Slashdot Mirror
What Happened To SMP For AMD processors?
Christopher Cashell asks: "Does anyone know what is going on with AMD and support for multiple processors (SMP)? I love AMD CPU's, but I've also come to love dual processor machines. Ever since the Athlon was still an 'in progress' chip code named the K7, and AMD stated that the CPU would support SMP, I've been drooling over the idea. Now, especially, I would love to have a dual CPU Duron box. Has anyone heard anything? I couldn't find anything on AMD's site about it. As I understand it, the CPU supports SMP, so it's just a chipset issue, right? Is AMD working on a SMP chipset? And if not, are any of the other big mobo/chipset manufacturers considering it?"
Almost right..
Indeed, Hotrail to cancel their 8 way AMD chipset, but the company who took it on themselves to work on is Alpha Processors Inc (API)
Hetz (Heunique)
It's a known feature of the Athlon. In fact, one of the things AMD was proud of is that in SMP configurations, each CPU apparently gets its own frontside bus into the chipset IIRC. Even the K6 (Maybe, definately the K6-2 and K6-3) supported SMP. Problem was, it was not the same SMP scheme as used by the classic Pentium, and there was never an SMP mobo made for it. I hope the same thing doesn't happen to the Athlon. (Although from what I remember, an SMP chipset was due relatively soon on AMD's chipset roadmap. As in by the end of the year, I think.)
retrorocket.o not found, launch anyway?
That's why the K6 had SMP support.
But unfortunately, the workaround made the K6 SMP protocol incompatible with that of the Pentium. And AMD wasn't big enough at the time for any chipset/mobo manufacturer to justify creating a new SMP mobo for a small market.
retrorocket.o not found, launch anyway?
Just to let you know - The K6 SMP support was OpenAPIC. Too bad no one made a mobo for it. Don't know if the Athlon is or a different protocol. (I think it may be different - Considering how different the bus/cache design is from the old Socket 7 designs in general.)
retrorocket.o not found, launch anyway?
You named exactly the opposite reasons that one should consider buying a Dell. They are not quality systems (really - they use lesser quality components, unless you pay big bucks for their higher grade models). They are not easy to maintain (their "thumb-buster" case design is my arch nemesis). Or at least, not any easier than any other PC in the world, and in many cases harder, with their poor case design.
They are not even homogenous. Go to Dell's site, and look up a given model, then check, say, audio drivers. There will be 15 different audio drivers mentioned. Yes, they really do use different audio chipsets in systems with the exact same badge designation (i.e. Dimension XPS600R). And no, they can't tell you what chipset you have in your system if you go to their web site and punch in your unique "service code" from the back of your system.
About the only thing homogenous about Dells are the ugly, thumb-busting, nasty cases.
The ONLY reason to buy Dell is that they will replace any nonfunctioning part of your system, at their cost (shipping included), if you are patient enough to sit through 3 or 4 tech calls and go through their standardized rigamarole (sp?) for determining what component is bad (yeah, you may know that it is the video card, but the Dell techie has to convince himself as well).
I am NOT talking "out of my hat" (a phrase I am unfamiliar with, but which I assume means, making things up).
We have had:
* A Dell system which would shut itself off after 3 or 4 minutes, every single time it was powered on. After being patient with Dell tech support over the course of several weeks while they tried sending new memory, a new motherboard, and a new power supply (not all at once, but over the course of several weeks), they finally took back the whole system, after determining that it was their goofy power switch (I suspected this from the beginning). The good news is that they replaced a two year old Pentium 233 system with a brand new Pentium II 400 system at no cost to us.
* Dell's audio cards in their Dimension models of two years ago used this crummy, Dell-only version of the Montego sound card which the Turtle Beach people would not provide support for. Dell had cut a special deal with Turtle Beach to provide a cheaper version of their card that Dell would support, so that Dell could save a couple of bucks. The only drivers that Dell could provide for Windows NT would blue screen the system when you tried to install them.
* We have had many, many of the mice on Dell systems go bad. OK, they are Microsoft Mice, not really Dell's fault, but still. They could have chosen better mice.
* We have had two or three Dell monitors go bad. They were replaced without cost to us by Dell but it was a bit of a pain.
We have about 65 Dell systems where I work. Maybe the above problems are par for the course for PC's, but I would have a hard time believing that they are better than average.
