Slashdot Mirror
Why Can't Intel Kill x86?
jfruh writes "As tablets and cell phones become more and more important to the computing landscape, Intel is increasingly having a hard time keeping its chips on the forefront of the industry, with x86 architecture failing to find much success in mobile. The question that arises: Why is Intel so wedded to x86 chips? Well, over the past thirty years, Intel has tried and failed to move away from the x86 architecture on multiple occasions, with each attempt undone by technical, organizational, and short-term market factors."
Intel is still the major manufacturer of laptop, desktop, workstation and server chips...
What if they're not the main provider for cheap toys? It's mostly a matter of price anyway. Whatever they do, Intel chips will always cost significantly more than ARM chips due to their business model.
Really? I mean the Atom line processors are pretty great. The technology is well developed both for hardware and software and Intel basically owns that market. Why would they want to kill it off when they're still making money hand over fist with it?
Much like redmond can't ship good software. Despite all those ultra-smart people both employ. Curious how that works.
Because there's no point in killing it. Also, most companies are not going to spend the money to port and/or rewrite all their code. There are also a shit ton of legacy apps for x86 that there is no reason to replace.
This has been true for decades. Technology wants to evolve from CISC to RISC. The x86 brilliantly hid this by translating CISC to RISC superbly,
But once you lose the x86 tag Intel would just be one of many vendors. The closest thing to competition they have had for x86 has been AMD.
What intel needs is a superior architecture that can successfully microcode intel instructions with minimal performance cost.
If you have such a huge install base and abandon it for a new architecture you are going to lose portions of that install base. Duh.
This is /.
I'm sure we can find a way to blame Microsoft or Windows, this is an easy one! /sarcasm
There is plenty of market outside the mobile market. Use the best architecture for the job. This is a moronic trollish non-story designed to incite debate. SoulSkill can go fuck off this site.
Discuss.
Intel is the one to beat.
Ask AMD, good luck with that.
Intel (with pressure from Microsoft) long ago decided that backwards compatibility was a sacred cow. The profit opportunity they found -- namely, having millions of cheap PC clones which can all run the same software, even if that software was written in 1989 -- was judged to be worth the extra engineering "drag" of having to support ancient operations and addressing modes. There is still hardware in the Core 2 that exists only to support opcodes from the 8086! How much money, I wonder, has Intel spent on engineering, all just to avoid having to tell the Windows fans that they'll have to stop using any program that's old enough to visit a bar?
There's not much mystery why they can't move away from X86. They consciously made a profit-seeking management decision that shackled their ability to engineer radically. If they didn't want their engineers to be shackled, they'd cut of all the old baggage that keeps them weighed down. Until they take that step, all their talk is just hot air. But I suppose hot air is the norm at a company that created the Pentium 4.
They could drop 32bit at some point, but I don't think even the legacy instruction sets hinder them much.
there is still demand. x86 just works.
That said, I didn't buy any Intel Products in Years. The last thing was one of the early Netbooks. I'm disgusted by UEFI, High Prices and unlock-performance-with-a-code. So.... I propably won't buy any x86 Hardware in the Future either, but that's just me.
http://www.anandtech.com/show/6529/busting-the-x86-power-myth-indepth-clover-trail-power-analysis/4
Because killing is criminal you dummy... :P
The reason has something to do with the billions of x86 chips currently in operation in the server/desktop/laptop market and the massive amount legacy software written for x86. Intel tried to implement a new non backwards compatible CPU architecture before, IA-64, and it failed to catch and the backwards compatible AMD 64 bit x86 variation winning out.
the question is idiotic. sounds more like "asking a question just to ask it". Why should even intel kill x86? Would anyone even WANT to kill his cash cow ? It sounds more like wishful thinking from the camp across the atlantic ( arm *wink* *wink* ). Sure they would like to initiate or induce an inception of such an idea, but Intel has no reason at all to abandon such a successful platform.
He didn't have to deal with an installed base.
Windows, Word, Excel, and Games.
Microsoft is just starting to make cross hardware platform applications and development. So we have decades of legacy software that depends on the x86 architecture.
Back in the 90's when Java Was becoming Popular, Microsoft put an end to that, and gave us .NET that runs slightly faster than Java but only works with windows on x86 and didn't put any effort in making cross platform, trying to keep a hold on the market. If apps could start working cross OS's and Hardware platforms then people will no longer want Windows, or more to the point, they could choose not to use windows.
If something is so important that you feel the need to post it on the internet... It probably isn't that important.
Because the world runs on legacy software, and that legacy software runs on a legacy platform called x86. The answer is really that simple.
You can come up with a superior platform for power (ARM), it has been done and it worked really well on phones where there wasn't a large legacy base of software already in place. You can come up with a superior platform for 64 bit processing (Itanium), it has been done and it worked really well in a very limited marked (servers that handled large databases). However that market was too limited and large lawsuits have been filed to try to get out of that market.
Other examples abound and have been made, the payoff to whoever could succeed would be in the billions of dollars (Even the Chinese are trying their own homegrown CPU architecture). Every single one of them that has tried to enter the desktop market has failed though for the simple reason that it couldn't emulate x86.
Even Microsoft would dearly love to get out of the x86 business, the payoff in terms of killing legacy software support and selling all new software would be huge (hello Surface RT). I think you'll notice that sales of Microsoft RT products have all been a dismal failure with manufactures declining to make new products as fast as they can.
Until you can build a chip that can emulate x86 and support a different architecture and do so more cost effectively than just an x86 chip x86 will live. You can't kill it, Intel can't kill it, AMD can't kill it, Microsoft can't kill it and you sure as hell can't nuke it from orbit. It's embedded in billions of computers and software programs worldwide, and that is a zombie army that you just can't fight.
Vendor Lock-in.
