Slashdot Mirror
AMD's 64-Bit Chip
EyesWideOpen writes "AMD is set to release a 64-bit chip early next year which will be completely backwards compatible with the Athlon line. The current 64-bit offering from Intel, Itanium, is an entirely new chip that has no backwards compatibility with its x86 line of chips (from the 8080 chip to the Pentium IV) and is designed only for high end servers. AMD's solution to this problem is the Opteron chip (product info) which will be in servers, desktops and laptops. Here is a wired article."
I've seen duplicate stories, but this beats them all! :)
They that quote Benjamin Franklin on liberty and safety deserve neither.
Slow day, huh?
Pedro Côrte-Real.
Hmm, I don't know about your experiences but I've found Intel's to be more reliable.
Ok so the AMD's have more performance, but they run hotter and are more liable to blow up (yeah ok, no evidence to back this up), The Intel chips jut seem to be built to more conservative tolerances.
Just my 2 cents I guess...
Running programs in a hybrid 32/16 bit environment puts a serious strain on the Windows OS: It crashes. Pure systems do not crash as often. I really wonder if the problem will be magnified in a 32/64 bit environment?
We're Doomed
Ford Motor Co. is set to release today a new car, the Model "A", based on the award winning and famously popular Model "T". The new Model "A" is backwards compatible with all previous 4 wheel gasoline powered Model "T" cars produced by Ford and its competitors, and can run on the same roads as them.
Infuriate left and right
I've been waiting for these bad mama jamas. I must have spent a week trying to install OS X on my Athlon XP. I couldn't believe how rude the tech support at Apple was, even though I tried to switch. I thought that Quartz would make the windows framing my porn look pretty, but I haven't had a chance to see. I hope this 64 bit CPUs change all that. I used to write 64 bit assembler programs all the time, but they never compiled and linked right. I blame the makers of GeOS, that had to have been the worst IDE I've ever seen for ASM.
What's the point of making something that is unsupported by a large chunk of today's software
Because you end up with a CPU that has layers of compatibility upon layers of compatiblity.
you'll have real mode, protected mode and now probably something like 64 bit mode.
imho it's better to get rid off all the old junk and start over once in a while.
Hmm, but you don't expect your DVD player to play your VHS videocassettes?
Backwards compatibility is fine where practical, but sometimes the past needs to be buried. Who would buy a computer now with a punch card reader? Or a 5.25" floppy drive?
Well, I suppose your reaction to this depends on your personal product loyalty (or possibly lack thereof). Basically, a CPU will inherently run slower if it is backwards compatible with a completely different architecture. What AMD needs is a chip that solely does 64-bit ops, like the Itanium. Now, I realize that this would require all programs to be recompiled/rewritten, but isn't that what PDA's require anyways? And I'm sure the conversion from 32-bit to 64-bit is a lot easier than 32-bit to Async (could someone familiar with that process verify/refute this?).
This is, in essence, what I'm saying: AMD should come out with 2 64-bit processors, only one of which natively supports 32-bit apps. Why? Otherwise Intel will absolutely rip AMD to shreds in the benchmarks test. Being a loyal AMD user, I don't want to see this.
IWARS.
People, in general, disappoint me. Politicians even more so.
If Windows runs on Itanium and not on AMD, that's the end of AMD.
-fb Everything not expressly forbidden is now mandatory.
people complain about slashdot rerunning stories. have you looked in the newpaper recently? There are full of repetitions and dupilicates, sometimes much more frequently. How many times have you seen violence in the middle east or northern ireland? How many times can you read about some crappy Martin Lawrence movie? All the slashdot editors are doing is trying to keep up with the newspapers.
Of course, Itanium is backwords compatible with x86 code. It just isn't particularly fast when doing so.
...considered part of the x86 family? The first processor in that lineup is the 8086. I think the 8086 might've been source-code-compatible (to some extent) with the 8080, but you can't take an 8080 binary and run it on any x86 processor (emulation doesn't count).
20 January 2017: the End of an Error.
I found my Dual AMD box to be as good, if not better from a never going down standpoint. Same goes for my gaming boxes. After building a bunch of systems, however, I do have one major beef with AMD...
... Things are a little better these days because the quality heat sinks - with paranoid mode on - are less likely to crush a CPU than when folks were trying to strap a socket 370 heat sink on an Athlon, but I still feel like it is a crap shoot every time I have to remove the CPU. I end up trying to stay about the $100 mark for CPU's as a result. (Yes, the MP's cost me much more, and I was very nervous when I mounted them)
For the love of god, put a coat of nickel or something on the CPU!
I chipped a couple when rev 1 of the Chrome Orb came out. Fool me once
+++ UGUCAUCGUAUUUCU
Itanium uses a very advanced VLIW-esque architecture which, coupled with a sufficiently smart compiler, allows for incredible performance.
Two problems:
1. making a really _good_ compiler for such an architecture is beyond the means of computer science now and in the near future
2. we gave up back-compatibility for a fairly slow (in its own right) ia32 emulation
DNA just wants to be free...
> imho it's better to get rid off all the old junk and start over once in a while.
Unless of course you've got an installed base somewhere in the billions, 20 years worth of compiler optimization, a factor of, what 100, more people that know the assembly language, etc. And it doesn't help if good compilers won't exist by the time your chip comes out. And if the internal interface teams have difficulty communicating, you're going to be late, hot, slow, and over-complicated.
Starting over is nice from a design perspective, especially because it feeds the urge for creativity that most engineers have. Unfortunately, that do-over is not always executed well, and it turns out to be a little underwhelming, just like Itanium.
Fight the urge to think that all new things are good. Please.
Outside of a dog, a book is a man's best friend. Inside a dog, its too dark to read.
Hmm, but you don't expect your DVD player to play your VHS videocassettes?
Actually, I do, and so do a lot of other people, apparently, which is why I can go to K-Mart or wherever and pick up a player that does DVD and VHS for under $200.
Most people don't buy these because they already have VCRs, and there's little problem using both. However, most people would be really pissed off if you told them that if they bought a DVD player they would no longer be able to watch their VHS tapes.
Itanium sales reflect that fact. Regardless of technical merit, lack of backwards compatability will kill Itanium.
Under capitalism man exploits man. Under communism it's the other way around.