Furthermore, Dell's cases DO suck. Trust me. Let one age for about a year and a half and then try to open it. You WILL bust your thumbs. Their case design is STUPID. I have worked with many cases. Screwless cases are not a big deal anymore. They are a dime a dozen. But 99% of them are designed MUCH better than Dell's cases. We have at least three or four systems on which the case no longer fits properly because of the force required to open them after they have frozen shut, and the fact that they are held in place by some rather soft plastic.
I have an Enlight case at home and it is by no means special or unique. It is FAR easier to work with than any Dell case I have ever used. No modern case is going to make it difficult to add hard drives. Trust me, I work on these things ALOT, I know a crappy case when I see one, and Dell's really suck.
I am NOT "spouting hackneyed Slashdot wisdom". I have quite a bit of experience supporting Dell systems. I know what I am talking about. Dell has its good points and bad points, but the original poster did not hit on the good points and spoke of some of Dell's bad points as being good.
Why would that be impossible? The Cray T3E managed to make use of that many Alpha processors, although they're using the EV5 and not the EV6 architecture, like the Athlon.
No doubt he meant 1000 processor SMP machine. The T3E is an MPP machine. A 1000 processor SMP machine would suffer way too much bus contention to be worthwhile, even as a 'because I can' hack. It would be hard to imagine an application that could benefit from such a set-up. Even harder to imagine one that wouldn't work as well or better on an MPP machine.
The Apple SMP systems seem fairly nice, but rather expensive.
It's not a realistic alternative unless I can specify the specs rather than living with whatever AAPL tells me I can buy.
I've idly watched the OpenPPC project; apparently the direct offspring, Pop Computers, found that they had severe procurement problems.
This does not a viable market make.
If you're not part of the solution, you're part of the precipitate.
Actually, he's been to a great deal of hardware sites, but he found very little solid information.
First, he went to AMD's web site. He browsed it thoroughly, and then did a number of searches. He discovered that AMD has essentially no information on their website about SMP, period. Almost no mention of it, even.
He then checked out Tom's hardware, Ars Technica, as well as 4 other similar hardware review and news sites. Do you know what he found there? A few rumors about AMD's supporting SMP, a single mention of 760MP chipset, without any dates or informaiton on it, and a couple of claims that AMD would have an SMP motherboard out sometime in 2001.
He wasn't impressed with most of the information out there.
You see, I wasn't looking for vague rumors, I was looking for facts. I was hoping that someone might have heard something from a reputable source, or someone might have a URL or two to a site that I haven't checked.
I'm glad that you were able to provide some insightful commentary here. Thank you for helping 'make slashdot what it has become'.
--
Toph
Topher
The hardware support for those G4 machines is sparse.
OS X doesn't offer a command line unless you buy server or developer versions for a lot more money.
Linux, I don't think is running on those machines.
OS X is slower than Linux on the same platform.
Yes, there IS a SMP alternative to Intel- and it's called Alpha. They're insanely great machines- just insanely expensive as well. G4 an alternative to Intel? Only when Motorola or IBM get off their duffs and sell SMP machines with the G4's processors.
I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
Intresting. I havn't heard about this issue. When my old (duel) PPro motherbord died, I bought a K7, and a K7 motherbord, and PC100 memory (ECC, because I'm that way), and an ATX mobo and PC Power And Cooling power supply (because it is quiet, and the same brand I've been buying for years, and was one of the few things that AMD did say there were compatability problems with). I took all the cards and drives from the old system and moved them to the new system.
They all worked. Including my generic NCR SCSI. Including my oddball ISA sound card. Including my cheap as crap ISA video card. Both of which I had bought for the PC I first built in 1992! I did get rid of the ISA sound card a few weeks later (FreeBSD funally stopped supporting it a decade after the makers went bankrupt, and I didn't feel like fixing the driver myself). I got rid of the graphics card too (so I could be ISA free -- it is my home server, and runs with no monitor most of the time, so the crappy graphics were just fine). Both were replaces with "store brand" PCI cards costing about $30 each.
Zero compatability problems.
Have you personally seen any compatability problems? Did the failed part work in another machine?
I hope so too. But I have little hope of Compaq doing it. After all they have the multi-way chipset (40+ CPUs) for the Alpha which would take little or no tweaking to work with the K7 "just" a BIOS, which shouldn't be hard for the first compony to reverse engener the IBM PC BIOS....
