Slashdot Mirror
More Juicy Dual-Processor Goodness
ronmon writes: "I was cruising around bp6.com (I still want one) and happened to see a link to some pics of dual Socket A motherboards. It's in Japanese, so I can't read most of it, but this particular board caught my eye. It's a SuperMicro sporting two sockets and five DIMM slots, plus four drive connectors (IDE RAID?). Yummy!"
And credulous reader Jim writes: "This one gave me a start when I woke up this morning. 2CPU.com has a screenshot of sisoft sandra scores from a Dual 1.53GHz (11.5x133)T-bird box. Apparently from an anonymous email. The scores are nothing short of amazing. Check it out." Grain of salt, remember.
haha
And the 4 IDE connectors are almost certainly an embedded UDMA66/100 controller (like ABit puts on a number of their boards) and not IDE Raid.
trying to get the trolls to leave? ha! we're not that gullible.
weenie.
Are you sure? I don't know about that particular board, but I have an Abit KA7-100, featuring the Highpoint HPT 370 which does indeed do hardware RAID. So there definitely are a few chipsets out there that support it...
Wild Boards!
Wow! With one of those boards, a pair of 1.2GHz Tbirds and an NV20 board I might get decent frame rates in Unreal! Woo-hoo!
-Rat
Boobies never hurt anyone. - Sherry Glaser.
I've had good luck with the Abit boards I've been using. A PIII-500 coppermine in a Abit BE6-2 and an AMD Duron 800 in a Abit KT7. Since I haven't really bought a different kinds of motherboard in 4 years I must wonder, are people out there still configuring your motherboards with jumpers to set the clock rate? What a pain in the ass that was. I can't imagine having to go back to a motherboard where I'd have to pop off the case and play with jumpers to just change the CPU clock or bus clock speed.
PCI is _twice_ as fast as ISA? Really? I think it's a little faster than that...
If you're talking PCI as it is in a typical PC config (33 MHz bus clock, 32-bit bus width), then it's at least 8x as fast (with most ISA cards using a 16-bit-wide bus, with an 8 MHz bus clock). 32x faster if you happen to be using a system with a 64-bit, 66 MHz clock PCI bus.
Also, there are a few "real" (non-soft) PCI modems out there. Hardware modems, whether ISA or PCI, will cost more than software modems.
_____
Sam: "That was needlessly cryptic."
Max: "I'd be peeing my pants if I wore any!"
two 1.5 gighz procs are aprox 50% faster then 2 1.ghz procs... wow imagine that...
*sigh*
I will wait till the bugs are worked out.
I have allready been bitten by bad dual cpu motherboards/chipset (via). Can't use the spiffy new geforce 2gts on this tyan tiger mb reliably.
maybe I should put it on ebay...
Forget make clean did we?
--
Program Intellivision!
Uhm, news for you: PCI is much more than 2x as fast as ISA. ISA is 16-bits wide and 8.33MHz. PCI is at least 33MHz and 32-bits wide. That's rougly 8x faster. And then there's the 64-bit variant, etc...
--Joe--
Program Intellivision!
More likely, 2xATA-33 and 2xATA-100.
--
--
Don't like it? Respond with words, not karma.
CPU Integer 3656.756 MIPS CPU Floating Point 1496.255 MFLOPS The above numbers are for a T-bird 1200 133 FSB from wintune (basically dhrystone and whetstone.) The 256KB on chip cache makes a significant difference Even if you normally expect only 1.5x real-world performance on a dual system, remember these are synthetic benchmarks, and, based on my experience with other dual systems, the numbers are usually much closer to 2x than 1.5x on these types of tests. Based on all of this, I believe the numbers - they are within the range I would expect for this configuration. To echo what the orignal poster said, these things *are* truely amazing - that's the only word that seems right :-)
Umm... go look again, please... It's definitely labelled in the article as a TYAN motherboard. Which would be the same board that an earlier Slashdot article covered.
Of course... further down the page, there is another picture of a similar looking dual PIII board which is the SuperMicro board.
(It doesn't hurt to double check facts every now and then.. -_-;;)
- Wing
- Reap the fires of the soul.
- Harvest the passion of life.
- Wing
- Reap the fires of the soul.
- Harvest the passion of life.
You might not have personal experience with SuperMicro dual-processor boards, but I do.