Hmm, but you don't expect your DVD player to play your VHS videocassettes?
I would if they were in the same media type. Just like that new holo-storage announce; they said the drives will be compatiable with CDs and DVDs.
Backwards compatibility is fine where practical, but sometimes the past needs to be buried. Who would buy a computer now with a punch card reader? Or a 5.25" floppy drive?
No neither one. But I see nothing bad with a 64bit x86. If all I wanted was a 64bit system, why pay Intels high prices when I can go Sparc cheaper? If there is no compat, there is no reason to stay Intel.
I don't think anyone's ever said that the Itanium won't run x86 code--just that it does so very, very poorly.
Perl - $Just @when->$you ${thought} s/yn/tax/ &couldn\'t %get $worse;
The Mac shows how a great system with all the best features can not be worth a damn if they don't have the products to back it up. Think of the tens of thousands, the hundreds of thousands, of small 32 bit programs are out on download.com right now. You can't use one of them on the itaniam. No kazaa, no winamp, no aim, no small shareware/freeware apps, and no GAMES!!! If Intel thinks they are going to get a desktop switch over to 64 bit in the next two years b/c they have a faster chip then they must have accidently hired some old Apple employees.
:)
And I have no clue if the mac OS is more stable, faster, etc. But I'm just going from what mac people tell me
In other news... Pentium IV processors can now use DDR memory, you can now get dual-processor Athlons systems, and the Intel Pentium-3 processor has new instructions that will allow it to "revolutionize your internet experience" dubbed "SSE"
Computer Science is no more about computers than astronomy is about telescopes. --E. W. Dijkstra
steve snyder
Vote Quimby.
The only time I've seen a successful migration from one platform to another was when Apple managed to migrate from 68K up to PowerPC, and that was only really possible because they controlled both the hardware and the software.
they were also successful because in part so many developers spent a lot of time making fat binaries that would run on either 68K or PPC platforms. The developers made things backwards and forwards compatible at the same time in one package.
neye
A properly designed 64-bit CPU does not need to 'run slower' to run 32-bit apps. AMD came up with a simple solution to the 32-bit limitations of X86 code: they added a new 'mode' to the processor to run 64-bit binaries. when this mode bit is set (similar to the old Real and and Protected modes of X86 chips), the chip utilizies the full 64-bit-wide pathways for data and cacluations, when this bit is not set, only the lower (or is it upper? AMD isn't saying...) 32-bits of the pathways are used. The same exact logic units are used for all 32-bit and 64-bit calculations, only the bit-depth precision changes. Thus if it takes an ADD instruction 16 cycles to add two registers and store the results in a third register, it takes 16 cycles reguardless fo whcih mode the processor is in. Of course, AMD also added an extra 8 registers for use in 64-bit mode... very useful.
The itantium does not get the majority of it's speed from being 64-bit - this is a common mistake people make. It has a _very_ different design and instruction set - EPIC - which places the burden of parallel instruction determiniation on the compiler. Basicly, they used the oldest software refactoring trick in the book, but on the whole processor design: they examined the amount of time spent executing, and looked for the bigest runtime performance-hit that could be moved from a O(n) to a O(1) penalty by simply moving the calculation. In this case, modern processors spend a great deal of time trying to handle multiple instructions at once, which may or may not be parralellizable (is that a word?) - thus the processor has to figure out, on the fly (in a P4, for example), if it can execute the next four add instructions in parallel, or if they are interdependant and cannot... By placing the burden of parellelism determination and instruction scheduling on the compiler, intel made the compiler writer's job much harder, but at the benefit of increased performance.
Oh, and most PDA processors are much more traditional, and thus don't require complex compilers like the itanium, so actually porting a compiler (or an assembly-lang app) to a PDA from x86(32-bit) is easier than creating one for the EPIC architecture.
And yes, I know the above is an oversimplification, and Intel and AMD both did a lot more, in a lot more detail, on thier 64-bit chips.
Oh, and I think the next few iterations of itaniums _will_ beat the AMD 64-bit chip on bechmarks. But not by a landslide.... And with the differences in price (EPIC chips are Expensive... capital E) the AMD chips will win the hearts of many and be the performance-price ratio king. And who wants to pay 3 times as much for 20% more performance?
man is machine
>> very advanced VLIW-esque architecture
Ah, yes. EPIC. Explicitly Parallel Instruction Computing. AKA VLIW. EPIC is market-speak. Intel didn't want to admit that it was making a VLIW chip for two reasons:
1. There is only one company that has every sold a VLIW chip that actually worked, and that people bought: TI makes DSPs, where are VLIW. They make tons of money. They are the only ones that ever did it right.
2. There is only one company that has ever made a good VLIW compiler: TI, again.
Lets think briefly about how great EPIC is, using the two main selling points I remember from a presentation I saw on it a few years ago (sorry if my memory is bad, no coffee this morning, I'm not responsible).
1. Instructions are Explicitly Parallel. So, the compiler tells you that these two or four or however many instructions can be executed without worrying about data dependency. Terrific. Assuming that the compiler actually works, which is still an open question.
The only difference between this setup and what's in your Athlon or Pentium4 is that the looking-for-independence is done in hardware on your Athlon instead of by the compiler on your Itanium. This means that there is the *possibility* that EPIC does better at finding independence because the compiler *should* know more about the code when its in a higher level language. *Should*. Essentially, until the science of compilers takes a quantum leap or we start using programming languages that makes these things easier (correct me if I'm wrong, please), Itanium will be at most as fast as a superscalar processor that finds independent instructions on its own and does register renaming.
2. Predicates and conditional execution. While the whole notion of the predicate in EPIC is more complicated and complex than just conditional execution, its not entirely more useful IMHO, or at least that was my impression the last time I heard someone talk about it. Alpha has conditional execution. ARM has conditional execution. I can append checks to the condition codes in ARM assembly. I don't really understand why this is so nifty.
I've said it before, and I'll say it again. Resist the urge to think that whatever marketdroids tell you is new is actually good. Sometimes its not.
(If more knowledgeable people are lurking, please correct any errors I've made, but I think I've got this right.)