P.S. now that I think back on it, the ISA video ard may not have been from '92, it may have been newer then that. But the sound card was that old (ProAudio Spectrum bought at a Microprose employe discount not all that long before the compony that made 'em vanished, nice card though)
Slashdot has been really disappointing me lately. Besides the usual crap that is posted in the news, look at what Ask Slashdot has turned into. The dumbest questions get posted. Like "I want to find a job" thing yesterday. Now this. If Christopher Cashell ever bothered to go to any of the hardware sites, he would find the answer to his question in about 5 minutes. Let me guess, the next Ask Slashdot question will be "What is Linux?"
___
___
If you think big enough, you'll never have to do it.
In fact, they are SMP limited by the chipsets: If a chipset existed, you could run a box with 1000 Athlon processors - of course, designing such a beast would be impossible...
Why would that be impossible? The Cray T3E managed to make use of that many Alpha processors, although they're using the EV5 and not the EV6 architecture, like the Athlon.
I doubt that designing that beast would be desirable, though. There aren't that many companies who know how to build systems scaling up to 1000+ CPUs. Cray is one such company, and SGI seems to be another (allegedly building a 1024-CPU Origin 3000 on special order).
But impossible? No.
(This is, of course, blatantly ignoring that "chipset" is a very, uhm, simplified way of speaking about multi-CPU support.)
I e-mailed ASUS and one other Mobo maker asking this question. I got one reply, 'nothing in the works right now'. About two months ago.
Can it use the old celeron chips?
Muchas Gracias, Señor Edward Snowden !
Sorta. As I recall, Intel owns patents which cover the APIC (Advanced Programmable Interrupt Controller), and these patents are related to APIC programming in an SMP environment. The K5 and K6's used the OpenPIC standard to avoid this, but there were no OpenPIC boards and so effectively no SMP w/ K5 and K6.
The EV6-style SMP that Athlon uses avoids both of these issues by using an SMP model which has existing boards (the SMP Alpha boards) and which isn't covered by Intel's patents. (Of course, the Alpha boards can't be used directly for some reason, but at least they're closer than the non-existant boards for the K5s and K6s). I imagine the EV6-style SMP requires different OS support, though. That is, Linux, BeOS and WinNT would all need different SMP drivers for the SMP Athlon boards to replace the APIC code. (Basically, they avoid the Intel patent by not designing to Intel's MP spec, but that would imply that OSes need to have differen't MP drivers to support it.)
--Joe--
Program Intellivision!
Any limitations on the number of CPUs is going to be in the chipset either in the number of pins (about 100/cpu) it can support, or in internal queuing stuff in the chipset keeping track of outstanding cache probes - the bus protocol itself is refreshingly free of such stuff.
Are you sure about that? Nothing like each chip being assigned a number out of a small and fixed pool of chip IDs?
The number 14 looks very suspicious to me in this regard (2^4 - 2).
OTOH, I haven't read detailed specs on the EV6 bus, and Compaq has recently started offering SMP systems with maximum processor counts greater than 14.
You can currently do that with intel boards. A friend of mine tried it on a BP6 using a 500 Mhz celeron and a 400 Mhz celeron. As for OS support to intelligently schedule jobs that take more time it could be done. As far as I know no OS has logic for differently clocked chips. It is something that could be added but would it be worth the trouble. How many people are really gonna buy several different speeds for a box.
Exactly, its the motherboards. I wonder if they are running into trouble? I hope not, because I have been specifically holding out for a dual AMD system. It sure feels like a long wait...
Try to hack my 31337 firewall!
I participated in the Nextech 2000 conference in Austin Texas. One part of the conference was taking a trip to AMD where we got to ask questions of the various interns there. One of the people there was working on getting the glitches out of the motherboards with multi-processors. If I remember correctly, he said that they were very close, but there were still a couple of issues that had to be resolved. Unfortunately, they weren't allowed to tell us much more than that. The trip was the end of July, just in case you wanted to know.
I don't remember the source offhand, but I remember reading that SMP Socket A (Tbird/Duron) mobos would be out around January.
"If ignorance is bliss, may I never be happy.
-- Veni, vidi, dormivi
> From what I understand, the 760MP should be finished between December and January, and on store shelves late Q1 2000.
Can you build SMP systems incrementally? I.e., buy a 760 motherboard and a single CPU for starters, and then add a second CPU as an upgrade a few months later?
--
Sheesh, evil *and* a jerk. -- Jade
The BX chipset gives you 2 CPU support 'for free' -- we'll see if the AMD chipset does the same, or if you need to buy a special SMP chipset.