...or is that three
AccountKiller
A single CPU architecture across operating systems and devices has worked out well for consumers. There are a wide variety of operating systems and user leve software on that platform. We'd need a viable cross platform application VM architectures, JVM/CLR style, if we want to avoid application islands.
95% of the processors on tablets and smartphones are ARM processors. ARM Holdings licenses out ARM to a number of chip vendors. In theory, Intel could license ARM also from ARM Holdings and start to manufacture ARM chips. Given the difference in margins, it is unlikely they will do so until they feel there is a significant threat to the business. Even better for Intel (in terms of non-x86 revenue) would be a cross-licensing agreement with ARM that gives Intel a slice of the ARM pie. So, it is not impossible for Intel to compete in the non-86 market, it is simply very difficult to establish a new processor architecture and gain significant market share. The ARM architectural roots are nearly as old as those of the X86 architecture.
This is not to say that Intel has not blown opportunities in the past, but a new architecture today would be very difficult. Intel has deep pockets, but were Intel successful in a new architecture today, it is plausible that US monopoly regs would stomp on Intel for using existing money to develop a new market.
As the parent mentioned, Intel DID come up with a clean sheet chip that threw away legacy. And guess what? It gets regularly trashed here for not being x86 compatible. WTF Slashdot? So you can't have it both ways Slashdot. Either you discard legacy, or have it. Make up your mind.
Cisc instructions still have a use where on risc you have to everything yourself
As much as people like to dig on x86, ISA simply doesn't matter. The benefits of programmer familiarity and tools infrastructure (not to mention installed base and compatibility) dwarf any possible technical advantage (of which there are few)
the reason i use ARM on my iphone, ipad or android phone is that there are hundreds of thousands of applications to choose from to do different things
every non-x86 platform for the desktop market has had a lack of software. the OS is useless by itself.
The last attempt Intel made at a non-x86 architecture was Itanium.
In 1995.
And it wasn't an attempt to ditch x86. The Itanium was a server product from the ground up, and only partially a technology vehicle for VLIW because HP (the partner at the time) largely drove that aspect of the ISA.
This article is pointless. The RISC/CISC debate is moot. Or, more aptly: an academic exercise, free from real-world constraints.
https://www.accountkiller.com/removal-requested
If Intel moves away from x86, it just becomes a fab company. Intel has much IP in the x86. The x86 is Intel's bread and butter.
It was called Xscale and it was among the best at the time. They sold it to Freescale (I believe).
I think the unique ability to run existing x86 (or recompile manually optimized for x86 c programs) at native speed is one of the reasons many of us stick with Intel architectures, it gives them a very strong tactical advantage. It would also alienate many strong key advice givers in the machine purchasing decision making process if their home Intel game collection was suddenly nixed by a capricious ogre like hardware maker.
The purpose of existence is to make money.
These articles are constantly missing the point.
x86 is fine. The flaws of the architecture are mostly superficial, and even then, x86-64 cleans a lot of it up. And it's all hidden behind a compiler now anyways - and we have very good compilers.
ARM has an advantage in the ultra-low-power market because they've been designing for the ultra-low-power market. Intel has been focusing on the laptop/desktop/server market, and so their processors fit into that power bracket.
But guess what? As ARM is moving into higher-performance chips, they're sucking up more power (compare Cortex-A9 to Cortex-A15). And as Intel is moving into lower-power chips, they're losing performance (compare Atom to Core).
The ISA doesn't really affect power too much, as it turns out. It affects how easily compilers can use it, and how easily the chip can be designed, but not really power draw or thermal performance. Given the lead Intel has on fabrication, any slight disadvantage of the x86 architecture in that regard is made up for by the software library.
Close Source Applications.
They're not stupid like Microsoft is, they know that closed source and multi-arch don't work together.
"When information is power, privacy is freedom" - Jah-Wren Ryel
.. of computational horsepower. The idea that one architecture rules is nonsense. Videocards used in X86 systems are not x86 processors. The truth is computing has increasingly undergone specialization. It makes much more sense to specialize computational units towards places it makes sense.
It's already a bad day for Redmondians. Haswell is slated to be introduced in 2014 will mostly offer the BGA designed Broadweil "System-on-a-Chip CPU", pre-sodered on an Intel motherboard like Atom chips are now. There will be nothing to upgrade - in effect this will be a device in PC clothing. There are rumors of high-end LGA packaging, but the upgrade possibilities will be limited to a few paltry offerings. No one will be making consumer upgradable parts anymore. Another way of saying it is that It will become cheaper for Dell just to replace the whole "PC-thingy" than to repair it. Yet Another Way... Intel's Ivy Bridge product cycle ends in 2014. Its successor, Haswell, will not have a desktop chip. The English story: http://semiaccurate.com/2012/11/26/intel-kills-off-the-desktop-pcs-go-with-it/#.UTU5hjZMn2A As tablets and smart phones replace desktops and notebooks, Intel, Microsoft and the desktop manufacturers struggle for market-share. The end of the desktop in 2014 does not mean the demise of the notebook, or of Microsoft, or of the support jobs they bring. It does foreshadow their end though. This time its a question of what and who will be left behind. Intel's market-based decision will shrink the computer field in general, and IT departments everywhere. With a paradigm shift away from a smart-client/server model to a dumb-portal/Cloud one, the computer becomes just another office supply, and the IT department becomes marginalized. When in the cloud, other services seem more viable. Virtual storage and backup deals mean goodbye to lots of servers, and that backup guy too. No longer dependent on the IT department, HR, Customer Service - hey, every department can find alternatives in the cloud. And those alternatives in the cloud will be supplied by the same people who make the software installed on their computers now. By putting Office online, Microsoft separates their biggest revenue stream from their troubled operating system. Microsoft will want to make up for the loss of revenue. They will “incentivise” their cloud products, making services cheaper than anything an IT department can provide. The stakes are even higher because Microsoft has to move into cloud, which is Google’s home turf. Google enters the market meeting Microsoft head on, feature-to-feature and with a better price - for now. Both competitors want a piece of the IT department, especially in these changing times. So count on predatory pricing to make the move even cheaper. These giants are in a fight for their corporate lives, so don’t think for one moment they’ll do anything that’s not in their financial interest. Every perk will have its price. The original story: http://translate.google.com/translate?sl=ja&tl=en&js=n&prev=_t&hl=en&ie=UTF-8&layout=2&eotf=1&u=http%3A%2F%2Fpc.watch.impress.co.jp%2Fdocs%2Fcolumn%2Fubiq%2F20121122_574440.html
That when Mankind actually launches ships to other star systems, the computers on board will be running a descendent of the x86 ISA, even if it's running 1024-bit words on superconducting molecular circuitry.