Outside of a dog, a book is a man's best friend. Inside a dog, its too dark to read.
my 800 MHz AMD is still doing the job. I don't really need a 64 bit chip until 00:00:00 UTC, January 1, 2038.
nahhh, gotta be a coincidence.
Religion is a gateway psychosis. -- Dave Foley
Can you say .NET? Seriously though, making a runtime will make this simpler...
"You can't make a race horse of a pig"
"No," said Samuel, "but you can make very fast pig"
"64-bit code is twice as big as 32-bit code" bloatware excuse
Unfounded. Though I find Itanium's instruction coding (16 bytes per 3 instructions) bloated, not all high-"bit" machines have to have bloated bytecodes. The ARMv4 architecture, used in processors such as the ARM7TDMI in the Game Boy Advance, has a standard 4-byte-per-instruction encoding, and an optional 2-byte-per-instruction encoding called "Thumb". Thumb code runs at about two-thirds of the speed of ARM code on machines with fast memory because some operations take more instructions on ARM than on Thumb, but Thumb code really shines when running on small or slow memory and can help drain less battery power on mobile machines. Apps will often have most of the app in Thumb but some of the time-critical inner loops in ARM.
Will I retire or break 10K?
It is just not native code, therefore it is slower. But it runs 32-bit versions of Windows and Linux JUST FINE.
Except the FUDsters are right this time, as software written for x86 doesn't run on Itanium. Rather, it crawls on Itanium. The difference is most noticeable in soft-real-time applications such as video games.
Intel could have done the x86 emulation much more efficiently; read my other comment. Efficient recompilation in silicon is the approach AMD has used since the K5 processor and perfected in the Athlon product line.
Will I retire or break 10K?
With an opteron running a 32 bit app is that app limited to a 4gb limit, or can it address above 4gb?
Depends on the operating system. Some kernels support allocation of memory through "far pointers" that refer to a "segment" of large memory, then a smaller offset within that segment. The Windows/286 operating system, versions 2.03 through 3.1, used far pointers as the common memory allocation type because the 286 limited offsets to 64 KB. Likewise, with the 4 GB offsets on the 386, 32-bit apps running on a suitable OS will be able to allocate multigigabytes of memory in 4 GB chunks. For instance, non-Celeron PIIs, PIIIs, P4s, and Xeon processors already support up to 64 GB of physical memory, given an appropriate motherboard. I'm not as sure about the Athlon, given that it still uses an older socket.
Will I retire or break 10K?
- "While the first 32-bit processor came out in 1995, the average PC used 1 MB of memory, so 4 GB was both unaffordable and generally not needed."
Without digging too deeply, it can be found that Motorola came out with the 68020, a true 32-bit processor, in June of 1984, 11 years prior to the debut of the 32-bit processor according to the nimrod author. I don't have solid dates but I know that within a year of this timeframe Suns and Apollo workstations were using this chip.How disgraceful.
blakespot
-- Heisenberg may have slept here.
iPod Hacks.com
Let's say you have an elaborately-customized server setup. Let's even imagine that some of your storage for both data and programs isn't sitting at a single PC, but is in network-attached storage. Now, you want to upgrade the hardware to 64-bit without having to recompile everything - or maybe just upgrade some of the servers while continue to share program code off the storage.
You get only one answer: AMD. You can take your complexly-configured servers and not have to redo them from scratch. And the hobbiest gains the same advantage - swap drives, compile yourself a 64-bit kernel, and forget about doing a virgin install of Debian 64.
___
"with their freedom lost all virtue lose" - Milton
Sure, everyone 15 years ago thought 4 GB of memory would be PLENTY. But how about in another 15 years? Will an exabyte of memory still be able to run the highest end applications including Microsoft Office 2017?
If you want a "fresh" architecture that isn't full of old junk, buy an Alpha. Or for that matter a MIPS, SPARC, or Power4. All of which are 64-bit and have either always been 64-bit, or at least had their original 32-bit designs planned around 64-bit expansions.
Personally, I think it's amazing how much old crap has been piled onto x86. It's really remarkable it runs at all, and it's even fast! I used to turn up my nose to the x86 given how they piled all the 32-bit extensions on the old 16-bit core. It's really a travesty. And the actual instruction set and register set looks like a damn train wreck compared to MIPS or PPC. But they are soooo cheap I eventually got over it, and just try to avoid thinking about any level lower than 'C' now so I don't go insane.
Bill (Gates) announced that Microsoft(tm) would support Opteron. Jerry (Sanders) gave nice pro-Microsoft(tm) testimony at the anti-trust trial. Funny how Microsoft(tm) seems to encourage competition in the x86 market. Oh well. I'm not complaining if it keeps AMD and the x86 market viable.
Whats probably best is to have a hybrid 32/64 chip for a little while
This is exactly why Intel built x86 compatibility into Itanium. From what I understand, this compatibility will be removed in future Itanium versions once the user base is weaned from the crappy x86 code.
"The defense of freedom requires the advance of freedom" - George W Bush
The vast majority of the old cruft in the X86 architecture that nobody uses any more has been demoted into microcode or other non-optimized crevices. Ever since the Pentium came out, good programmers and compilers have been using an almost RISC-like subset of the X86's myriad possible instructions, operands and addressing modes. IOW, all that old stuff really doesn't slow things down in the real world.
Anyway, recent CPUs have been transforming X86 instructions on-the-fly into bizarre internal parallelized architectures anyway. This hidden logic is an order of magnitude more complex than what is visible in the X86 instruction spec. The implementers are free to completely redo the hidden stuff with every new generation of X86 chip.
The Wired article has other errors as well. A 32-bit CPU isn't limited to 4GB; that confuses address space with physical memory. The definition of exabyte is wrong (1000 petabytes, not 1000 terabytes). The 8080 in 1981? Closer to 1975. And many have mentioned the bogus "no compatibility" claim.
One wonders if the whole thing wasn't a troll.
Just not the way you might think. An Intel Itanium-based computer running Linux64, Win64 (the codename for the 64-bit version of Windows 2000) or Windows XP 64-bit can run x86 (386, Pentium, Pentium Pro, etc) binaries unmodified. It will be significantly SLOWER than an equivalent x86 processor, because it does do it via hardware emulation, but it does do it.