;-)
Seems unlikely. The electrical interface that AMD uses (which is the same one DEC's Alpha uses) is not, in fact, a bus architecture, but rather a port. Each CPU has dedicated lines to the a "hub" chip. So you would need significantly more silicon to make a dual-processor chipset, as compared to a single-processor chipset. It seems generally agreed that this design is superior to the Intel bus design (it scales to 1000s of processors, compared to Intel's eight), but it does make thing a bit trickier on the low-end.
SMP Systems have a huge margin advantage over single CPU systems.
Very true. Why do SMP systems cost so much more? Because the people using them are willing to pay so much more.
With SMP you can start using the magic words "server" and "workstation" which translates into higher profits for the resellers.
"Server". There, I just used the magic word on a single-processor system!
dragonhawk@iname.microsoft.com
I do not like Microsoft. Remove them from my email address.
Well, I do the buying for our company and yes, I am one of you guys (techie, not PHB); I'm senior mgmt and run Linux on my desk :-) We are a startup and $$$ are important.
My policy is to buy Dell returns from DFO for desktops (they are new, just no warranty) and yes, they all come with Intel CPU's (Dell's choice not mine). We are now getting P3-733's with 133 frontside as "used" Dells. They are single CPU boxes. The reasons for my Dell-only are quality, homogeneity, ease of maintenance.
The cost difference between AMD and Intel is not enough to materially impact this policy, as I'd need to go with a different vendor; most business class machines (most of my desktops are for developers) only come with Intel, Compaq's AMD boxes are all low end.
All our servers (production and office) are low end SMP capable, and are either USparc-III (Sun Ex50's) or 2-way P3 boxes (Penguin, Dell, IBM).
Where I'd really kill for a good AMD is in the laptop arena - we use midrange laptops, mostly for business people (sales and marketing) who don't really need performance. In keeping with our frugal policy, we don't get monitors but we do make sure laptops have decent displays (1024x768 TFT) and this means we are forced to buy Celerons, which suck.
We have one real el-cheapo laptop, a bottom of the range $1000 Toshiba which we use as a test console for the server rooms. It was mine before we got venture capital. It has an AMD, K6-3 I think. The screen sucks (800x600 DS) but it kicks a $2000 Celeron laptop for performance, and the power consumption is very modest.
If I could get a decent midrange laptop with a good screen and an AMD cpu, I'd buy one in a heartbeat; on both performance and battery life, AMD kicks butt in the mobile market.
Doesn't it only support 8? Or did they change the constant in R5?
A deep unwavering belief is a sure sign you're missing something...
Uh no. SSE is uses 128bit registers, 3DNow! uses 64bit registers. SSE contains more instructions that does 3DNow! Those two facts alone mean they cannot be clones.
A deep unwavering belief is a sure sign you're missing something...
But AMD's CPUs AREN'T clones of Intel. And that analogy is still wrong because while AMD and Intel CPUs are instruction set compatible, SSE and 3DNow! aren't. Lastly, you can't say SSE is a "hacktogether" response because I'm pretty sure they were working on it before 3DNow! was released. (What, you think you can put together an instruction set in a couple of months? KNI (Katmai New Instructions) were in planning a while before 3DNow! came out.) Also, SSE is in many ways superior to 3DNow!. It has more developer support, it has a wider range of instructions, and when fully implemented (which it is not in the PIII series) it is twice as fast (due to the whole 128 vs. 64 bit thing.)
A deep unwavering belief is a sure sign you're missing something...
Intel moves to .13 in 8 months? I would like to see that..
.18->.13 die shrink of the P3. Northwood, the .13 P4, isn't due until Q4. (It'll be a lot easier for them to test and get the kinks out of the shrink with an old core that they understand very well than with a new.) Remember, they're already working on the shrink now.
Here's the latest roadmap, in Japanese pastel no less.
8 months at the earliest--i.e. Q3 of next year. And that's for Tualatin, their
The real story about Intel that their Pentium 4 is being manufactured here in Israel (in Kiryat Gat), and they got LOTS of problems with that (like low numbers of chips on 1 waffer, only 2 machines to product the P4), so until Intel gets more machines to produce those chips - this will take long time.
This story was posted here in the Israeli newspapers..
Well if it was a repeat of this piece at The Register, it may have misinterpreted things. For one thing, 70% yield on a new chip is quite impressive. For another, while a yield of 70,000 chips/week is not enough to substantially affect the x86 market, 70,000 represents more CPUs made in a week (and before launch no less) than there were 1 GHz P3s made for the first 6 months after its "launch"!