And also that the geeks who know anything about them will be bitching about the <expletive> ancient POS instruction set.
Lacking <sarcasm> tags,
i remember one of my old friends use to have a 386 or 486 computer that ran Windows 3.1. It even had a 5 and 1/2 inch floppy drive. it would be interesting if Intel put a 486 into a smartphone. but then the phone's battery will be drained fast, I bet. ok, i'll stop rambling. lol
x86 architecture failing behind mobile phone? hahaha
like Unix.
Let me know when I can full system compiles on my video card or run real world business applications on my video card.
Are Blender and other 3D modeling applications not "real world business applications"?
... but it won't happen until you can plug a keyboard, mouse, and monitor into a tablet and run MS Office and/or play hot 3D games in a familiar environment.
But, there will never, ever be an ARM port for MS anything. So, there you have it.
Nobody is going to try and build an ARM-based supercomputer (or an AMD based one for that matter). Intel is pretty much the only game in town, despite a spirited challenge from Nvidia (GPUs require specialised code and aren't suitable for many tasks.) We target Intel chips with Intel compilers, and have no plans to do anything else.
This discussion ignores all this and assumes that the entire future is based on small, low power devices.
x86 runs practically every desktop, laptop, server and every dedicated, non- embedded computing device (with the exception of phones and tablets, but those are so crippled by comparison to a real computer that I struggle to identify them as non-embedded). The general software support is inescapably vast and the user base even more so.
ARM software still sucks and is comparatively crippled. My phone is roughly as powerful as the laptop I bought 5 years ago, but is nowhere near half as capable because it's hogtied by piddly, ad-laden, consumption-centric software (y'know all that great freeware that you can get for x86 systems that people wrote because they enjoyed it and not just to squeeze ten fucking dollars a month from ads and other bullshit? Yeah, that doesn't really exist in the mobile software space, at least not yet)
Yes, at this point, maybe it is a kludge of workarounds on top of compatibility layers on top of conversion systems, but it works and works well nonetheless.
People have this false notion that Intel only makes x86 microprocessors, when in reality they make many other things. What truly makes them elite is their semiconductor fabrication plant ("fabs"). They are vastly superior to the competition, but they are most profitable/efficient only when the lines are full. When x86 chips became in greatest demand THAT became their focus. Now that they can no longer run their fabs at capacity with their own products they are opening them up to other companies (e.g. Altera, Netronome). Their fabs' process technology is currently 12-24 months ahead of the competition, and everyone would love to build their (ARM) chips their if they could (affordably). Back to chips, there is yet to be a high-performance 64-bit ARM implementation on the market for consumption. Until that day x86 will remain in Windows/Mac PCs and Linux workstations/servers without a question.
I see absolutely no evidence that one instruction set is going to give better performance or energy efficiency over another. CPU speeds have to do with the design considerations (circuit design as well as substrate technologies) taken when designing the chip.
During the 1990's there was a "war" between the RISC (Reduced Instructions) and CISC (x86 Complex Instructions) instruction sets. (I can't be bothered to reference this but you can try googling it.) RISC is considered to be simpler, equal in length, though more numerous, and faster to process by the instruction loader and decoders at the time. CISC, was unequal in length, and required more decoder time. And in-fact for some time there was competition between RISC processor manufacturers: Alpha made by Digital, SPARC made by Sun, and PowerPC made by IBM (to name a few) where king of the hill. Today, Digital is no more, Sun was snapped up by Oracle for Java and SPARC binned, and IBM, well, is a Jauggernaught that can't be stopped even by solar radiation. IBM still sells RISC based PowerPC chips for their bigger iron boxes and specially applications.
At one point it was revealed that CISC was actually a superior instruction set to RISC because it was more compact and made better use on on-chip "close to the iron" L1 cache space. As fetchers, decoders and branch prediction became better and chips shrank the performance differences boil down to process technology and chip design. At a market level, product support, reliability and serviceability are all key...
The main benefit of x86 is that its widely supported. The problem with creating a new Instruction Set is that compilers and development platforms need to be updated to target the new platform. This considerable time requires companies to invest. Optimizers etc etc all represent additional investments for newer platforms. And yet are available for x86, even for free. Even it an i7 is not reliable enough for the redundant services banks require, it's cheap and available enough to replace when it goes bad or becomes obsolete.
In reality, CPU's are general purpose chips. The basic operations like fetch, add, subtract, push pop etc have been around from the beginning of the computing age when CPU's where as big as rooms and constructed from vacuum tubes. And that is probably not going to change. The tendency is to add graphic related vectorial processing instructions to speed up that department.
Perhaps, the only gain in a new instruction set is where the new instructions are more purpose specific, or where that purpose requires fewer or different operations that make the x86 instructions bloated. That even then, the "waste" comes from engineering efforts go into implementations are are unused, and perhaps, just perhaps, a tiny bit of substrate area that happens to be associated with that unused area. But with the onset of SOC (System on Chip) the tendency is to want more features rather than fewer.