Where the Itanium (and, I'm assuming, the Opteron/64-bit Athlon) really matter is in in large database and high-end workstation solutions. Basically, anything that needs more than 4GB of RAM. In these uses, it's not actually the processor speed that is needed, it's the RAM. The Itanium is meant for servers, yes. That is all the Itanium was designed for.
The cleverly named Itanium-2, however, is a horse of a different color. Not only is it faster (both MHz and IPC,) but it's cheaper, too! (You can get an Itanium-2 based system for about $3000.) The Itanium 2 at 900MHz is about twice as fast as the 'old' Itanium at 800MHz, performance-wise.
The only thing AMD has going for them (literally) is x86 compatibility. If it can run x86 code reasonably fast (i.e., a 1GHz Opteron running Pentium code at least as fast as a Pentium 3 1GHz) then it will be likely to take over the Workstation market from the Itanium 2. Unfortunately, I don't think anything could cause the Opteron to win over Itanium 2 in the high end server market.
Another non-functioning site was "uncertainty.microsoft.com."
The purpose of that site was not known.
If your motherboard was designed right, it would notice the overtemperature, react, and shut it down. The real problem is the heatsink falling off- if that happens, your CPU will emit magic smoke faster than the temperature sensor can react.
Tim
Omnia vestra castrorum habetur nobis.
Gimme a break! The reason has nothing to do with Windows XP. It has to do with databases more than anything. Crikey.
I am sure that Intel is really happy that the chief architect for their partner in their 64-bit efforts is endorsing the competing technology.
"The statement: "...is an entirely new chip that has no backwards compatibility with its x86 line of chips.." on the front page is simply not true.
"
Bullshit. Itanium is as compatible with x86 as my Athlon is compatible with the processors used in the NES, Nintendo 64, Atari 2600, Game Boy, Sega, and Commodore 64.
Tim
Omnia vestra castrorum habetur nobis.
Unreal Mode (aka Voodoo Mode, Flat Mode, or Big Real Mode) was a weird hack to allow 32bit addressing in real mode. It works because when you turn an IA-32 CPU on, it may boot in real mode, but it's not really doing true 16-bit addressing. It's just setting the limit register in the MMU to 1MB.
Therefore, you can set a single-entry GDT with a 4GB limit starting at address 0, kick the CPU into 32bit PE mode, and then go back into realmode, and the end result is realmode capable of accessing 4GB of RAM using the 32bit registers. Just use 0000 as a segment. (Of course, this weirdness also means you won't be able to use movsb or other string operations that expect realmode segments.)
People stopped doing this around the Pentium era because 16-bit instructions ran up to four times as slow as 32-bit instructions, and Windows was becoming ubiquitous, so it was easier to just use 32-bit protected mode with the same single-entry GDT.
(BTW, V86-mode is a hack where a task handle is put in the GDT and it runs for all intents and purposes as if it were a real-mode program. When an interrupt occurs, it goes through to an alternate IDT instead of the normal low-memory block, and the handlers run in protected mode.)
AMD processors are more reliable? What, did you have more system crashes on Intel cpus? Did you diagnose more crashes on your Intel system as CPU failures rather than software problems? Have you had any Intel cpu just quit working on you?
Lets remember the video on Toms Hardware with the Athlon burning up within seconds of loosing its cooling system. AMD did address the problem, but they put a large amount of the burdon on mobo designers. The Athlon uses a full 10 watts more than the P4.
There may be valid reasons why somebody would choose an AMD processor over Intel, but reliability isn't one of them. And for the past 6 months, neither is performance.
"The defense of freedom requires the advance of freedom" - George W Bush
I don't see it as being underhanded. Intel and HP have been very open about the new architecture. They want to move away from x86 (right now, in the server market at least). They have been saying that since Itanium was announced.
"The defense of freedom requires the advance of freedom" - George W Bush
Are there any ports of bochs that pass system calls through to the native system so that none of the actual OS is running inside Bochs? This would allow you to, say, run x86 Linux code on Linux PPC or Win x86 apps on Win ia64. This assumes, of course, that the system call numbers and arguments are the same across architectures. Maybe it would require too much OS-awareness in Bochs in order to fix the endianess, but it would be nice to move away from hardware x86 decoders.
Please someone tell me that all of the 64-bit mode instructions are the same length. (Maybe the caryover instructions from x86 need to be padded with nops.) Varaible-width instructions absoutely kill hardware or software decoding speed, especially if you're trying to parallelize it. Maybe we can all migrate to pure x86-64 instructions and slowly rid ourselves of the old x86 instructions?
Ideally, AMD would come out with a RISC cpu with an open source x86 emulator for the OS vendors to integrate with thier OS. I would love to be able to have comodity RISC or VLIW chips on pricewatch. x86 decoding is a waste of heat and chip realestate.
Copyright Violation:"theft, piracy"::Anti-Trust Violation:"thermonuclear price terrorism"<-Overly dramatic language.
Ah. You sound like the people who made Java. They confronted every potential problem with C++ and solved it by removing the offending feature. Multiple inheritence can be dangerous? Just remove it. Pointers can be dangerous? Just remove them. I like Java well and all, but I'd really like to see an extension to C++ that adds some of Java's new feature (like introspection and dynamic classes). Anyway, back to the topic at hand, what do you do when you need to program some device on the serial port? Or when somebody hands you a floppy disk because they aren't hooked into the net? Or your USB keyboard craps out and all you have is a modly PS/2 one lying around? All of these have happened to me recently, and boy did kick myself when I had to copy something to a floppy in the next 2 minutes before heading out the door and I realized I had disabled my floppy drive in the BIOS. Most of the legacy stuff these days hangs off the Super-I/O chip and stays out of the way of the rest of the system. They're not hurting anything and don't cost anything, so why get rid of them?
A deep unwavering belief is a sure sign you're missing something...
The story so far:
:-)
>Think nothing is impossible? Try slamming a revolving door
1 block of wood inserted between door "wing" and building edge.
then the door bounces off the wood, and spins the other way, pushing the 2 by 4 out onto the sidewalk.
The Opteron is compatible with all software made since the 8086. Therefore, the Opteron cannot truly be called new technology. It may have evolved in certain ways, but at its core it is no more advanced than the 8086. You can read AMD's whitepaper, and it will confirm: AMD knows that RISC, or specifically VLIW, is faster than CISC, but doesn't want to switch because of the installed base.