In any case, no OEM is going to buy many of the initial P4s, because Intel is planning a packaging change in March or so. That means new motherboards, new systems, and another entire validation process for OEM's--something that most of them are not going to want to waste time doing twice. Thus they'll only offer a couple models, in limited quantities, of the original P4, so it doesn't matter that only limited quantities will be fabbed.
In essence the first P4 will be positioned quite a bit like the original PPro. Whether the Willamette core ends up as successful as the P6 core has (remember, the PPro introduced the core now powering Celeron and P3) will be interesting to see.
Who else is waiting for the PIII 1337?
No, AMD licensed EV6 to get around the Intel Slot1 licensing, not the SMP licensing.
General Relativity: Space-time tells matter where to go; Matter tells space-time what shape to be.
I think the reason people get the impression that AMD chips are not top of the line, is because of the poor compatability with other components. I've had a lot of problems getting AMD chips to play nice with my other cards. For a server, I don't see this being an issue, as you can always find *one* SCSI card, video card, to work just fine with the chipset. Hopefully DELL, Compaq and and company will release "Interprise Level" servers featuring 8 way AMD chipsets. Doing so would add a lot to the credability in the mindset of the average buisnessman. -Jon
this is my sig.
The company that AMD had licensed the Athlon bus to was HotRail, formerly Poseidon Technology, decided in June that it was no longer profitable to remain in the PC chipset business, and put their 8-way chipset on the back burner, in order to focus all their strenghs on Networking switches and transceivers.w s/story/OEG20000501S0051">Here</a& gt; is an article about their switch of plans
<a href="http://www.ebnonline.com/ecomponents/commne
Stop over-analyzing your analizations
AMD is working on the 760 chipset which is the DDR chipset, later this year or early next there will be another revison (760MP) which will support SMP, all current Durons and Thunderbirds are SMP capable.
How do you think that SSE & SSE2 will be implimented in upcoming AMD CPUs starting with the mustang.
IIRC, KNI/SSE is just a clone of 3DNow! which debuted in the AMD K6 line.
<O
( \
XPlay Tetris On Drugs!
Will I retire or break 10K?
With all other variables equal (bus speed, HD speed, etc.), a 500 MHz PowerPC G3 is approximately twice as fast as a 1 GHz Intel PII/P!!! at doing hard-core integer number crunching (Photoshop, some 3D games in software mode, d.net) according to the SPEC integer math benchmark. And the vector unit in G4 is quite a bit easier to code for than the vector unit in PIII, giving even more speed and boosting Team Slashdot's RC5 rating.
<O
( \
XPlay Tetris On Drugs!
Will I retire or break 10K?
Intel holds the patent for APIC, the smp structure. AMD will need to use openAPIC or their own standard before they can release. This means the kernel may need some work before supporting it.
Eat right. Stay fit. Die anyway.
You killed the first post kiddies! You bastard!
This has got to be the first time in recent slashdot history that a "post #1" has gotten any moderation other than down...
I think this deserves a toast, a slashdot story, and a redundent story, just to be sure!
I would think that it would not currently be profitable for the mobo makers to get into the dual-processor game, with the technology moving so fast and the next generation of chips right around the corner.
is to say I have a "Duron-Duron" setup....
One of the problems with 21x64 and thus Athlon chipsets is that the busses are designed to handle high front side bus speeds. To do this, the bus is a point to point bus with the processor connecting directly to the chipset. This translates to more wires and pins on SMP chipsets. The issue is further complicated by caching issues. Alphas use an L3 cache which the chipset controls to share data amoungst SMP nodes. This increases the compexity of SMP chipsets for the architecture. Other people have also mentions, and correctly from what I have heard, that AMD had no desire to produce chipsets, and this has probably been a reason they where slower to ramp up additional higher end chipsets. VIA might very well do similar sets, but I imagine AMD does not want to place all its eggs in that basket.
One of the side-effects of that merger is that AMD and Intel have a very comprehensive patent cross-licensing deal, which was "inherited" from National Semiconductor. This is why Intel has not attempted to sue AMD for patent infringement since the days of the K5 cpu. It's also the reason that Intel forced the "Slot 1" bus architecture down the throats of most of the mobo companies a few years back: it was "new" and thus not covered by the agreement. Unfortunatly, it also turned out to be both a political and an engineering mistake, leading to the current chilly relationship between Chipzilla and most of the taiwanese mobo companies...
News for Nerds. Stuff that Matters? Like hell.
