Slashdot Mirror
Is Prescott 64-bit?
unassimilatible writes "According to The Inquirer, Intel's new Prescott has 64 bit instructions lurking inside. Could really rain on the parade of those who thought the new Athlon 64's would be supreme - especially when you look at Intel's price roadmap. Don't run out and buy an Athlon 64 just yet..."
I haven't even run out and bought a 32bit. Contiki and c64 1mhz 8bit power keep me warm at night.
Be you Admins? nay, we are but lusers!
Intel, those typical insensitve clods, are spreading FUD around to kill AMD.
Or there will be virtuly no software for it when it comes out and for months to come. AMD has had books on the x86-64 instruction set for years now. Not to mention emulators have been available for almost as long.
"I'll stick with Intel, thanks. Any of you guys actually have a *good* AMD processor?"
My 450MHz AMD K6-2 worked fine, at 4.5x100MHz, my 1.2GHz Athlon Thunderbird worked fine, and my Athlon XP2400+ (2.055 GHz after some interesting bus overclocking) works just fine.
I've never had a problem with them. Do you know what you're doing? Setting the voltage levels is required on the older boards, and that actually means reading the provided motherboard manual.
I'm looking forward to a dual-AMD 64 bit configuration for home at some point, it looks pretty sweet.
Do not look into laser with remaining eye.
This is a disaster for Intel, and if you follow along, HP, which is trying to sell Itanium solutions to counter IBM. I love big blue and AMD, so I can't say I'll shed a tear.
SCO: 800-726-8649
Verisign: 800-361-8319, 888-642-9675
Diebold: 800-433-VOTE (8683)
I'll be able to afford it, and then run a thirty-two bit extension and GUI shell to a sixteen bit patch to an eight bit operating system originally coded for a four bit microprocessor and sold by a two-bit company that can't stand one bit of competition, on a sixty-four bit processor! :) Can't wait! 178$ - wow! :) ;)
Uhm... Maybe I'll buy it and install linux on it, all the big good games are coming to linux now it seems
Doubt that I will totally be rid of windows in the next few years though. You know. Work and such
Any technology distinguishable from magic, is insufficiently advanced.
The "Prescott" has 2 32 bit cores, but the "secondary" core is missing an AGU, among other things, and this is pointing to the fact that most Prescotts have some sort of 64 bit functionality in them, but their keeping their lips shut about it?
So it seems to me that possibly Intel are waiting to see how AMD's 64 bit chip goes, and if its going better as a 32/64 bit chip then Intel's Itanium, release their Prescott with "fully" added 64 bit functionality?
Am i correct in my logic? I can't really follow why they're keeping tight about it.
Think nothing is impossible? Try slamming a revolving door.
"I'll stick with Intel, thanks. Any of you guys actually have a *good* AMD processor?"
Yes. Both my machine at work and at home are dual Athlons. Not only do they work great, but they also were very good price wise. On top of that, Lightwave (3d Rendering App) competes very nicely with the P4s, though LW is heavily optimized for Intel.
I'm not an AMD zealot, but I guarantee you that if I had the problems you did, my AMD'd be in the trashcan in a heartbeat. I can't afford aborted renderings.
"Derp de derp."
II aamm rruunniinngg aann eevvaalluuaattiioonn PPrreessccootttt rriigghhtt nnooww,, aanndd II aamm nnoott ppoossiittiivvee,, bbuutt II tthhiinnkk tthhee eexxttrraa bbiittss mmaayy nnoott bbee iinnddiivviidduuaallyy aaddrreessaabbllee,, bbuutt aarree jjuusstt uusseedd ffoorr rreedduunnddaannccyy..
It's hard for thee to kick against the pricks.
Hey! There's a rumour that Intel might have not bothered marketing their new developments properly. You should ignore what is on the market and working just in case you can tweak a future chip to get something a bit like it. You heard it here first!!!
Idiot.
The allegedly informative statement is couched in so many conditionals that they of course can once again squirm their way out of any uncomfortable spot they might get stuck in.
Plus, I don't see Microsoft supporting not one but TWO Intel-specific 64-bit platforms.
Hold on to your Athlon64 pre-orders, boys and girls.
Since 1997, all my machines have been AMD's. The K6-2 is still alive, actually. One of them (a Duron 600) has been running 24x7 for the last 3 years. My gaming rig's a dual Athlon MP2000+. My current workstation's an Athlon XP2400+. I've NEVER had any problems with them, either hardward or software (Linux).