Intel, with the Itanium, takes the opposite stance. They know that CISC sucks, and that x86 was doomed from the start. It's not that "new things are better"; VLIW processors could have been developed in 1980, and if they were there would be no need for Itanium. But they didn't. So Intel wants to use 64-bit as an excuse to throw out x86, and start over the way they should have from the beginning.
Let's hope that Intel uses it's 75% marketshare power to win. It'll be unfortunate if AMD does.
There may be valid reasons why somebody would choose an AMD processor over Intel, but reliability isn't one of them. And for the past 6 months, neither is performance.
The magic word is VALUE. Compare, say, an nvidia nforce chipset system with an AMD cpu against similar Intel hardware. the AMD solution is way cheaper, like half the cost or something.
It may lag in performance, like maybe DDR isn't as fast at RDRAM or something. Actually I have no idea. All I know is that my Athlon 1.4GHz is way faster than the PII-350 I used to have.
(That last part was supposed to be funny...)
And go figure, I'm about to put an ATI video card in my nvidia nforce chipset mobo. So much for brand loyalty. Now if only I could see the point of installing Linux...
(puts on fireproof vest)
Their PA-RISC machines have been pretty popular for things like airline reservation systems and low-end graphics workstations.
They also helped out with Itanium.
The story so far:
>However, most people would be really pissed >off if you told them that if they bought a DVD >player they would no longer be able to watch their >VHS tapes.
What strange world would this have been? Not ours, for sure. Were manufacturers going to create DVD players that generated some sort of magnetic field that rendered all VHS tapes within the home unreadable?
Yes, it's a little-known clause in the DMCA.
Yes, the lead acts as a lubricant for the valve/seat interface. Beefing up the valve seats fixes that problem. Lead is only required as an ugly, toxic hack to allow the use of cheap valve seats.
These are my friends, See how they glisten. See this one shine, how he smiles in the light.
We don not want an 80x86 compatible chip. It's crap. It's an outdated CISC architecture. Has anybody read the complete white paper on the IA64 chip? Intel may have completely screwed up the implementation, but as specified (by H.P. originally) the chip is kicking ass and taking names.
Some extremely large percent of the silicon (and thus power dissipation and heat generation) on the athlon line is spent on the insanely complex variable length instruction decode needed to break old legacy x86 instructions into smaller pieces to feed to the well designed and powerful execution units.
A well designed RISC with optimization and caching hinds built into the object code will kick the living shit out of a 64 bit CISC hack built on top of an instruction set that was designed to run pocket calculators and automatic washing machines.
---
Play Six Pack Man. I
Or a DEC/Compaq Alpha workstation, or a Silicon Graphics MIPS Workstation.. And arent HPs PA-RISC machines 64bit too?
But the funniest of all:
A lot of WinCE pocketpc type things are using 64bit capable embedded cpu's. (a lot of the embedded MIPS clone CPUs used in handhelds can do 64 bit, eg NEC VR series).
I use Friend/Foe + mod-point modifiers as a karma/reputation system.
Completely different issue here. Pentium -> PPro -> PII, the ISA was basically the same. Incremental changes only, cost of the hardware itself was the limiting factor. x86 -> IA64 is a completely different ISA, many more factors involved, with software now being the primary issue.
So in other words, both companies are doing most likely the same thing, though Intel is slightly more underhanded about theirs
;)
No, not at all. Two completely different paths. AMD's strategy is obvious. Intel otoh is a bit more complex. They want people (high end people that is) to move away from the commodity x86 chips (currently P4/Xeon) and to this new ISA. Why? Well, two primary reasons, the first one is a bit lost in the shuffle, the second one plain as day.
Firstly, you have to know the history behind the Itanium/Merced. It was conceived back when Intel still considered RISC to be a major threat and the likes of AMD were really no competition at all. So the thinking was that x86 would never economically scale up to the levels that RISC could, and therefore a complete departure was required to ensure that they kept the performance edge (boy howdy do times change!). Their primary goal was performance, x86 compatibility was an afterthought (and it shows). It wasn't until it became obvious that x86 compat. was important (and the surprising upramp in x86 clock rates) did Intel realize that they were going to have to put way more effort into x86 compat mode then they ever wanted to.
Secondly, and this is probably most important now given the huge advantage Intel has over rivals in the clock rate category, is the simple obvious fact that there are no IA64 clones. If Intel can convince the market to move to the new architecture, they will once again have free pricing reign over the market. They can also make sure that follow any clone activities much closer (i.e. the lawyers would follow the clone activity much closer). This my friends is the big buy. Once again, a sea all to themselves, once again massive margins (well ok, more massive margins). No niggling AMD nipping at your toes.
So, there you go, more than you ever wanted to know and probably care about.
The point is that people have a lot invested in legacy software, and they don't like being told they can't use it with their new machine. This is what Itanium sales reflect.
When someone buys an Itanium they pretty much have to buy all new software as well (if they can even find what they need ported to Itanium). If we are to continue with the (admitedly weak) DVD/VHS analogy already established in this thread, it is equivalent to preventing buyers of DVD players from watching their VHS tapes.
"But I can have a DVD player and a VHS player at the same time, and there's no magical interference field or anything" you say. That's true, but most people aren't willing to suffer the inconvenience of having 2 computers that they must switch between in order to perform certain tasks. Therefore, when someone is buying a new computer that is incompatable with some psoftware which they prefer to use, say "this doesn't support that" is equivalent to saying "you can't do that if you buy this system".
In other words; if you want people to buy your system, it needs to do what they want it to do, and in the computing world, and especially the business computing world, that means running the software they have already invested in. There are a lot of companies running old Unix database systems that the users have to telnet into just becase that's what they've used for 10 years, it works, and switching to a new system would cost more than maintaining the old one despite the inconvenience. It doesn't matter if the new system is technically superior, if people can't run their old software they won't buy it, as is proven by the lackluster Itanium sales.
I hope that clears up my position a bit.
Under capitalism man exploits man. Under communism it's the other way around.
No, binary compatible is the 8088/8086, the former
8-bit to the mainboard and thus exclusively used in
IBM PC and XT, the latter one year earlier, but with
its 16-bit bus making mainboards twice as expensive.