Hey all, I posted a story that got declined by the forces that be at /. but "The Register" a uk web rag
reported last week or the one before that the 760
chipset is going to be out 1st Quarter 2001 in SMP fashion for T-Bird and Duron chips.
The Register
Its alos the home of the BOFH stories for those that are interested.
From there I think it likely that AMD will produce 4, 8, 16, etc chips-per-die instead of plugging them into a motherboard. Although I guess it depends when they run into the bandwidth bottleneck.
Suppose that's when they'll start pumping that EV6 DDR bus up to 400mhz, etc.
Drooling profusely here.
All Socket A Athlons and Durons are multiprocessing enabled. They only require a new chipset to do so. AMD is developing the 760MP for this purpose. It will also support DDR SDRAM. This topic will be discussed in much detail at the Microprocessor Forum next week. VIA may or may not develope a multiprocessor capable chipset for AMD processors. Hotrail was developing a chipset that would support 4 and more processors, but they dropped the project. I'll be covering all news of the 760MP next week at AMDZone, and expect Tuesday to be the big news day if you are interested in the 760MP. There should be a load of new information, and possibly a press release or two.
ignorance is bliss. googlefiberatx.com
According to the hardware spec of the Thunderbird and the Duron they support SMP. As for the older Athlons (with the slower cache) they do (from what I understand) NOT support SMP! So be careful, and don't buy an old one.
:-)
Both AMD and VIA are planning SMP chipsets. Check out http://www.viahardware.com/roadmap2000.shtm for VIA's roadmap. There should be a chipset out just in time for Christmas if we're lucky!
I remember reading an article about a test of a Dual Thunderbird board, but I can't remember where... It was a beta board with a very early release of VIA's chipset anyway. They didn't actually test the SMP, they guys were more interested in the DDR support...
Cheers,
There have been LOTS of problems with AMD chips. Actually, more precisely, the problems have mostly been with the chipsets, which AMD themselves didn't want to make -- they were really hoping the third-party market would do that for them, but nobody stepped up to bat with a really good chipset for the Athlon until very recently.
Quite some time before the Athlon came out, I myself bought a K6-2/300 (which I still have, happily chugging away as a Linux server) and had all kinds of hassles with the Aladdin V chipset and various and sundry cards. I eventually ended up buying a BX-chipset board instead, and was much much happier with it. Even now Linux doesn't run as well as it should, as it doesn't seem to have any support for that IDE chipset and leaves all the drives in non-DMA mode. It doesn't do much work so that's okay for me, but it's a bit annoying.
With Athlon machines, the biggest problem was simply inadequate power supplies. Those chips suck power like nothing else before them (I think the new P4 will suck even more!) and if you put a nice fast GeForce DDR (another power hog) in there, many motherboards and/or power supplies were simply overwhelmed by the demand. Your purchase of that PCPAC power supply was probably the best money you spent in that machine, and may have saved you lots of trouble.
There have also been AGP driver issues with some of the Athlon chipsets, though I haven't yet owned one and don't know the details. I CAN tell you that the compatibility problems have been severe enough that I held off buying an Athlon. It sounds like the KT133 chipset has it pretty well together, finally, but I will probably hold out a bit longer and go SMP when those ship. I haven't done an SMP machine yet for myself, though I have wanted to for a long time.
Oh, another thought: it sounds like you had pretty good luck with your system, but remember that you weren't running fast 3-d graphics and/or Win9X either -- video drivers have been especially problematic. You were running against 'old' standards that are very well documented and easily testable. A lot of people are buying Athlons to game with, and gaming taxes a system harder than almost anything else you can do with it. AGP appears to be something of an evolving standard, too, so there are all sorts of niggling little details that differ from chipset to chipset, and can cause weird behavior that you would never see on a BSD-based server.
My $0.02.
Tyan is almost done with their SMP Athlon board. It should be released soon.
-- Ted tsikora@powerusersbbs.com
Should be possible with Athlon, too. You can leave CPU sockets without CPU in Intel SMP configurations and the remaining CPUs are used properly. However, you have to put in termination dummies in order to not degrade the CPU bus signal quality. Since AMD uses the Alpha way for SMP, it won't have a CPU bus but separate ports on the chipset for every CPU, so you may not need terminators for those.
In fact, they are SMP limited by the chipsets: If a chipset existed, you could run a box with 1000 Athlon processors
Actually, my understanding was that limitations in the bus protocol limited SMP machines to 14 processors, as with Alpha machines. For more than that, you use a hierarchical scheme or clustering.