... I know. I've administered Intel-based servers).
My biggest problem is what to do with the old mobos and processors that I put aside due to upgrading.
No, I've never had a reason to spend more for so little (it's even arguable whether you get more for spending more
Chip Architect was speculating on this way back when intel's 64 bit extensions were still called Yamhill. They make some interesting observations that lead them to belive the second 32 bit ALU was to allow for 64 bit integer operations in a 2x32 bit format. And not to assist with eliminating resource shortages in HT as some others had suggested.
And even if that does pan out it's highly unlikley to appear in desktop Prescott core chips anytime soon. Seems much more like something you'd find in Xeon MPs and later DPs to eliminate the need for that hack they call PAE.
Though i hardly see how 'somebody told us a seinor exec said' makes Slashdot.' (I understand that's what the Inquirier bases most of their news on, i thought we had slightly higher standards of reliability)
For one thing, I wonder what its physical external address bus looks like. Can it address more than 4GB of physical memory without paging? The Athlon64 and Opteron can. In addition, they are discounting the benefits of an on-board memory controller. This feature alone is a huge performance boost. To top it off, AMD gear comes with HyperTransport and a host of other goodies associated with AMD, like nForce chipsets with the best on-board sound of any integrated solution (and I don't suspect this will be different with nForce3 chipsets). In short, it seems like Intel is starting a concerted marketing blitz against AMD but with little avail. With the Prescott and this new extreme edition P4 with 2.5MB cache (I shudder at the yield hit that much cache has per wafer for them), we have a lot of sudden refocus towards Intel just prior to the launch. Product quality counts, but so does marketing.
You're reasoning is faulty. In fact, it is the presence of competition and the inability of Intel to completely shape the PC platform that is holding back technology.
Look at the Apple Macintosh. The platform is completely proprietary, and controlled by Apple. Apple successfully switched architectures in the mid-90s, from the CISC 68000 to the PowerPC. These architectures were completely incompatible at the machine code level. But since no one was selling souped-up knockoff 68000 Mac clones in competition, the PowerPC was the only way forward and all users eventually upgraded. Apple is going to do the same thing again, this time to the 64-bit platform, and no one is dragging their feet this time either.
This happens all the time when hardware is proprietary. Sun, HP, and SGI all used to sell workstations based on the 68000 architecture, and were able to transition to RISC architectures due to the proprietary platform. But look at the PC architecture...the same outdated CISC architecture that was used in 1981 is still there in today's PC's. If Intel had complete control, we'd all have switched to the i960 or Itanium by now. But no, there is competition, so we're stuck in the past.
Itanium is a full fix to the problem. The horrendous x86 ISA is completely replaced by an explicitly-parallel (EPIC) instruction set that has all the nice properties of a RISC machine (easier to compile for, less stress on the memory system as you get 128 registers, easier for the machine to decode the instructions as they're fix format and don't require RISC conversion, etc.). The problems with it are:
1. You need a compiler that "knows" how to bundle instructions effectively (a VLIW-compiler). GCC clearly isn't there yet (it's not uncommon for the intel compiler to beat gcc by 30->50% when running computationally-intensive stuff)
2. Being completely different than x86, it can't be very efficient at emulating x86 programs.
AMD partially fixes the problem by extending the x86 ISA to 64 bits, *and* adding 8 general purpose registers. Because they just extended the ISA, running old code is just as fast. Furthermore, new code can benefit from from the extra 8 registers, and run even faster.
For the short term the Opteron is a pretty impressive chip, but I really don't see how AMD is going to stay on Moore's curve with such a shitty instruction set architecture.
P.S. Clearly 32 bits can only address 4GB of RAM, and for *some* servers more addressing space buys you something. But I'd say they are a very small minority.
The Raven
Fear, uncertainty, doubt ... not just the tool of Microsoft!
Let's see ... the Athlon 64 is out, officially, in a few days ... Intel's 64 bit part, the Itanium, is having trouble shaking its nickname, Itanic ... lots of developers are excited 'bout having a chip running 64 and 32 bit software.
Solution? Don't make a better chip ... just float a rumour that you'll be producing something better with some 64 bit instructions... Real Soon Now! With luck, you'll tank the sales of your competitor's chip, without doing any real work!
Blah blah blah.
No. It's just a really bad rounding error in the FPU.
If Athlon 64 doesn't take off, Intel could keep things bottled up untill needed, or even nerver turn it on, letting consumers get 64 bit computing in a future chip that they've had time to improve the instruction set on or something.