The 8080 has a different command set, but with some
macroes 8080 assembly source code can be re-compiled.
My Karma isn't excellent, damn it! (And
Why support AMD when they take advantage of open-source developers to get their product supported while at the same time embed Paladium / DRM garbage in their products which will be used by Microsoft to extinguish Linux?
I will certainly only buy another AMD processor if I hear they are dropping this ridiculous 'feature'.
Comment removed based on user account deletion
Can you say Macintosh, PowerPC, and MC680x0? I knew you could.
General Relativity: Space-time tells matter where to go; Matter tells space-time what shape to be.
No its not better.
In CPUs like the Athlon or Pentium, esoteric and rarely used previous generation features are microcoded and have little or no impact on critical path performance.
The ability to create a high performance processor while retaining backward compatibility is just a matter of design. The two ideas do not necessarily interfere with each other in terms of performance.
The x86 backward compatibility hierarchy is actually quite reasonable excepting for the lack of registers. "long mode" in x86-64 fixes precisely this problem.
With "fresh starts" we end up with garbage like "MIPS" (exposed branch delay slots, makes forward compatibility to deeper pipelined chips difficult if not impossible) or "Itanium" (so complicated, its hard to imagine how they will improve architectural performance over time) or "Crusoe" (an extremely "simple" architecture that cannot go toe to toe with its contemporary x86 brothers.)
In the end the better implementation wins. That is all.
Well I shouldn't think the compiler technology will be too much of a problem by the time the Itanium hits mainstream. For whatever else you ahve to say about Intel, they do know how to write a good compiler. While their new chip is no small undertaking and requires a radically smarter (so to speak) compiler, I think it is something that they can pull off.
For that matter I think all in all compilers are evolving to this state already in a way. Orignally one of the major reasons for RISC design was that it was easier to compile for, compiler could generate more efficient RISC code and hence you didn't have ot tweak assembly language just to get deceant speed. Well compilers have gotten a lot better, espically high quality ones like the Intel compiler and they can generate very optimised code, even for CISC ISAs.
Ultimately, I think an EPIC (to use Intel's term) type architecture will be one of the most powerful.
A little like how Apple included the Classic and then Carbon compatibility level, so that Mac OS X can run old software. It meant that the first day I got the public beta, sure - there were a lot of little unix apps I could run at the command line, and a few little Aqua apps, but the majority of software I ran was through Classic.
As time has progressed (about 17 months later) there are plenty of Carbon and Aqua apps so that I almost never launch classic anymore.
- passion
This may be nonsense since I don't claim to fully understand how 32bit and 64bit differ at the application level, but here's my guess at the reasoning from a marketing standpoint:
Pure 64bit with no backward-compatibility -- this CPU is intended for dedicated software that's designed for 64bit from the ground up. I'd expect this to be aimed at primarily at the server and database market, where new apps are likely to be written simply because the old ones no longer handle the load -- while you're at it, might as well design 'em as pure 64bit from the gitgo and ditch the compatibility kludges. This CPU is likely to be a higher price bracket since the target audience is essentially the enterprise market.
64bit with 32bit compatibility -- that's for 32bit software that is already overstressing 32bit's 4gig memory-addressing limit. I'd expect this CPU to be aimed primarily at the CAD and video graphics market, where existing apps are likely to remain in primary use for some time (because they still do the job and are too expensive for their current market to replace). This CPU is likely to inhabit a lower price bracket since the target audience is independent designers, small studios, and the like.
IOW, it looks to me like AMD and Intel are courting two completely different and separate markets.
~REZ~ #43301. Who'd fake being me anyway?
Thank you©© That© was ¥ very helpful¥©
It's 10 PM. Do you know if you're un-American?
Have an "old style" chip and a new-style chip. Big apps then use the 64-bit chip and the old one's use the 32-bit chip.
Table-ized A.I.
Will the crack-addict moderators who modded this guys post to -2 flamebait please report to the front desk for an ass whumping? I don't agree with him(though in the days of the K6-2 I'd likely agree wholeheartedly -- I would have killed for a PII 366 to replace my K6-2 400 at the time!), but he's hardly flamebait!
If I had mod points, I'd right this wrong myself.
It's been a long time.
A dual 1Ghz Mac can emulate x86 and performs as well as a 266Mhz PII
A 667 mhz 64 bit Alpha can emulate x86 but is only as fast as a 200mhz Pentium Pro
An 800 Mhz Itanic emulates x86 as fast as a 166Mhz Pentium.
Linux can emulate a cluster on a single machine.
Any PC with two network cards can emulate a Cisco router.
Intel stopped marketing Itanium's x86 emulation mode because it is abysmally slow. The emulator is of course compiled on Itanium's still very immature compilers so it will improve in the future.
The Sledgehammer contains a complete x86 core and a complete 64 bit risc core. At 800mhz it outperforms a 1.6Ghz Pentium 4 running stock Windows XP and stock applications.
Running 64 bit SUSE, the Sledghammer performs as well as an Itanium at the same clock speed.
Sledgehammer is expected to ship at 2.0 Ghz. It's should perform as fast as a pentium 4 at 3.4 Ghz. Each processor has it's own memory controller so there is no shared memory bottleneck for multiprocessing. 2 processors should be exactly twice as fast using multithreaded applications. Sledghammer scales to 8 processors.
If voting were effective, it would be illegal by now.
I work on EPIC compilers and I concur on every one of your observations. The caveat is that every EPIC compiler optimizations is applicable to superscalar processors too. Hence the only reason why EPIC would beat superscalars is the reduced complexity of the processor itself which would (hopefully) make it cheap and fast.
.
In theory EPIC compilers should give amazing performance. In practice due to pointers, aliasing and lack of interprocedural optimzations they dont. It is interesting to note that Java, because of lack of pointers produces impressive code for EPIC processors and is highly amenable to optimizations targetting EPIC.
"There is only one company that has ever made a good VLIW compiler: TI, again."
I would like to think TI and Us
Nice and simple. Two chips, two busses, and interactions between the two different systems add up to a bloated, buggy, and (most importantly) expensive system.
steve
Oh, you're not stuck, you're just unable to let go of the onion rings.