A have a SuperMicro P6DNE dual PPro board that's been up since early 1997. It happily runs WinNT 4.0 sp6a supporting ~50 users at a factory in Connecticut. WinNT goes 3-4 months at a time without a reboot, and I believe the motherboard contributes to this.
Some might remember that another SuperMicro dual board was used by Andy Grove to demonstrate the Pentium Pro chip on a tour sometime in 1997.
I think their board designs are good, and I'm sure their manufacturing is as good as anyone on the market. Maybe you should reserve your judgement for products you have personal experience with?
-jb
WRT the '142% improvement' benchmarks floating around, there are several issues. First, they used make clean, not make mrproper between compiles. make clean leaves some stuff lying around (and consequently speeds up the compile). Second, they used -j3 on the dual compile but not -j2 on the regular one. -j2 can speed up compiles on uniprocessor systems not insignificantly. But, even with that, I can't imagine the difference getting too far away from 100% improvement. Not that that's suprising...compiles are 'embarassingly parallel.'
Um perfect for a 1U case? How exactly do you fit FIVE pci cards in a 1U case? Vertically none the less.
I'm running mine with drives on the ATA33 and ATA66 side, not a problem, run Linux, W2k, BeOS, QNX without a hitch.
You may have had one of the bad ones as others have pointed out to you.
Overall I'm very satisfied with the board, but as the manual says, the board is a beta product anyways and not meant for regular users to be using.
Whoa there captain obvious!
Pretty sure the fuss here is over the dual socket a -- the kt7-raid is one of many boards to have the hpt370 controller integrated.
...that these boards are all legacy-free? Start chucking those isa cards now, PCI isn't available from the big boys.
I agree with what you say, but let me explain why I put that remark there: Most of the messages in this story were from people drooling over MIPS ratings and saying "I want one, too!". So I thought somebody asking a question like mine would be considered a troll (wouldn't have been the first time), and nobody would have answered. This is also the first time I posted at +2, because I wanted people to see my question and maybe answer it, even that late in the discussion.
I think you are right in criticizing the moderation system, but the question is: how else could you do that? How do you get rid of the junk and do that fairly?
EagerEyes.org: Visualization and Visual Communication
Athlon boards will require as many as 8 layers to fit into an ATX form factor. IIRC, the industry standard is 4 or 6, so this would be a new (and likely expensive) manufacturing process that may require new tooling to produce in bulk.
The Asus A7M266 (Socket A, 761 northbridge, 686B southbridge, working PC2100 DDR support) is an 8-layer board. It's available now, though in short supply. Lowest price on Pricewatch is $185.
Dual-CPU boards always cost more than equivalent single-boards, but the delta doesn't have to be huge. If we assume that the good 761 boards will use 8 layers anyway, then the incremental variable cost of MP is whatever extra AMD charges for the 762 versus the 761, plus the extra CPU socket.
I predict that the first dual Socket A board (Tyan's?) will be ridiculously overpriced ($499?) due to lack of competition and will be aimed straight at the less price-conscious server market. But once competition kicks in, we should see several 760MP boards in the $200 range.
-- David Ripton
Could somebody well-versed in Japanese please put up some kind of translation, into something that the fish can handle (i.e. french, german, english) so us mere mortals can look at this stuff too? Lotsa ppl would appreciate it...
"Caffeine is not an option. Caffeine is a way of life."
Good to see someone with a brain for hardware. As far as I've heard, Tyan will be the only one with dual AMD boards for a while.
How about a link to a review of the Abit VP6?
I certainly HOPE it's intended to be humorous. :) If not someone is seriously misinformed.
Get off my virtual lawn, you damned virtual kids!
PCI is newer than ISA and twice as fast. PCI involves a 33 MHz bus, as opposed to ISA's 16 MHz bus. ISA modems cost more because they tend to handle the actual signal processing on the modem card. That is the only case where I can think of an ISA card costing more than a PCI card.
There are no ISA slots on many newer motherboards (including my Asus A7V) because ISA is over ten years old. Even the newer PCI is scheduled for an overhaul in the form of PCI-X, due out this year.
Get off my virtual lawn, you damned virtual kids!
As I recall Itanium is 64 bit, has yet to see commercial release, and has a hard enough time running as a single processor. Athlon/Thunderbird is 32 bit, is over twenty months in widespread commercial release, and competes against P-III/P-4, not Itanium.
From what I have seen, running standard 32 bit software the Itanium should not be compared to any current 32 bit chip -- it gets beaten into the ground by anything faster than a 486.