It really is an interesting idea, and quite a consipracy theory. Is it true? Who knows! But with all the hub-bub around the Opteron and the upcomming Athlon 64, I wouldn't be suprised if Intel were to drop a bomb like this soon. Just think. Intel first steals AMD's thunder by anouncing the P4EE. Not only is it announced first, but it trounces the competition in benchmarks (this is speculation, I haven't seen any numbers). If the P4EE is fast enough in benchmarks and the price is competitive with the Athlon 64, AMD could be in some trouble. Now if in a few months, Intel announces something like this, AMD's savior that they seem to be betting the farm on could be in BIG TROUBLE. If this happens, AMD's best hope is that Intel DOES use their instruction set, because if they don't things could get very ugly.
So will any of this happen? Who knows! But that can't stop me from speculating! There is one last thing I'll comment on. If Intel does release a 64 bit processor soon, and doesn't use AMD's instruction set, there is a small possiblity that THEY (Intel) could be in trouble if the Athlon 64 (and friends) make a big enough splash. They might come too late to the party to make big decisions (like which instruction set rules).
These things seem a bit more likely, given that Intel seems to be in trouble right now (IMHO). While they are ratcheting up the P4 fast, the fact is that they weren't planning on 64 bits any time soon, AMD has forced the issue on them. If AMD is right, that will put them in trouble. And anyone who follows this kind of stuff knows that Intel has some major heat issues. Current opterons put out what, 70 watts? And some of Intel's upcomming chips are looking at 120 watts during usage (maybe as high as 150 under full load). Between heat, stagnation, and pathetic sales of the Itanic, Intel seems to be in Trouble.
The last thing that I'll say is a message to Intel: when you move to 64 bits (or even if you are just going to stick with 32 for a long time more)... DROP THE NAME PENTIUM. I'm tired of it. There have been FOUR of them (not counting all the different core revisions of each one). I know you have marketed that name for years, but it's time to move on. When will it stop? The Pentium 5, which you might call the Pentium Pentium, or Pentium Squared? Will I have to wait untill the Pentium 17 before you get a new name? Come one guys. Time for a name change.
Comment forecast: Bits of genius surrounded by a sea of mediocrity.
I don't know if this "dual 32-bit" thing is very plausible. Being able to do 64-bit operations is perhaps the most useless feature of the upcoming 64-bit processors. The big things about AMD64 is the larger compiler-visible register file and the ability to address > 4GB of memory.
A deep unwavering belief is a sure sign you're missing something...
Athlon chipsets sucked rocks for a long time, and were really unstable. But VIA finally got their act together, I think with the KT133A.
AFAIK, other than stomping on occasional driver bugs, Athlon chips have been pretty excellent ever since. I have an Athlon 1900+ on an ASUS A7V333 that's rock solid, and a new Athlon 2500+ on an Nforce2 board that's not quite as solid, but which is still pretty good.
I'd like to see some improvements on the NForce2 chip stability. It's not all the way there yet, in my opinion. But the VIA chipsets are extremely solid.
Float a rumor that their next chip wil have some 64 bit instructions, so don't upgrade to that competitor that is true 64-bit?
You do realize that there has been no such thing as a "CISC processor" since the Pentium Pro came out. Underneath the X86 bytecode VM, Pentium IVs, Athlons, etc. are highly advanced RISC cores with multiple concurrent execution units.
The main reason that the huge expensive power sucking Itanium scrapes out a small lead in benchmarks over X86 CPUs is because of its expensive huge power sucking cache.
AMD should kick it Old School.
What would happen if they succeded in causing AMD to go out of business? Then the only other option for consumers are VIA C3's and Transmeta which will never match Intel's. So then once again at the top of their game and after buying auctioned assets off AMD's corpse they ratchet prices back up and we lose out on innovation.
Can you say "astroturfing to ruin the release of Athlon64 coming up next week"?
I knew you could!
In other news, SCO sues Intel for the use of their na... SLLLLLLLAAAAAAPPP!!!
Ouch, that hurt!
getSexySig();
As the author of the article, I had to REALLY make things vague. The people involved would be hurt badly by Intel if their names got out. Some of the situations that were told to me make it quite apparent who was leaking. That was as specific as I could make it :(.
-Charlie
I'll stick with Intel, thanks. Any of you guys actually have a *good* AMD processor?