I can't wait until the Voodoo3 is announced! Damn, that chip is gonna be sweet!
/^[A-Z0-9._%+-]+@[A-Z0-9.-]+\.[A-Z]{2,4}$/i
If Itanium fails, well then we've learned something about the processor marketplace. If, on the other hand, it languishes for a while, comes out in a new form (faster, better, stronger) and (after two or so more years) builds up enough inertia to be bought fairly heavily then it will be about as smooth of a transition as was made from vinyl to CD's.
The thing is, the lesson has already been learned with the Alpha. The Alpha was always faster/stronger/better and everybody knew it. It even had an NT port, but it didn't run people's apps so it did a slow death spiral until finally being bought up by Intel last year.
The Itanium occupies the same market niche that the Alpha did, but doesn't have the relative strength in design that the Alpha enjoyed, and that kept it alive for so long. Everybody drooled when Alpha was mentioned, but I don't see anybody drooling over Itanium.
Under capitalism man exploits man. Under communism it's the other way around.
Barton is coming out real some now, is 512K L2
cache on an Athlon enough for you, or would you
prefer to wait for the 1M L2 on the operaton?
Luckily there going to put a metal heatspreader
on the clawhammer. The K6's had then to, i'm
not sure why this skipped it on the Tbird to T-bred athlons.
Well, Itanium has backwards compatability to ease porting, but nothing worthy in support or 64-bit software for running a mid-end server yet. Say IA-64 software begins to appear and we come to the point where a good server might run a good chunk of IA-64 software and some x86 too (no good ports, perhaps). What would stop Intel from designing a riser card like the SunPC (right?) with a Pentium chip on it? For those that need fast x86 as well as fast IA-64, this could work.
"Open Source?" - Press any key to continue
Anyone who really needs the power of a 32 bit desktop was already happily using a VAX workstation 15 years ago.
Best Slashdot comment ever
and here is the IHS (integrated heat spreaders) at anandtech.
yes, it is in fact very good news they decided to add it. putting and removing heatsink+fan on AMD CPU is not that easy to do nowadays (thermal paste and so on). fortunately it's not something you do every day....
I would have killed for a PII 366 to replace my K6-2 400 at the time!
To be fair, K6-2 performace was VERY dependent on the motherboard and the size of its cache. With a decent board it was competitive with the PII's. I had a K6-2 500, on a decent MSI board and it performed nearly identically to a Celeron 400 OC'ed to 500 mhz. The real neat thing about the chip was that it could be used in old socket 7 motherboards, a 2 times multiplier would be interpreted as 6 times by the chip.
My company began really using AMD's in workstations and (non MP)servers for our clients when the T-birds came out.(Athlon classics ran way too hot) We spent alot of time testing to establish stability and found that once again that the motherboards made all the difference in the world. We settled on Gigabyte boards with the Via kt133 chipset, these had the addded security of dual-bios. After a couple years in service we havn't run into any problems with the 50 or so systems we built. Intel, whom we also use did have some serious well documented stablility issues, remember the pentium III 1.13 that had to be recalled.
Could Jesus microwave a burrito so hot that he himself could not eat it? (HS)
Yet the Alpha seems to be forced into complete non-relevance by the mass media. (Witness the Wired article, where it's not even mentioned.) Is the Alpha really that irrelevant? Is the experience gained (both technically *and* marketing-wise) to be tossed and never thought of again?
The performance thing is pushed because the review sites need something to write about.
:-)
They need to start comparing value for value... An athlon 1900 costs the same as a P4 1.6Ghz. The benchmarks wouldn't be so close, then
The main reason I ditched Intel is the whole backward compatility thing.. I was sick of having to buy a new motherboard (and sometimes even a new case) just to get a faster processor. With AMD they're all Socket-A... if you bought an old Duron two years ago you can just stick in the latest Athlon without any extra hassle (except perhaps a BIOS flash).
At work we're entirely AMD now because the Intel line don't support dual processors - a requirement if you're doing development on Win2k.
Actually, the Athlons do have a thermal diode, just like the Intel chips. It's only that older motherboards dont use the cpu thermal diode, but use their own external one instead. And that one cant react fast enough to a heatsink removal. It will react to fan failure tho.
If you get a motherboard that does use the internal one you dont have a problem.
Of course... I cant say I find it likely that a heatsink would fall off. You'd have to drop the box from a pretty fair height to manage that.
When is it time for us to move on from such an old architecture? Surely there is some luggage in there we can now do away with?
Even software languages have broken compatibility at times to advance. Can't hardware do the same?
"I basically just didn't get the part about DVD making VHS _not_ work"
/. (or most other places) huge great flamewars break out over X-Box vs. Gamecube vs. PS2 vs. whatever, because some people won't buy them all. VI vs. Emacs arguments are legendary, and you can even run both in adjacent windows of the same machine!
It does sound strange, I know, but I see people all the time who don't want too many boxes in their living room.
Also, every time consoles are mentioned on
Lots of people seem to think they can only own one of a type of technology at once; it seems to have turned into a hardware equivalent of supporting sports teams or something.
"I Know You Are But What Am I?"
in integer performance it could keep up, but FP is what I was looking for. since 3dnow didn't really catch on(or speed anything other than Quake II on a voodo2 up significantly).
It would have been interesting to see what the chip was capable of at 4*100mhz, rather than the 6*66 I had it clocked to, but my old and crappy PC Chips motherboard didn't support that(or 400mhz, for that matter, I had to underclock it to 6*60mhz to run without crashes).
It's been a long time.
Comment removed based on user account deletion
Physical memory was extended to 36 bits by the PII (or was it PPro?).
For greater than 4Gb virtual, you can still use segmentation. A process can have (what, 12 bits of local segments, 12 bits of global segemnts) 8192 segments, each with 4 Gig memory. Hardly a hard limit. It just means data has to be broken up. All you old 286 programmers know how to do that don't you.
Note: AMDs X86-64 will supposedly discontinue support for segmentation in 64-bit mode.
There is nothing so silly as other peoples traditions, and nothing so sacred as our own.
As you said, each processor has its own memory. If a processor needs data in another processors memory space, it has to request it over the Hyperchannel bus, not quite a local request (NUMA).
Multithreaded programs share a lot of data between threads.