On the other hand an Itanium running native 64 bit code should outperform any current 32 bit chip in FP operations, which makes it ideally suited for FP-intensive tasks where the Pentium III/4/Thunderbird cores will never dare tread ...
In terms of performance, in terms of architecture, in terms of intended market, comparing an Athlon to an Itanium is silly.
Get off my virtual lawn, you damned virtual kids!
What you say makes sense assuming the software running on the processors is capable of harnessing that feature.
What I still do not understand is how these two processors could function at greater than two times combined individual performance if the bus linking them is a fraction of their core speed?
What I am wondering is if the techniques with which you are familiar also apply to x86 architecture. It is usual to see something more like a 1.5:1 improvement overall in Pentium-based MP machines. From what I have seen of benchmarks that's the norm.
The Athlon IS a different architecture, and EV6 IS a different bus. I suspect that if putting two Athlons together MORE than doubles their speed that will make considerable waves in x86 computing.
On the other hand as has been pointed out elsewhere the posted Sandra scores are based on a motherboard with virtually no features that would allow an overclock from 1.2 GHz to 1.53 GHz. In addition these types of unsubstantiated "announcements" are the norm on many smaller tech sites looking to boost their readership.
If you could answer my questions or point me towards resources where I could do a little research it would be appreciated. Thanks for your comment.
-JoeGee
Get off my virtual lawn, you damned virtual kids!
True, I was just going by bus speed in MHz.
You're right -- my bad.
Get off my virtual lawn, you damned virtual kids!
That's true. The "fair" test results would have been interesting to see. I did not notice the -j3 compile flag, just gave the results a cursory glance and thought "yah, right ..."
I still cannot understand a > 2 times increase in speed as aminorex asserts, but I'll take him on his word. I assume with different bus architecture such a thing might be possible, and there IS a whole world out there above and beyond x86.
I have a hard enough time keeping my facts straight with x86 without worrying about Compaq, Sun, HP/IA 64, and IBM.
Get off my virtual lawn, you damned virtual kids!
Intel SMP systems gain 70% for the second processor in these sort of tests (this was taken directly from the discussion at Aces's Hardware when the linux compile results came in). EV6 is a much better SMP plaform than Intel's shared bus system so it should score higher than Intel's 70%, in fact in the same conversation we decided that 90% would probably be close to correct for these boards (due to such things as DDR Sdram, large caches, etc)...
Now real world tests (aka executing actual apps) will probably only gain 60-70% (against 50% for an intel machien real world result). I've seen a couple dual alpha boxen & they seemed to get results similiar to this (can't bench with sisandra of course though, so I could see synthetic benches on it even if I was allowed to run them)...
we are all invisible unless we choose otherwise
I was at a career fair Friday looking at Solectron's booth. They had a board they made for Cisco, about 50% larger area than an ATX motherboard. I asked him how many layers it had, and he said '16'. I was a little aghast, but I picked it up and looked, and he was right. Also had column array packages with *thousands* of pins each. So this making, say, an 8 layer Athlon board isn't unreasonable.
Dear articulate coward:
2 00 10124/17.jpg
I was not on the wrong article. Had you been able to fully read the article, you would have noticed the image referred to (in the slashdot article) at:
http://www.watch.impress.co.jp/pc/docs/article/
Although I was wrong about which slot was in backwards: the first 5 PCI slots are keyed in the wrong direction.
You should check your information before considering yourself so much better than everyone else.
When I die, please cast my ashes upon Bill Gates -- for once, make him clean up after me!
Note the key on the PCI slot closest to the edge: the card would have to face backwards. The Tyan board also had 64bit PCI slots -- those look like 32bit connectors -- not server class hardware.
The board looks less densly populated than the Tyan. Either it's a cleaner design -- or maybe this is just a mock-up (so it's okay the slot's welded on backwards).
When I die, please cast my ashes upon Bill Gates -- for once, make him clean up after me!
Actually, the RAID included in the Abit boards and almost all other IDE RAID boards are BIOS-based solutions..
they are still softare, just at a lower level (BIOS vs. Driver) but as far as true hardware raid, only one company makes a microcontroller-based IDE RAID card, which, at the price of high end adaptec raid cards, isnt too lovely an idea...
Stop over-analyzing your analizations
mod this funny Jane Fonda loving bastard up!!!
XML causes global warming.