SOC (System on Chip) also takes other factors that are normally put on a northbridge (like until recently a memory controller) and puts them together. But this type of integrations is normally more limited by protocols that are a level higher than the instruction set. DMA, PCIe, USB etc etc. It's this integration that increases efficiency, not the instruction set or the nature of the protocols themselves.
Intel has some very very brightest engineers working on their process technology as well as their chip design. The only reason ARM has a head-start in mobile space is because Intel was caught with it's pants around it's ankles in a war with AMD in a performance race, which they have won. They had been focused on the 130W power envelope that Microsoft pushed for the PC era when Bill was at the helm.
Rest assured that now with the new goalpost in sight (mobile space, iOS and Android) Intel will utilize its massive resources to compete very well with technology that they are very very confortable with, x86 instructions, the very best manufacturing process as well as chip engineers. They do not need to change what does not need to be fixed. They need only get Samsung, LG or even Motorola onboard with one successful product to convince everyone x86 suites mobile space.
Isn't the PlayStation 4 and SteamBox going to be x86? Maybe we should ask them "why the x86?"
Personally my favorite topic is the MS OS/2 2.0 fiasco, about why did it took *10 years* after Intel introduced the 386 before 32-bit programming became popular:
http://yuhongbao.blogspot.ca/2012/12/about-ms-os2-20-fiasco-px00307-and-dr.html
But what happens when cheaper, more power efficient ARM chips are powerful enough for desktops and laptops?
Operating systems for ARM computers will probably still not be able to display more than one application. Right now, the most common window management policy on devices with an ARM CPU is all maximized all the time, even on a tablet that's big enough to run two or three smartphone applications side by side, and even when the device is docked to a Bluetooth keyboard and a (comparatively huge) HDMI monitor. This plague is even starting to affect x86 PCs running Window 8 (lack of plural on purpose).
There is still a very viable pc desktop market, especially among custom builders and pc gamers. People still buy desktops from stores, they still buy custom built ones from speciality companies and they still buy parts to build their own. So if there is a market why would they stop making money?
Then you have the fact their only competition in that area is AMD and AMD cant make a decent product to save their asses. Intel has vastly superior performing products for a lot less. Sure intel would drop x86 processors but why quit when youre the king?
I dont know of any company that would say "well guys we make product X and we make money at it, our only competition are complete imbeciles but I feel we have been doing it too long. So I propose we get out of that market and stop making profits off of. Why you ask? Eh, just because. No real reason".
Then you have the plain and simple fact that there is nothing wrong with x86 at all. Does it have a few short comings? Yes but nothing that impacts its usefulness at all or makes it undesireable. People argue stupid stuff like oh arm chips are getting faster and so on, but their performance doesnt improve that much while their power consumption and heat rises while intels x86 get cooler and use less power as they are improved.
It's mostly in netbooks now
What "netbooks now"? I thought 10" laptops were discontinued for good months ago.
I wouldn't be so sure about that. Look at Windows 8. They now have the legacy desktop semi-isolated with an entirely different interface. They also have a VM solution standard on all desktops. How much of a change is it really to move to a "classic" environment where the Win 7 desktop doesn't load by default? Then say a decade later (they'll probably have to go slower than Apple) where it isn't included by default? .NET4 is mostly going to treat .NET3 the way Apple upgraded people to Cocoa. With their newer interfaces playing the role of Cocoa. And just as Apple slowly moved Carbon lower level....
Intel sold it to Freescale (they actually sold it to Marvel).
The commic bookk companyy that sharres a corpporate parrent with Pixxar? Methinks you forgot a letter.
Sorry, pet peeve.
Practically all free GNU/Linux applications can be recompiled. What needs are still unmet by free software?
It would also alienate many strong key advice givers in the machine purchasing decision making process if their home Intel game collection was suddenly nixed by a capricious ogre like hardware maker.
Like the move from Xbox to Xbox 360?
Then perhaps closed source is the part of the problem that needs fixing.
Because of this monolithic inflexible market dominating operating system and related application landscape (MS-Windows) that runs practically only on x86.
Windows, Word, Excel, and Games.
Xbox 360, PS3, 3DS, PlayStation Vita, Wii U, iPad, Kindle Fire. All have games. All lack Intel.
Q: "Why can't Intel kill x86"? A: Because nobody wants to use anything else.
They should have put an x86 inside the ia64, so software for either architecture can run. That could have been a separate core, or even just the bits and pieces to make an accelerated emulator. Then people could have run old and new software side by side. And of course, as mentioned, they already needed a damn good ia64 compiler ... and make it free and run on all the OSes.
Eventually (not even now) they can phase out the x86 to purely software emulation.
now we need to go OSS in diesel cars
The reason for Intel and X86 is the IBM PC and it's back room marriage to Microsoft basically boiled down to a simple choice of who would supply the microprocessor with the most desirable terms. Intel won, with their X86, not because it was better or faster, but because they agreed to IBM's terms. The rest of the history is about the symbiotic (some would argue incestuous) relationship between Microsoft, Intel, and PC manufacturers has little to do with what would have been better technically.
The Motorola 68000 series processors where much more capable, flexible and MUCH easier to program (at least at the assembly level). Had Motorola won, we would have enjoyed an instruction set that did not change for the life of the 68000 processor. But as it was, with the x86 progression, 286, 386, Pentium and following, each introduced multiple instruction set alterations in an effort to keep up with the PC's needed expansion and performance requirements. None of this would have been necessary with the 68000 through the same progression.
Another advantage of the 68000 would have been 64 bit floating point math would have been standard and using 64 bits would have been seamless to the programs that used it. Operating systems would have been easily ported to 64 bit hardware, because it would have been a device driver exercise, and ONLY for devices that required 64 bit so the migration would have been piecemeal instead of the hard cut to 64 bit we have now.