Sooo. what you really mean is that two independant processes will each run as if on a dedicated processor (which they will). But multithreading will still have some memory and bus contention.
There is nothing so silly as other peoples traditions, and nothing so sacred as our own.
Not to knock the Doozie, but note that many cars today will reach 120mph with about 1/3 the engine displacement. Not to mention they probably produce 1/10 the pollution and are immeasurably safer. So, see kids, the automobile industry is capable of making progress.
Hey kids, there's only 5 days left 'til Yak Shaving Day!
Bullshit. Itanium is as compatible with x86 as my Athlon is compatible with the processors used in the NES, Nintendo 64, Atari 2600, Game Boy, Sega, and Commodore 64.
Untrue. Your Athlon cannot emulate the 68000, Z80, or 6502 chips (what's in the N64?) without additional software. The Itanium has x86 emulation built-in. Thus, it is false to say the Itanium has no backwards compatibility with x86. The fact that it performs emulation is irrelevant to the discussion. In fact, most of the modern x86 chips could in a very real sense be said to be emulating x86, as they translate x86 instructions to a more RISC-like instruction set internally.
Hey kids, there's only 5 days left 'til Yak Shaving Day!
What would stop Intel from designing a riser card like the SunPC (right?) with a Pentium chip on it? For those that need fast x86 as well as fast IA-64, this could work.
What is stopping them now? Seriously, such a thing would make an Itanium system much more attractive to potential early adopters. Of course, they'd still have to deal with the fact that the Itanium offers no real price/performance advantage over competing architectures, but it would at least remove one black mark against them.
Really, though, I'd say the boards are already done. It seems like it would be a fairly minor matter to have an x86 slave processor in an NLX (or similar backplane style) based system. IANAEE, though.
Under capitalism man exploits man. Under communism it's the other way around.
Interesting. I was under the impression from the articles I've read that Compaq killed Alpha off and sold everything except the engineers who defected to AMD to Intel.
Under capitalism man exploits man. Under communism it's the other way around.
All of these cars could be tuned to run on 87. Yes, you'd give up a little power - but on a 95 Camaro, or a Porsche 911/996, you wouldn't miss it much.
(The Porsche may be an exception, you'd have a hard time getting the turbo model to run nice on 87)
These are my friends, See how they glisten. See this one shine, how he smiles in the light.
Value is a reason many people cite, but the difference is not that much.
Example: According to Toms Hardware, the Athlon XP 2200+ performs just about as well as the P4 2.2 GHz (which is why AMD choose to name it the 2200). In a few of the benchmarks, it outperforms the 2.2 GHz, but mostly it is right on par. Now, comparing the latest prices (according to pricewatch), the AMD is about $210 and the P4 is about $226 (a whopping $16 difference).
What about the cost of the entire platform, you might ask? Look at the large OEMs- similarly equipped Intel P4 2.2 GHz PCs are usually within $50 of AMD XP 2200+ PCs. AMDs are cheaper, but its really not that big of a difference.
Granted this comparison doesnt scale to the middle of the road processors (the 1.6 GHz, XP 1600+), but the price difference from OEMs is much less pronounced. And Intel just announced a 63% price cut the 2.5 GHz when they release the 2.8 GHz later this year.
"The defense of freedom requires the advance of freedom" - George W Bush
Yes, it is also now the main processor on their XVR-1000 graphics card. It seems that a VLIW CPU is very appropriate for graphics, where parallelism is trivially found. It is somewhat odd that the 'J' in MAJC is "Java"; I guess they originally envisioned different applications for it.
I have yet to see competitive comparisons between the XVR-1000 and other cards, but I suppose it does fairly well with two 500MHz CPUs on board.
Healthcare article at Kuro5hin
There aren't many features of Java that couldn't be implemented in C++. Runtime introspection, for example, *has* been implemented in C++, as has object serialization, all without putting a VM under it.
A deep unwavering belief is a sure sign you're missing something...
> Can you say Macintosh, PowerPC, and MC680x0? I knew you could.
:-)
Amazing how something so condescending could be so wrong.
The Macintosh line could easily move over to PPC from 68k because of one factor and one factor only: the PPC was so much faster than the 68k that 68k apps could run in an emulated 68k environment at decent speeds. And then of course, a huge motivating factor was that the 68k CPUs had reached the end of their useful life and ween't going to be produced at higher speeds.
Fast forward to today, and the situation with Itanium is very different. It can't run x86 applications at near-native speeds. It doesn't offer a compelling upgrade path for users, especially since all their old software and games won't run on it and nothing it offers makes up for that. And then we have AMD's x86-64 Hammer, which offers all the advantages of a 64-bit CPU but with complete backwards compatability with existing 32-bit x86 apps. Many people bitch about the x86 instruction set and the limitations of the architecture, but the fact is instructions sets are nearly meaningless nowadays now that instructions are predecoded into micro-ops and treated as they would be on RISC processors. And the limitations of the x86 architecture are solved in the expanded x86-64 architechture.
Backwards compatability with no discernible performance drawbacks is always preferable to no backwards compatability.
Chasing Amy
(We all chase Amy...)
"The more corrupt the state, the more numerous the laws"-Tacitus
> which still beat x86 designs, megahertz for megahertz.
But which can't seem to scale well enough to compete performance-wise in the real world. A 1GHz PPC may well beat a 1GHz P4, but since the P4 is available at 2.5GHz, that hardly matters.
Chasing Amy
(We all chase Amy...)
"The more corrupt the state, the more numerous the laws"-Tacitus
Wow, thank you for the brilliant and insightful remark. Incidentally I'm pretty sure my mom's pussy didn't stink in 1929, as it did not exist yet. My grandmother's might have, but I can't very well ask her as she's no longer with us.
Hey kids, there's only 5 days left 'til Yak Shaving Day!
> Sledghammer scales to 8 processors.
;-)
8 AMD processors?
I'd have to swap out my 500 Watt power supply for something that plugs into my oven's outlet.
You certainly will. Sledghammer is expected to suck power like a cat in an exaust pipe. It is also expected to have more flop/s/watt than anything yet.
I'll be happy to pump 1000 watts into the Sledghammer server that replaces four 500 watt servers.
If voting were effective, it would be illegal by now.