Almost nothing worth buying is ISA anymore (I'll ignore sound cards and modems in that statement). Servers don't need those. And a server can get an external modem if need be.
PCI is much faster, especially when you get into 64 bit and 66MHz PCI busses.
ISA is a leftover technology now, with slots there for backwards compatability. Similar to what Apple did with getting rid of the floppy drive (except it's easier to attach a USB -> floppy converter than an PCI -> ISA device if such a beast even exists). For a server motherboard in 2001, I can't see why you'd want an ISA slot. I'd rather get an extra PCI space and toss the ISA stuff.
Yeah. I ran "Sisoft Sandra" on that dual Intel P4 @2GHZ, but I haven't yet taken the time to anonymously e-mail the screenshot to 2cpu.com .
Also, I developed a version of lnxoskrn.exe that will successfully run in place of ntoskrnl.exe for either NT 4 or 2000. Unfortunately, I work on that alone 50 feet below the earth in an abandoned Atlas Missile Silho.
Oh, also I scraped the cure for cancer off a rusted "tin" container of Juicy Juice.
The List of Grievances with Slashdot.
That benchmark looks too good to me. I'm not sure I buy it. The Athlon 1GHz gets 2792 for its ALU. The dual T-Bird 1.53GHz processors get 8576. 2792 * 1.53 * 2 = 8543.52. Now the 1GHz I believe is an original Athlon and not a T-Bird. Also the 1.53 GHz processors are running at a 133 Mhz DDR FSB while the 1GHz is running at 100 Mhz DDR FSB. Between the increase in FSB and Athlon vs. T-Bird I would think that might by it an extra 10% performance at equivalent clock speeds. 2792 * 1.1 * 1.53 * 2 = 9397.872. 9397.872 * .75 = 7048.404.
However, there is always overhead when running multiprocessor systems even with good multiprocessor archetectures like the EV6. Generally I expect to get 1.5 the performance with two processors compared to one. Anything much above the 7048 above doesn't sound right to me.
Anyone have any explanations?
This message is encrypted with Quad ROT-13 to protect the author's copyright under the DMCA.
I think Grace Hopper's analogy still applies: when one horse isn't strong enough to pull a cart, you don't try breeding a bigger horse, you just add another horse. Intel and AMD have created the ugly situation were their companies must revolve around creating faster and faster processors (stockholders love that) when they should really put more effort in making it more efficient for multiple processors to work together. The crap of the matter is, that when that happens (think: Xeon) the price of the chip is so high that it's cheaper to buy the commodity version and then run multiple machines.
Someone will get it right, sooner or later. Until then, the consumer just gets stuck with the bill.
I/O Error G-17: Aborting Installation
The original rebuttal quoted 3Dfx. It kinda slipped off track of main processors, but the argument still stands.
I/O Error G-17: Aborting Installation
Didn't you just contradict yourself there? If there are "teeming hordes" of single-processor boards, why build another one to compete in an already saturated market?
IMHO, the reason why multi-processor boards are expensive is because they target corporate consumers. You see this all the time in the prices with, for example, rackmount vs. desktop computers and the like.
I/O Error G-17: Aborting Installation
Tried what? Making a card that ran with two or more processors? The Nvidia Geforce cards seem to do it pretty well.
I/O Error G-17: Aborting Installation
imageine a beowulf cluster of these :)
The Dhrystone score is 58% higher vs. the PII, which is not such an amazing feat considering the clockspeed is 53% higher. --M
There is nothing strange about a factor of >2 speed-up in a 2-way run of the same job. Working on MPPs most of my life, I'm very familiar with the effect, which is due to the fact that the 2 cpu system has twice as much cache, and hence significantly less bus traffic occurs overall. The cpus don't stall as much waiting for memory to load.
-I like my women like I like my tea: green-
>Multiprocessor Thunderbirds will rock, I am ... Yup.
>certain. But this is as ridiculous as the
>recently reported SMP Thunderbird Linux kernel
>compile that supposedly demonstrated a greater
>than two times increase in speed between one
>processor and two
That is common. notice he used -j3. This causes 3 files to compile at once, not 2. This helps because one compiler may be waiting on i/o, but there is still 2 compiles to use the cpus. If the test was done with -j2, then it would have been "fair".
That's funny resourse utilisation doesn't go up when you use them, plus HDDs about double their benchmarks Mind you SCSI's the go in this regard, especially the more upmarket cards's which have a SDRAM slot or 2 on the card, that make fantastic caches or RAMdrive swapfiles.