Well, I've got the old K5PR133 sitting on a shelf now that did years of sterling service, first as my main desktop CPU, then moving back to the firewall CPU, until it got replaced by the K6 II 350 (which is still in the firewall machine, and has been there, running solidly, day in, day out since about 1999, and was my desktop CPU before that,replacing the K5).
Replaced the K6 II with an Athlon 700 (original slot A). That ran fine, until the board fritzed, due to the old capacitor problem that ran rampant in late 99, early 2000.. The board lasted until 2002. The only reason it's not in the firewall box now is that I can't get another board for less than it'd cost to put a faster athlon in a cheap board..
The 2002 set I bought was a nice Athlon 1700 in an Abit board. Ran stably and never a problem.. That's now in the girlfriend's machine (built one for her from the last generation of hardware I had lurking), and now I'm using an NForce2 board with an Athlon 2500 (Barton core).
No problems with any of them.
If the college computer broke, are you sure it's the CPU? Not memory, motherboard, power supply or any of a myriad of other issues?
For the laptop, is it a problem with the manufacturers not putting good cooling and airflow in the laptop (or, heaven forbid, a desktop CPU in a laptop case to save money/add a little extra speed)?
By all means, stick with Intel if it keeps you happy, but I've had a long history of using AMD chips, and I like 'em. If I saw a reason to use Intel's chips, I would.. I just never have to date...
People just assume that G5 consumes this enormous amount of power because of all the fans in the G5 desktop. This isn't true. Even the 2G takes only about 40 watts or so. One P4 3G takes in the range of 80 watts of power. All of the extra G5 fans are to make the cooling quieter.
I'm glad to see someone finally point this out. The exact wattage number is 46.7 watts for the 2 GHz PowerPC 970 "G5" running at full speed (2GHz CPU and a 2:1 multipler for a 1 GHz FSB).
A 2.4 GHz P4 (400 MHz FSB) uses 62 watts, newer P4s use even more. Prescott is expected to use 100 - 105 watts. (And this is totally ignoring the even further power needs of the "extreme" edition with its added transistors for on-die L3 cache)
Apple has always seemed to overengineer the heatsinks and fans in their desktop model, for about as long as I can remember. Oddly, many of the PowerBooks use a much different "transfer the heat from the CPU, Chipset, and GPU right to the bottom of the case" cooling method.
However, the compact legacy CISC instruction set does conserve on instruction cache space. This offsets much of the cost of the conversion logic. Moreover, it allows custom optimizations for the exact architecture du jour without affecting binary compatibility.
And any way you look at it, you have to read from memory a lot more often with 8 "general purpose" registers than 32 real GPRs, which is what most sane CPUs have.
Many modern X86 CPUs have more than 32 real GPRs which are utilized by register renaming. Like quantum mechanics, the processor state for any given instruction is smeared out over time and space, and the CPU is operating on many instructions simultaneously. The number of visible registers just doesn't matter as much as it would seem on the surface.
Itanium doesn't have to do this, PowerPC doesn't have to do this, no modern ISA requires this nonsense.
They will when somebody figures out the next architecture trick that doesn't match the assumption of the designers of their ISAs. Take a look at history; remember when MIPS stood for "Microprocessor Without Interlocked Pipeline Stages"? What did the R4000 introduce? Could it be - interlocked pipeline stages? Exposing CPU implementation details to the software is not something that wears well over time.
It doesn't just "scrape out a lead" in floating-point benchmarks, it absolutely destroys the x86 competition.
That's because the FPU has not been very important in the X86 market up to this point. Business and multimedia apps just don't need it. If AMD or Intel put their efforts into an X86 with ultimate FPU performance, it could match or beat the Itanium.
I suspect that Intel took advantage of the huge schedule delays in the Itanium to throw in more FPU horsepower because it had to beat the consumer-grade chips on something.
And oh yeah, its running at what, half the clockspeed of the P4? If Itanium had the same economies of scale behind it at this point, there would be no competition.
As I said, the cache and memory architecture is the primary factor in the performance of CPUs today. Clockspeed, instruction set, registers ... who cares? Everything that's not cache is only a small fraction of the die size.
All of that hardware architecture stuff is a red herring. Worrying about those non-issues has caused the Itanium schedule to slip nearly a decade while they desperately tried to write a C compiler that could statically wring out performance from their brittle concurrent execution model without the benefits of the run-time statistics information available to the X86 code translators.