We are stuck with x86, not because it was or is the best, but because it was the chosen one. X86 is the one supported by IBM back when this all got started, and now it's the primary platform for Windows and Microsoft. These past decisions where for business reasons and not technical ones. There is a lesson in all this.. :)
So.. As long as the relationship between Microsoft, Intel and Hardware builders remains in tact, and the PC remains the premier computing platform, we will be stuck with the x86. The question is how long will this last? Apple tried and fell back to x86 hardware, but I'm not sure Intel is going to control the mobile computing market which seems to be able to make inroads into the desktop market.
"File to fit, pound to insert, paint to match" - Aircraft Maintenance 101
I don't program ARM assembly language, but it appears to me that Sophie and Roger made a few calls on the instruction set that proved awkward as the architecture evolved:
These design decisions made the best desktop CPU for 10 years, but they came at a price.
How easy is it to port existing applications that were not originally conceived as web applications to Chrome OS? Or how easy is it for an end user to enable developer mode and get an Xfce-based distribution going?
ISA is pretty much completely abstracted that this point.
It would take almost zero actual silicon modification to take a Cortex A-15, switch out the microcode and have a x86 processor. You might have to add some new features to the instruction decoder logic but that's it. You wouldn't sell such a chip until after you optimize the A-15 design to efficiently execute x86 instead of ARM of course :)
Same thing on the Intel side. They could take an Ivybridge chip, write some new microcode and poof, its an ARM chip.
In essence, ISA really doesn't have much impact on power vs. performance anymore, its essentially the "API" that sits on top of the actual processor. What DOES matter is the chip design itself, and the process technology. ARM has spent 20 years optimizing their design for power, whereas Intel has spent 20 years optimizing their design for speed.
Now that the two markets are starting to overlap, Intel's designers are optimizing for power, and ARM's designers are optimizing for speed. The big question is which company is going succeed in the other company's home market first?
Seriously, lets stop pretending that the current state of affairs has anything to do with ISA. /. is acting like a bunch of old techies that remember the CISC vs. RISC war and want a rematch.
Can you say goose?
There are loads of proprietary, binary software around. Some people even run OS/2 because they won’t port their software to something newer. FreeDOS is around and used in production. Alpha emulated x86 quite competently, and current x86 processors are actually Risc chips with an x86 translation unit.
Until most software is based on open standards and free components that can be trivially recompiled, all platforms will live much longer than people would like them to.
Leandro Guimarães Faria Corcete DUTRA
DA, DBA, SysAdmin, Data Modeller
GNU Project, Debian GNU/Lin
...sorry for the slip, for anyone paying particular attention.
What killed the RISC alternatives to x86 was the Pentium Pro. Before the Pentium Pro, the industry consensus was that the way to faster machines was RISC. Then Intel developed a superscalar x86 machine and beat out RISC hardware.
It was an incredible technical achievement to make an instruction set designed for zero parallelism go superscalar. All previous superscalar machines, from the IBM 7030 and CDC 6600 of the 1960s, had imposed restrictions on what programs could do to accommodate the problems of concurrency.
The Pentium Pro didn't do that. All the awful cases were handled. Exceptions were exact. Storing into code just ahead of execution was allowed. It took Intel 3,000 engineers to make that work. Nobody had ever put that level of effort into a CPU design before. The design team for a MIPS processor was about 15 people.
The Pentium Pro was designed for 32-bit code, but still ran 16-bit code. Intel thought that by the time the thing shipped in 1995, the desktop world would be 32-bit. After all, it had been 10 years since the 386 introuced 32-bit mode. The desktop world still wasn't ready. Many users ran Windows 3.1/DOS on the Pentium Pro and complained of slow performance. It ran Windows NT quite well, but NT hadn't achieved much market share yet, much to Microsoft's annoyance. So the Pentium II had more transistors devoted to 16-bit support, fixing that problem. The Pentium II and III use modified Pentium Pro architecture. The Pentium 4 (late 2000) was the next new design.
That was the beginning of the end for RISC. RISC could get a simple CPU to one instruction per clock. Superscalar machines could beat one instruction per clock. Superscalar RISC machines had all the complexity of superscalar CISC machines, combined with a lower code density and thus higher demands on memory bandwidth.
As it turned out, x86 wasn't a bad instruction set to make go fast. RISC thinking was that having lots of registers would help. It doesn't. On a superscalar machine, commits to memory are deferred, and most stack accesses are really coming from registers within the execution units. So there's no win in having lots of user-visible registers. Also, if you have a huge number of registers like a SPARC does, time is wasted saving and restoring them. On the stack, you just move the stack pointer.
Also, RISC code is about twice as large as x86 code. Making all the instructions the same length bloats all the small ones.
The Itanium was an attempt to introduce a proprietary architecture that couldn't be cloned. The Itanium has lots of original, patented technology. It was very different from other CPUs. However, it wasn't better. Just different. Compiling fast code for it was really hard. It was a "build it and they will come" architecture, like the Cell. Except they didn't come.
Intel has a compiler business but the standard is Microsoft's Visual Studio and Microsoft plans Visual Studio around its own releases, not Intel's.
"The standard"? Eh, for Windows/Desktops, maybe. But gcc and Linux have far surpassed VS and Windows on x86 server hardware (not to mention virtualized x86 hardware!) And actually, the last Win32 port we did used MinGW because it was so much simpler to port in the existing build environment...
And besides, the massive adoption of ARM in mobile devices, ARM and MIPS in consumer electronics, and PPC in game consoles (up to now, at least) shows alternative architectures can flourish and compilers will be developed *if* there is a strong technical or business reason (one other than "Intel wants higher margins").
Is what you advocate using going to save enough money to pay for the replacement cost? If not, then why do it?