That Supermicro board in a VIA dual Socket 370 board.
After Supermicro only supports Intel CPUs
it can be put at the bottom (like on the Epox, Iwill & Abit boards) sharing a backing plate slot with a PCI slot, thereby adding flexibility without any sacrifice such as losing a PCI slot.
Its much better than having a stupid useless upsidedown AMR slot there, or sorse giving up a PCI slot for a conventional AMR slot.
Besides jumpered ISA modems leave PCI modems for dead.
by sacrificing the bottom ISA slot, you just lose flexibility, as they are shared.
---
Anybody know what kind of memory these will be using?
If I can eventually buy one of these boards with a $70 dollar 256 meg stick of RAM from pricewatch Then it would be pointless to do any upgrades on any of my computers. I just wish I could quit drooling over this! Look Out Intel!
Well, not a big deal or anything, but the actual introduction of ISA (as far as the PC goes) would have been with the first IBM PC, which my copy of "The PC Programmer's Handbook" by Sanchez and Canton (this book rules, BTW) says the first IBM PC was released August 12, 1981.
:) here's a few about that machine:
Since we're rhyming off pointless facts
- 64k RAM
- 40k ROM BIOS + BASIC
- (TLA time) 1 SS, DD MFM 40 tpi 160k FH 5 1/4" FDD. MFM made it totally incompatible with GCR C64 floppies meaning one had to build an adapter...
- DOS 1.0 included (duh)
- 8088 processor (the 8086 PS/2 25 & 30 came later)
- 5 slots
- 1" card clearance (later shaved down to today's 0.8" for 8 slots)
- 63.5 Watt linear P/S (or maybe it was switching after all... one tends to forget)
- Choice of MDA or CGA video
- No HDD
- 83 key keyboard (the IBM M boards don't even begin to compare to this IIRC ALL-steel tank).
- EXTRA LOUD PC beeper
Now, wait a minute why do I care and where were we, oh yeah:
That makes ISA almost 20 years old! ISA is dead! Long live ISA!
If you could be told what you can see or read, then it follows that you could be told what to say or think - BoC
rr
Quidquid latine dictum sit, altum videtur.
Very fast no doubt but it will have a price tag to match and of course is aimed at the server market.
The K8 on the other hand will go back to using a shared bus (lightnig data transfer) and hence make multi processor motherboards a lot more common and they should be availible a lot closer to the launch of the K8 which has not benn the case with the 760MP and the Athlon.
A journey of a thousand miles starts with a brutal anal raping at airport security
i didn't see anything along this line in the threads http://babelfish.altavista.com does a nifty translation of urls
Why couldn't I find people who believe things like these speedup claims when I was selling electronics? I'd have made a fortune in commission. :)
Recall that Byte magazine used to regularly post adds for Write-Only Memory chips in their April issue. Great performance specs, too! The funniest part was that they always got a few serious inquiries...
i think a company called 3dfx already tried that and failed.
A not well founded statement. Intel's big super computers for example are made up of many individual nodes which are each SMP machines. In highly parallel systems the network is huge bottleneck and so by having greater number crunching on each node you gain better performance at a relatively low overhead. The future is not plain SMP or MPP but MPP'ed SMP nodes.
--
BACKNEXTFINISHCANCEL
they had some info on it as bp6.com It has scsi on board, dual socket A, yes, 64 bit pci slots. Bet it's gonna be an expensive bitch
oh yeah, dual onboard nics, too
because there are apparently already a few dual athlon mobos out there. They are starting to appear on the news sites, and chances are, they'll be available in 6-9 months. Maybe less. (Please!)
knee-jerk? check. post? check. okay, time to read the article.
Sufficiently Slashdotted...
Well Done.
rock on
That said, I'm not sure where the author you're quoting got the idea that it's difficult to overclock a 1.2Ghz Athlon to 1.53Ghz. I've seen a 1.1Ghz Athlon clocked that high, and 1.2Ghz clocked significantly higher (1.7 isn't a stretch at all).
It might be more correct to say that without changing the FSB speed, it's very difficult to get 1.53 out of a 1.2 Ghz Athlon, and that would be a reasonable caveat to make in this case, since the Tyan board doesn't provide for changing the FSB speed easily, but that's not what the passage you quoted said.
Ix
More like a rock of salt. My mobo has four IDE channels, and it is NOT IDE-RAID. Check the labels on the board next to the channels; if they read 1, 2, 3, 4, then it isn't RAID.