It is not the CPU but the Motherboard that can't support SMP. Maybe the MB manufacturers are waiting on a reference board or chipset from AMD? Does anyone know if the Irongate chipset is the holdback or is it simply the Motherboard configuration? (I have a feeling that the AMD motherboard chipset is not ready)
--------------------
Kyle at HardOCP posted up a link to an interesting synopsis of an AMD seminar. AMD confirmed that VIA would be supplying an SMP chipset which will be able to run two Socket A CPUs.
These should be available this fall. As we all know; VIA has no problem introducing a new chipset to the motherboard manufacturers and no doubt will have no problem getting boards made with their chipset. As many of us know, the same Irongate chipset that many of us use for Slot A Athlons was perfectly capable of running the Socket A CPUs as well and was always available to motherboard manufacturers. Despite this, manufacturers still opted to use a VIA chipset instead and delay availability of their boards because of this.
Plan to see SMP as production exceeds demand or at least keeps abreast of it. Also it has been widely reported that AMD is concentrating on introducing the "Sledgehammer" with an SMP MB chipset almost from introduction.
"Science is about ego as much as it is about discovery and truth " - I said it, so sue me.
The boot 'ROM' code is probably in Flash, so could be blasted with x86 code. The peripherals which have boot code have x86 code already, as the Alpha supports x86 emulation even at the lowest levels. FatPhil
Also FatPhil on SoylentNews, id 863
Can someone who has knowledge about operating systems, especially process cheduling, comment that a bit? Can users be sure that the most CPU hungry thread/process is run on the fastest CPU?
--
By the way: AMD has no plans to cripple it's Duron processors.
p ost_id=15000265 - all Durons will be SMP capable, and, AFAIK, they should take full advantage of the 133MHz / PC266 DDR motherboards.
http://www.aceshardware.com/Spades/read_news.php?
The Alpha motherboard is close, but probably won't win the cigar.
Motherboards include ROMs, for things like setup, BIOS and booting up in the first place. Alpha uses a different instruction set, so a 386-instruction chip like Athlon wouldn't read it.
The mobo vendor could probably do an AMD port of the board. API is unlikely to since their job is to sell Alpha processors, not mobos. Somebody else with an Alpha SMP mobo would be a better candidate.
The reason SMP is still unavailable for the Athlon line of CPUs is that the current chipsets have no way of supporting it.
The first Chipsets capable of supporting SMP, AMD's 760MP(a SMP-enabled version of the 760, due out very late this year or early next year) and the 770 chipset, which is expected to have support of up to 4 CPUs, and due out early next year.
As far as VIA and otehr third-party chipset manufacturers, they are still awaitin a chipset from AMD, before they can begin making their own SMP chipsets.
AMD's plans for a more advanced (4 and 8-way)chipset also had to be canceled when the company they were working with (forgot the name at the moment) decided to leave the server chipset business, leaving AMD to work on the chipsets on its own.
Stop over-analyzing your analizations
You are correct in regards to the processor's support for SMP. The current crop of Athlons (including the Athlon classic) are SMP capable.
p ost_id=10000214
In fact, they are SMP limited by the chipsets: If a chipset existed, you could run a box with 1000 Athlon processors - of course, designing such a beast would be impossible...
At any rate, Ace's Hardware has been covering AMD's products fairly diligently. They've posted several articles about the 760MP (The SMP capable Athlon chipset.) One good example is available here: http://www.aceshardware.com/Spades/read_news.php?
From what I understand, the 760MP should be finished between December and January, and on store shelves late Q1 2000.
From what I read awhile back, the next VIA chipset will have SMP for Thunderbird AND Duron. It will also have DDR support.
I bet that not long after this chipset hits that companies like Abit have dual socket A motherboards, and for an attractive price. Especially considering that 700 MHz Durons, which are in some respects (FPU) better than a 700 Mhz P3, can be had for under $90...
It sucks how slow SMP for Athlon has been coming, but I think when it does hit, Intel will lose a lot of face. The "Mustang" core Athlon is supposed to be the next stage, a chip that competes with Intel's way WAY overpriced Xeon line.
In 2000 America, is a non-lawyer truly free?
As others have said, the first SMP chipset for the K7 family is the 760MP, due out in December or January. AMD has already released the single-processor 760 chipset--which looks to be the first DDR-capable PC chipset--to motherboard manufacturers; motherboards based on the 760 should be showing up at the end of this month.