I didn't consider timing when I wrote the story, or any of it's predecessors. Silly as I am going to the A64 launch tuesday. Anyway, I have been chasing this story since the chip-architect articles. The timing was unfortunate, but it wasn't an Intel plant, that much I can assure you.
For about 3 months, I have known there was 64 bit functionality there, but I didn't have enough to prove it to my own satisfaction. I chased leads, interviewed people, and got that info.
The fact that IDF brought me into close proximity with a ton of sources was the thing that got me so much info so quickly. There was only one thing from Intel directly, the rest were from third parties supporting the chip. If IDF had happened last January, I probably would have gotten the info then.
-Charlie
You can order amd64 systems from places like appro and Penguin Computing right now, with decent sized collections of 64-bit applications provided by popular distributions such as SuSE. Let's not forget that the amd64 CPU's can run ia32 binaries at speeds faster than many ia32 CPU's and on a system with an amd64 kernel allow for more aggregate address space consumption across processes and the ability to install tremendous amounts of physical memory for buffers and cache even if individual processes can only take advantage of a few gigabytes.
With other groups like the Debian project well underway in their amd64 porting efforts, you can expect thousands of popular applications built for the amd64 platform. There's tons of software available for amd64 already, and you can bet by the time that AMD releases their "Athlon64" or whatever they're targeting the low-end market with, there will be even more.
I'm afraid this line of reasoning just doesn't cut it.
Intel does not want a 64bit x86 on the market. They want to lead everyone to Itanium where they don't have those pesky AMD guys competing with them.
It's for this precise reason that everyone SHOULD run out and buy an Athlon64. If nobody buys them, Intel will have no reason to jump into the 64bit x86 market at all.
I for one can't wait for Athlon 64 to hit the market... I need a viable 64bit Linux workstation solution and I need it yesterday.
-cjs
And why isn't there an Itanium 2 or Power4 running at the same clockspeed as the P4? It's because they can't. To do more work per clock, they use more logic, and that takes more time. Don't you think that Intel would have a 3.2 GHz Itanium on the market now if it were technically feasible?
All of these CPUs use similar fabrication technology. This technology is capable of a certain number of fundamental logic operations per second per square millimeter. The P4 uses high clockspeeds only because it is marketed to users who think that MHz==performance. If marketing requirements were different, the P4 would have been designed to get the same performance out of 1/10th the clock speed with the same die size and the same manufacturing cost.
The AMD vs Intel war has been going on since the 386 days. When you bought a 386 or 486, you probably had an equal chance of getting AMD or Intel, and you didn't even know it. AMD won the war in the 486 market, remember the DX4120 or DX4133? Those were AMD chips. Intel just has a better marketing machine. Intel Inside(r) came around in the late 486 and Pentium days. AMD nor Intel will win or lose in this battle. The only people that will come out ahead will be the consumer. I just can't wait till the price cuts come around.
It's why they don't call them laptops anymore. Look on any major computer manufacturer webpage, you'd be hardpressed to find the word "laptop"
Dell - Notebook
HPAQ - Notebook
Alienware - mobile gaming
VoodooPC- mobile
They don't call em laptops anymore because, as you've noticed, they often don't work well in your lap
Then why, precisely, are you here chatting about Intel and AMD 64 bit chips when GNU/Linux has been running on 64bit DEC Alphas for over 5 years, And YellowDog Linux runs very nicely on the G5?
Article X: The powers not delegated... by the Constitution...are reserved...to the people
Ahhh. So there's another variable... you need x86 compatibility. that's a different animal than simply needing a 64bit Linux workstation as you initially stated. I will warn you: much open source software is written with x86 in mind and had things hard-wired to that platform. Even when things compile on a 64bit system, they tend to fail in interesting and unpredictable ways. I've found that ignorance and lazyness seem to prevail in the open source movement when it comes to truely portable software.
I don't know exactly what you mean by your last comment. I don't really know of any x86 specific hardware or peripherals except for the MBs and CPUs. Just about every other peripheral or piece of hardware will plug in to many Alphas and any recent Mac (drivers and firmware aside).
I've used many video and network cards off-the-shelf as well as standard RAM in my Alpha and use standard "PC" memory, drives and peripherals on my 5 year old Mac.
I have checked out x86 price/performance. It's not been enough to get me to purchase any of it. I get more real work done on my Mac per dollar than on x86 (no virus downtime, few system updates, etc), and my Alphas put out more heat per dollar than any x86 could in it's wet dreams.
Article X: The powers not delegated... by the Constitution...are reserved...to the people