> I'd like to set up the system so that her account is not an Administrator
Well, yes, duh. That's *mandatory*. Anybody who doesn't have deep knowledge about computers should NOT logging in as Administrator on a daily basis. The account she uses on a daily basis should definitely be Limited. If she wants to log in as Administrator every day, then you tell her she's on her own then. If she wants support, she needs to follow basic minimal safety practices, and NOT logging in as Administrator all the time is top of the list -- more even than keeping up to date in Windows Updates, and WAY more important than having anti-virus software.
Some users can handle being given the Administrator password also (so they can do things like Windows Updates on their own), and others cannot. You know your mom better than we do, so I won't try to advise you on this issue.
> and that I can easily (and securely) remotely connect
The best option for Windows, that I know about, is probably VNC. There's also rdesktop (which I think is maybe more efficient with bandwidth), but I'm significantly less confident about its security when routing the connection over the public internet. (It does use encryption, but I'm not entirely confident that it implements it in a totally secure way.)
Symantec also has a product in this space, but I don't know much about it. I'm sure there are others.
Cut that out, or I will ship you to Norilsk in a box.
There are some free open source business apps out there but nothing compared to paid ones.
Which major paid business apps are you talking about, and what do they have over the free counterparts?
Also the bigger firms have a shitload of custom in-house written apps that only run on x86.
Are they written in assembly language? If so, why? If not, why can't they recompile in-house what they wrote in-house?
Keep your legacy x86 PC to play your legacy x86 games, just as you keep your legacy NES console to play legacy NES games. Do your work on a new legacy-free computer.
Intel can't, r won't, kill x86 for the same reason that Microsoft won't completely change and why they made sure to keep Windows applications compatible across releases. It allows them to maintain their pseudo-monopoly on desktop. If there was a competition for a new PC architecture, IBM and a bunch of Chinese manufacturers would be right there competing on an equal footing. No way Intel allows that to happen. Intel would rather have an AMD than a proper competitor. AMD shields them from antitrust and so the x86 situation is brilliant for them.
AMD is the underdog, but everytime AMD gets to design a new part, they have a chance to better Intel. Famously, AMD won the race to 1GHz, and bested Intel's putrid 'Netburst' Pentium 4 design with the first 64-bit x86 architecture, and the first proper dual-cores. AMD only fell behind when corrupt management killed future R+D projects at AMD, because by doing so the corporate structure allowed for massively increased bonuses and stock-options. Only now is AMD beginning to recover from that internal sabotage.
Intel's mobile work is total rubbish. It consists of two strands. The utterly hopeless Atom architecture that enjoyed obscene success in the brief 'glory' that was the 'netbook' market, and so-called ULV parts that are either clocked very low indeed, or 'binned' as cheery picked chips from very large runs of CPUs usually designed to require a much higher voltage. Only the ULV parts have acceptable performance, and they are VERY expensive for Intel to produce.
Meanwhile, AMD has moved from its 'Brazos' line (very promising mobile parts that were just too under-specced in GPU and CPU) to parts based on its successor, using cores named as 'Jaguar'. The 'Jaguar' design has the same performance as AMD's original 64-bit desktop chips (per clock), but with powerful integrated graphics, and very low power consumption. In other words, everything idiots expected Intel's Atom to be.
The Jaguar is the ONLY x86 core that stands a chance against high-end ARM designs. Intel's 'equivalent' is the dreadful 'Haswell' architecture. Haswell achieves ultra low power operating modes by gating out hardware blocks, and emulating their functions in software. In other words, Haswell low power = no performance. The alternate mode for Haswell is bursts of activity where the power consumption goes through the roof, with the hope that the benchmark ends before the mobile device melts. Continuous high performance computing from Haswell is not going to happen in a tablet format.
Worse again for Intel is the fact that AMD is now implementing HSA, where the GPU and CPU clusters use the same hardware memory mapping. Intel is years from doing the same. And worse again is that AMD offers zero-latency video-encoding for remote-desktop streaming (including games), which is how the WiiU works. Intel's video encoding, on the other hand, is a nasty hack purely used for making bad copies of existing video for viewing on mobile devices that curiously lack the room for the original video file (all new mobile devices can easily play the same video files as desktop PCs).
Intel is a terrible company that has NEVER had genuine sustainable success as a result of its own efforts. Before IBM gave Intel the home PC market, and every bit of success enjoyed afterwards, Intel was a basket-case. Sure it invented the fist CPU, but the real talent would ALWAYS leave Intel and form their own, more successful, start-up CPU companies. The great 8-bit CPUs from the time were the Z80 and 6502. It matters not that the first was inspired by Intel's 8080, and the later by Motorola's 6800. When it was time to step up to 16-bit architectures, Intel's 8086 was easily the WORST of the 6+ competing 16-bit architectures, whereas embarrassingly for Intel, arch rival Motorola regained its glory with the wonderful 68000 design. We should all feel sickened that the 8086 was chosen over the 68000 by IBM simply because Intel's market failures meant IBM had Intel over a barrel at the time. IBM didn't want their PC to be good, performance-wise, or easy to program, which made Intel an even better choice.
Years later, when Intel was swimming in the mountains of cash gifted by IBM's choice, Intel tried buying tech companies. It ran every one of them into the ground, and its efforts proved to be a total waste of time. More successful for Intel was stealing the patented designs of competitors' CPU architectures, and then simply paying the fines levied by court losses out of their massive profits. Intel didn't go 'RISC' with the Pentium Pro (and then the Pentium 2
there is no way in hell any serious money making company is going to entrust their data to the cloud. the security of the cloud varies from questionable to outright bad, the availability of even of the most robust providers isn't 100%, and there is always the question of what happens in the event the cloud provider merges or gets aquired or goes bankrupt or gets the SEC or other fed agency coming down on them like a ton of bricks.