;) the market for SMP boxes is primarily business servers. This presents two issues:
.13um process--in about 8 months. The 512kb, 1Mb and 2Mb L2 cache Xeons won't move above 800MHz or so in that time. Meanwhile, the new P4 chips are *not* SMP capable (or at least there will be no SMP chipsets available for them). Itanium is a joke and will likely never be launched. Now, Foster, the "P4 Xeon" will be released, possibly as soon as January, but the large-cache versions of Foster won't be out until Q2.
It has been pointed out that one of the major reasons AMD has taken so long to get SMP going is that they already sell all of their processors anyways, and already have commitments for all the K7 chips they can make through the end of the year. This is true, but misses a more important point: with some few exceptions (i.e. nerds like us
1) Many if not most Intel SMP boxes are currently stuffed with Xeons. Thus, AMD has been perfectly content waiting for the release of their upcoming "Mustang" core tweak, featuring up to 1MB L2 cache--due out...you guessed it...in December or January--before rolling out the 760MP. Conversely, the big-L2 Mustangs without an SMP-capable chipset are dead-in-the-water.
2) By and large, business still clings to the notion of AMD as a cut-rate unreliable chip company. Despite the fact that knowledgable consumers have switched over to the cheaper faster Athlon in droves, Intel still has a nearly complete monopoly over the x86 business market. This impression of AMD as the "cheapo generic brand" persists despite Intel racking up delay after delay, errata after errata, recall after recall, embarrassment after embarrassment (i820, i840, 1.13GHz P3, Itanium) in the past year and a half. AMD knows that if they release the 760MP and it runs into one rumor of one conflict with one obscure 3D graphics card no business machine would ever contain anyways, their foray into the high-margin world of business computing is over before it began. (Never mind that Intel can release the i840, their new workstation-quality top-of-the-line chipset, with an "errata" which rendered it unusable with ECC memory!) Thus they are being very careful, and rightly so, with their validation process on this one.
Interestingly enough, if they get their act together (and purchasing departments take their heads out of their asses), AMD has a major market opportunity on their hands here. The Coppermine Xeon (i.e. a plain-old Coppermine P3 with $200 tacked on to the price) is incapable of scaling past 1GHz until Intel moves to a
That leaves AMD with a quarter as the sole supplier of GHz+ large-cache multiprocessing x86 CPUs. Will that be enough to get them into the lucrative enterprise market?? Hard to say. After all, you never get fired for buying Intel...
I have read on internet (www.aceshardware.com I think) that you can use existing alpha mb to put K7 into it; word is the K7 uses the Slot A (DEC Alpha) interface and so is supposed to support alpha style SMP. So the K7 should have nice SMP, scable up to 32 processors.
If you wanted just dual processors, this implies A UP2000 or suchlike would do the job, but I can't say for sure.
Here's a qoute from Paul Jakma
Interestingly the new *Alpha* 21264 UP1000 motherboard uses the AMD Irongate chipset.. they also have a dual 21264 UP2000 board based on a DEC chipset. So it seems K7/Alpha chipsets are interchangeable, so then K7 SMP is probably possible using the DEC chipset.
And here's a qoute from Acehardware.com:
Alpha:Slot-A:Slot-B:Athlon KH Yeap Wednesday,
June 23, 1999 (10:00 AM EST)
Ok, at the ongoing PC Expo in New York, Alpha
Processor Inc. is demo'ing its new
Alpha-21264 750 Mhz, which is expected to
come out in July. More interestingly a 1 Ghz
versions of the processor, which runs under
regular air-cool condition, is also demo'ed
along with a Slot-A motherboard, UP1000, and
a Slot-B motherboard, UP2000. For further
details check out this News.com report.
Now, a lot of people have been wondering
about the possibility of running a K7 on an
Alpha Slot-A or Slot-B motherboard. According
to Alpha, yes, this is possible. To make
things even more interesting Alpha's new
Slot-A motherboard, UP1000, uses a chipset
that is a hybrid between AMD's very own K7
chipset, Irongate (AMD-751), and ALI's
M1543C!! PC Watch Japan has a great shot of
this UP1000 motherboard. Also appears on PC
Watch is a photo of the Slot-B UP2000 and a
photo of the 1 Ghz Slot-B Alpha processor.
Special thanks to Daiki for this wonderful
tips.
So, you could try an Alpha dual-processor Motherboard but I can't give you any garuntees.
Michael
...another comment from Michael Tandy.
"Goodness me, how unlike the FBI to abuse the trust of the American public." -- The Onion