/arms on anyway, which means putting the thing down on a table, etc anyway.
(These things are almost at the tech level of the pads in "2001: A Space Odyssey" -- recall the scene where Bowman and Poole are eating while watching the TV (?) broadcast about their mission, each watching on his own web-pad-like device. At least something from that movie came about in the right timeframe. Too bad it wasn't the orbiting hotel and the bases on the Moon.)
-- Alastair
I think my village would rather have basic necessities met first, such as food, shelter, medicine, etc.
Give a man a fish, you feed him for a day. Teach him to fish...
Why do think such things are lacking? Knowledge is the key to fixing these things, and as long as the finite number of agricultural experts, civil engineers, doctors, etc have to travel to each location to help the people, they are limited severely. But transporting such devices to the villages in question so that villagers from several villages can consult with the appropriate expert all at the same time, or look up references to such information when an expert is not available...
Its amazing how people will find problems to apply technology to, when in fact, they should be trying to find technology to solve problems.
I realize that information is not the same as knowledge, but knowledge can be transmitted through information, and any technology which lowers the barriers to doing this is a very significant part of the solution to such problems.
It's amazing how people who want to help other people are so often not capable of seeing the forest for the trees...
Chris
San Francisco values: compassion, tolerance, respect, intelligence
The most exciting feature, for me, of Crusoe is code morphing. Reading the white paper on technology behind the chip (something that *a lot* of posters here should do before posting) got me excited even more.
...(do some stuff with %register)...
Basically, after a piece of code is translated to native code and optimized, it is cached. Next time it is executed, if it's still cached, the already translated and optimized verison executes.
The benefit of this is speed. A lot of people doubt this speed, saying things like "an emulator can't possibly run at 75% speed of the native system", etc. There are two reasons why Crusoe can outperform the native system, one of which is really not apparent and ignored by almost every person that criticizes Crusoe.
The main thing to remember here is that Crusoe has some radical, very different technology decisions.
First, as any experienced software engineer would point out (backed by experimental data), 90% of a program's execution time is spent in 10%(!) percent of its code. What this means is that if ONLY that 10% of the code is optimized, it will speed up 90% of program's execution time. Crusoe's code caching mechanism helps this immensely because as a program runs, these 10% become cached in native code and translation from non-native machine code is done only ONCE.
You may be saying, "So what, in the best case, the program will run almost as fast as the native system, but it simply can't beat the native system." That's where you're wrong.
The second reason is that the software layer not only performs translation, but optimization as well. You may now object that if the original program is optimized by the best optimizers, Crusoe's optimizer can't do better. Well, it can because of Crusoe's architecture. Note that, for example, x86 processors have a small number of registers (which are areas for data stored internally *in* the processor; such data is accessed the *fastest*). Crusoe's VLIW architecture, however, has a lot more registers and its out-of-order pipelining, branch prediction. Also being a very-long-instruction word processor, it executes a lot of small instructions (atoms) in one big full instruction (molecule). Molecules can be executed in parallel (pipelining). Crusoe's optimizer takes advantage of these features, making the translated code use more native registers, instead of accessing normal memory or L1/L2 cache (which are slower) and groups code to be processed in parallel.
Crusoe's optimizer performs really aggressive optmiziation. Perhaps the neatest feature is how Crusoe handles aliasing. Here's some pseudo-assembler code that loads from the same memory location twice:
load from %X to %register
store %anotherregister to %Y
load from %X to %register
add %register and something else
etc.
This is the tightest optmiziation a compiler can perform. The compiler can't eliminate the second load operation to the register because %Y may be an alias for %X (that is, %Y may point to the same memory location as %X). Such aliases come up rarely, but they can come up, and so the compiler can't risk eliminating the second load instruction because it can't predict whether %X is an alias for the %Y. Nobody can, not even the processor.
Crusoe takes a radically different approach in this situation. Its optimizer ELIMINATES the second load operation, assuming that %Y is not an alias for %X. However, in case it is, it marks an internal bit that protects %X from being overwritten by the store instruction. So the code that one ends up with doesn't have that load instruction and when the case of %Y being an alias for %X does happen, it simply generates the extra load instruction on the fly.
This may seem like an insignificant optimization, but in reality, it can be quite significant since things such as these happen in programs very often (and often %Y ends up being not an alias for %X). Elimination of extra loads permits better pipelining (more code executed in parallel), and an extra load may take quite a bit of time if the load has to be done from the memory.
There are a whole bunch of cool other things about Crusoe's technology which makes it a great all-around processor.
So, what this means is that thanks to the revolutionary architecture, Crusoe's optimizer can optimize that 10% BEYOND the original and actually run faster.
Users of computationally-intensive programs will especially benefit from this. For example, a 3d ray tracing program spends a lot of time in the small, tight rendering code. Having that optimized so well by the processor can have a significant effect.
Crusoe also uses filtering techniques to avoid caching code that is executed once-an-hour (thereby preserving translated native often-executed code in the cache as long as possible).
As the website mentions, most benchmarks only measure a bunch of tasks done in 10 or 20 minutes. The website asks: do you really repetitively do 10 different tasks on your word processor for half-hour or do you actually sit in front of a processor and type most of the time? This is indeed a valid rhetorical question.
Most benchmarks are too short to let Crusoe speed things up as much as possible.
Although I don't like the "mobility features" that Transmeta keeps pushing every other sentence (damn marketing) and I don't like the fact that their benchmarks mix performance with "mobility features" (even though there is some validity in doing tat), I think that Crusoe is a very exciting technology and wish I had one.
Stop thinking in terms of megahertz. As processor technology gets more advanced, all these things stop mattering. In one app, your 700Mhz AMD may perform much slower, in another it can perform much faster. It's never same speed all the time.
Did you see the slashdot plug on the Transmeta 'going mobile' page. It just gives ya the warm and fuzzies... Ahh.... Now where'd I leave that The Who album with "Going Mobile" on it??
-- What you do today will cost you a day of your life.
We need to consider just one thing. If we were releasing a new chip, capable of running any other (theoretically) architecture's instructions, and thie feature being in upgradable software, with the intention of putting something on the market in reasonable time.... Which ONE architecture would we choose to support?
:) That is, no VM layer.
I'm not at all devastated by the fact that 86k, PPC, Z80 or whatever else isn't supported. In time, the smaller segments of the market will either be supported, or will convert.
What I would really like to see now is direct Java bytecode support.
As for the power consumption aspect taking center stage, it's right to do so. The general public and casual user of portable devices isn't hampered by poor performance, they're hamstrung by battery life. This is HUGE.
Mobile users have never, EVER considered running multi-platform apps on the same machine. Code morphing is wizz-bang, but not much else to them.
Hey, will this thing multitask platform specific apps? Or is there a reboot needed for platform switches?? Anyone figure that one out yet?
-- What you do today will cost you a day of your life.
Don't count your chickens before they hatch... and don't count them twice before they hatch... :-)
This idea has been discussed to death over the last years. There are a few devices like this already shipping but they are very expensive. They used for special application like the healthcare industry (yes, they replace the doctor's clipboard).
The problem is not coming up with ideas, any slashdotter can do that. The problem is implementing them in a way that works and at the right price.
----
Stop worrying about the risks of nuclear power and start worrying about the risks of not using nuclear power.
Umm, I'm not sure why it's an "interesting rumor" that BeOS is running on systems built with a chip that, as Transmeta says, is "Fully x86 compatible" and "is compatible with the complete range of x86-based operating systems", given that BeOS comes in a version that's "x86-based".