Instead, for non-developers, if you want to keep your IT job, get one or more of these skills or stay current with them, think of it as "private cloud" skills if that suites you:
1. virtualization software, both server and desktop
2. storage / SAN
3. network engineering
4. backup / archival / HSM
5. DBMS and related softwares
6. end user device management
7. telephony
it wasn't really a jump off of x86 but it was a big move away from 16bit x86 but they put their eggs in the Microsoft basket and got their egg cracked. Windows 95 shipped and it sucked on the 32bit CPU while other 32bit OS's( UNIX, OS/2, etc ) saw upto a 200% boost in performance at the same clock rate over the 16bit Pentium.
It seemed like Intel started spinning things like the multimedia SSE into the CPU to give it some way for improved performance on that old DOS/Windows 95 platform the public got suckered into. Tied to Microsoft they had to keep with x86 because Windows wasn't portable and Windows NT was a bloated pig of an OS considering the hardware at the time.
So because they slept with Microsoft, I think it's the reason they could not give x86 the heave ho. And they surely didn't do a great job with StrongARM which they got from the carcass of DEC. I recall a 400MHz model running slower than a 200MHz earlier version because of some failure in cache. Lots of handheld vendors put those 400MHz StrongARM chips in their handhelds and the users cried at paying such a high price for slower hardware. It wasn't too long after they ditched StrongARM.
Any recent x86 chip from Intel/AMD will have an IOMMU which controls a lot of the I/O coming from peripherals. For instance, if a network card tries to DMA to main memory, the IOMMU will determine whether or not that write goes through based on which address range it's trying to access.
On older hardware the various peripherals could directly trample all of main memory, which could cause nasty bugs.
That article says that "Broadwell" will be BGA only, not Haswell. Haswell will continue to be offered as LGA. Also, the successor to "Broadwell" will apparently be offered as LGA as well, so I doubt this is the end of the line...
I believe this.
Ultimately it's the fab technology that gives Intel the hyper-carnivore advantage - being three process jumps ahead of other foundry operations.
Intel sells a server chip for every 20 phones sold - so all Intel needs to do is take 20% of that ARM market - and use it's superior fabrication ability to continue to own server, desktop and have a slice of mobile.
As a BIG COMPANY it suffers from the delusion that it can command markets.
Ultimately the battle to be fought over the next 10 years will be to continue to own servers, desktops, laptops - but take a significant segment of mobile.
The Intel CEO - will I'd guess - get the company back into the business of fabricating ARM chips - using Intel's far superior fabs to drive power performance numbers that the Qualcomm's of the world just can't achieve.
A .22 nanometer, or .14 nanometer fab costs about 8 BILLION dollars to build - and Intel is two process hops ahead of the competition.
It's Intel's game to loose
If Microsoft, apple and the linux foundation would agree to stop suppoting X86 on future operating systems and only support ARM on future versions of OSX, Window and Linux, this woully kill X86 quickly
Nowadays, the instruction set is just a front end and changing it won't do much in term of performance.
I believe that the reason why x86 is so prevalent on PCs and ARM in mobile devices is mostly historical : Intel focuses on expensive, high performance CPUs and ARM on cheap, low power SoCs.
Even Intel talks about Atom's abysmal performance. The good news is the next gen Atoms will be bringing real performance to low power. They're going to be completely difference archs.
Yes, but will they support more than 4 GB of RAM? No doubt it was Intel's marketing department that foisted that limitation on all previous versions. Can't "cannibalize" the lower end of your Core market, eh?
Well, that particular decision made me choose the AMD Fusion platform instead. Too bad, because I really would have preferred Atom-based systems.
My next purchase will be awarded to the 18 W TDP CPU system that supports 16 GB RAM and AES-NI crypto acceleration—and this opportunity is Intel's to lose (not that they care about it).
It would have been nice if TFA tried to address the issue of the point of a particular instruction set over another. What do you gain by jumping ship and trying something new you can't get by adding on to what you already have?
The conventional wisdom says ARM is unbeatable at low (mostly idle) power but nobody explains why x86 can't have the same characteristics. If the marketeers at intel are to be believed there is no reason.
I used to think the future of processors were those specialized java bytecode executing CPUs but nobody is touching this approach these days.
People talk about CISC vs RISC but anymore it seems instruction set is nothing more than an interface to an underlying structure that seems to have little problem evolving within the constraints of the interface.
Now my working assumption is that what really matters in the age of parallisim going forward is the memory / concurrency model of the system...yet I'm not so sure of even this what prevents you from introducing new instructions with fewer guarantees? As the number of transistors reach twoard absurdity does the total amount of effort needed to deal with the past increase or decrease?
It's funny the lack of historic context with all the comments here. Intel has tried numerous times to promote new architectures. At the same time that the first x86 chips were coming out Intel had a processor series supporting abstract software architectures, where did it go? The dustbin of affordability and compatibility. x86 could run CPM (sort of). Then the i860 and i960 series (controller and RISC processors) came and went. I worked a lot with the i960. I liked the RP series a lot, but after the shine of RISC wore off x86 could run DOS/Windows, i960 not-so-much. Then Intel bought the DEC StrongArm holdings (see - even Intel did ARMs) - which became Xscale (now owned by Marvell). But Intel's ARMs were relegated to embedded controllers only. Then Itanium came, is still here, and almost no one is using. Then a few years ago Intel decided that x86 instruction set would be the ONLY thing they would make (except Itanium for contractural reasons). It always came back to the market for Intel was/is windows desktop.
The summary makes unjustified assumptions, and it's laughable to suggest that Intel's not a leader any more.
Well it's not just fileserving. I have a software RAID5, so everytime I write a file, it has to calculate parity information in addition to writing to 5 SATA drives simultaneously, and the bottleneck is still the Gigabit network.
I don't know if you are aware of the risks that have been mounting for RAID-5, but I am constructing an 18 TB total capacity home NAS and I have chosen software-based, double-parity RAID-Z2 for that reason.
If you are interested in standard RAID levels, then consideration for a double-parity RAID-6 may be in order.
for those /.ers who appreciate the number.
Both very good CPUs. In fact, had Microsoft seized the initiative then, they could have had a 64 bit version of Windows long before AMD came out w/ x64. Also, had Silicon Graphics been less dogmatic about Unix and more open to making workstations like the Magnum that ran Iris visualization software on NT, Microsoft would today have had Windows 8 running fluently on MIPS based tablets or phones, w/ a rich suite of applications. Fact remains that Windows NT came out before Windows 95 even did, so Microsoft could have had Windows applications easily cross compiled for MIPS and Alphas.
It's more that there were hardly any third party vendors. All the hardware as well as drivers/software came from DEC, or from other Alphastation vendors such as Carrera, Microway, Aspen, et al.
Problem is that DEC always went for overambitious specs on the MHz, resulting in lower yields, and hence, the need to overprice them. Had DEC offered the Alphas over a variety of speed & price points, they might have done better. For instance, on the 21064, when they achieved 275MHz, they should have offered workstations or even laptops over a range of speeds, ranging from 100MHz to 275MHz. They'd have been a lot more successful that way - both in terms of the Alpha's acceptance, as well as profitability, since they could have lowballed the low end, sold the mid range for a minimal mark-up, and marked up the wazoo on the high end.
Heh, one of my favorites!
"It has a killer refresh rate.
P6 chip. Triple the speed of the Pentium.
Yeah. It’s not just the chip, it has a PCI bus. But you knew that.
Indeed. RISC architecture is gonna change everything.
Yeah. RISC is good."
I mean its not all about RISC, its that PCI bus as well, but you know that! :)
There were 2 ways of improving the performance of a CPU. One was a superscalar design, where there are more registers, ALUs and pipelines to execute more instructions in parallel. The other was a superpipelined design, where the pipeline was broken up into several simple stages so that it was easier for each stage to complete faster, and thereby, increase the clock frequency.
At the time in question, most CPU architectures went the superscalar route: MIPS, and later Alpha, were the only ones to start w/ the superpipelined approach. But as they went into subsequent generations, superscalar CPUs tried to become more superpipelined - remember Intel's Hyperpipelining, anyone?, while superpipelined CPUs tried to become more superscalar. So those were the 2 routes to become superscalar, superpipelined. MIPS, as one might recall, first was superpipelined w/ the R4000 and R4400, before becoming superscalar w/ the MIPS IV (R8000) and MIPS V (R10000). In case of DEC, the 21264 was when it became superscalar & superpipelined.
The advantage that DEC had is that since they were trying to converge their VAX and MIPS lines into one architecture, they had the liberty of starting from scratch. The few attempts that Intel made to start from scratch didn't come anywhere near the RISC leaders of the time. i960 ended up as being good for peripherals, while i860 was good for some higher end things, but never came close to touching the best that IBM, DEC, HP or even Sun had to offer. As far as the Itanium went, it turns out that RISC was the sweet spot b/w all the dynamic analysis being done in hardware vs being done in software. Even if the Itanium had achieved a perfect compiler, the savings in silicon area would hardly have given them the speed boost needed to outstrip the Alpha. In fact, given that Intel had all the Alpha related IP, as well as presumably the IP from the PA-RISC, they could have done a RISC CPU instead of an EPIC, and been much better off. Of course, they'd also need either native x86 compatibility (which AMD achieved w/ its 64 bit x86) or massive commitment to port to the new CPU. As it turned out, the Itanium falling far short of expectations not only killed projects like Monterrey or Solaris or Windows Server, it even killed Linux versions for the CPU. As it is, today, the only OS outside HP/UX for the CPU that killed the PA-RISC & Alpha and crippled MIPS is Debian Linux and FreeBSD (I'm not sure if NetBSD is yet supported on this platform).
At this point, Intel would do well to abandon Itanium altogether, and instead focus on a RISC CPU like MIPS, PA-RISC or Alpha, rebrand it, and have it scale from tablets (think of the Transmeta Crusoe) to servers. And don't even think about x86 compatibility here.
No, when Compaq ended the Alpha, they sold all the IP and rights to Intel (which already had some as a result of their settlement w/ DEC over a lawsuit). The only thing that HP owns is legacy Alpha business, or DEC customers who bought into the Alpha/OVMS.
Wasn't that true about others as well - SPARC team and Solaris working together, POWER and AIX teams working together, PA-RISC and HP/UX teams working together? Even for MIPS, since Silicon Graphics owned it at the time they were big, I'd imagine their teams would also have worked together. Also, for NT, DEC had I believe a team that ported it to the Alpha, while Microsoft developed it (ironically) on the MIPS DECstations as well as i860s.
They wanted brand recognition and loyalty. They got it. I know a lot of managers that wouldn't even consider anything else - the Alpha that Microsoft supported, the Itanium that was a far superior chip, even the MC88000 of the 1980s was way better. So their followers were so loyal they won't even use another much better Intel chip.
They could easily work out a deal with Microsoft (and everyone else) to simply and quietly move away from it. Maybe they can call the next chip - IGS (Intel Great Stuff) chip.
People who often talk about post-PC mobile world are forgetting that mobile computing is useless without huge serve farms. Those servers are here to stay with us. You also forget that in the work environment, PC is still the primary way of computing. When most of your work time is spent on interaction with a PC, it better have a nice screen, keyboard, and mouse. I just don't see the tablets replacing the workstations of office workers.
So what is the difference b/w Wine and what Sun used to have on SunOS/Solaris - WABI? WABI used to do what you are describing - convert win16 or win32 system calls into unix system calls, and have the application run, while looking like a native SunOS/Solaris app. That would also take care of not dealing w/ either x86 nor SPARC binaries directly