"Ancillary does not mean you get to rule the world." --U.S. Circuit Judge Harry Edwards, speaking to the FCC's lawyer
I wouldn't say these four slots on the board IDE configs have been typical for a long time, but they did catch my eye last time I was down in the computer market here in Taipei. I saw a bunch of them on both dual and single processor boards. It makes sense that they're for different buses.
But what impressed me about this was thinking about all the cheap IDE drives you could cram into it. If drive prices keep falling like they have then I think this is going to eat into the supposedly up and coming market for DVD writers. If we were to hazardously assume we could get 80 Gig IDE drives for around a hundred bucks next Christmas --okay, perhaps it's too hopeful, but maybe-- and you put six on one of these boards with the four way IDE slots then you'd have almost 500 gigs of drive space for about the same cost as blank DVD media, not to mention the writer itself and all the burnt discs you'd inevitably end up with.
Personally, I think Itanium will come out (June/July) before 760MP (August). There have been Athlon delays for over a year and all they're showing now are a couple of prototypes.
Well Itanium prototypes have been around a lot longer than these dual Athlon prototypes and Itanium still has a ways to go!
Heavy usage of these motherboards in the research stations located at the South Pole could in fact accelerate the melting of the polar ice cap!
:-)
Wow, I find that hard to believe. I have used Supermicro Boards for about 9 years now. I have probably sold/used around 200 of them in that time frame. Only once did I have a DOA and all the other boards met the garbage can before they were not functional.
I have had nothing but positive things to say about Supermicro. Their boards may not be the fastest, but I haven't seem them be flaky or never work. Currently, I have about 15 or so in place and being used.
Now if you want to see flaky and finicky boards let's talk about Intel motherboards!
dave
Sorry to burst your bubble, but ISA has been around since the advent of the PC. Granted this maybe older than the average /.er, but I believe that would in 1982. It went to 16 bit from 8 bit with the advent of the IBM AT in 1984.
Damn, I am showing my age!
dave
Most of the supermicro boards do have Ethernet on board. In fact some of their models even have dual ethernet! They also have slanted DIMM Sockets.
dave
The board in question isn't a SocketA (479) Athlon board... it's a dual FC370 P3 board, hence the name "Apollo Pro 266" on the sticker by the box, the Apollo name is given to Via's P3 range rather than their 'KT100' or 'KT266' type names given to Via's Athlon chipsets.
ISA is not supperior to PCI it cost more couse they have to add functions to the MB to handle it , as far as uses for ISA, as it stands now ISA true modems are curently the best modems avilable, not only are they 99% of the time true modems (i.e. NOT winmodems) you can use them in ANY OS that supports a com port , includeing dos,win,linux,beos,os2 ect. they are also cheap. the modem im useing now cost me 20$ 2 years ago and outperformed expensive pci true modems that I have tried.
hey dumbass, the tyan 760mp board is 6 layer, their foster board is 8.
ignorance is bliss. googlefiberatx.com
Here's what I posted to my site. Funny, I send in good AMD Hammer Linux info and I get rejected, but unlikely stuff like this and every article at sharkyextreme gets posted. Legit 1.53GHz Dual Athlon Sandra Scores? Reported by: Chris Tom At: 9:00 PM Source: e-mail 2CPU has what they are calling Dual 1.53GHz Athlon scores on a Tyan board. Well, there are a couple of problems with that which make me very, very unsure that these are legit. First of the Tyan 760MP board does not have overclocking features. I know, I held one in my hands at Comdex. So you think that they might add clock multiplier features in the meantime? Nope, what they had was the final revision of the board which only needs a final chipset from AMD to be complete. Second, it is very difficult to get a 1.2GHz Athlon to even 1.4GHz, much less 1.53GHz, and then you are telling me they got two to go that high? Lastly there are no details about this system at all except that it is using the Tyan board. Who knows, maybe it could be right, and I'm not saying 2CPU is making it up, but there are not enough details and not enough evidence, and there is too much logic keeping me from believing it. And for the fools that will say that I am jealous because I don't have dual scores to post here, don't even bother e-mailing me. That is ridiculous.
ignorance is bliss. googlefiberatx.com
It's nice to see AMD finally getting their SMP going. Though one would have to wonder, with the extreme heat of T-Birds, could these boards risk heat damage? ( Third Post ?)
Ashes of Empires and bodies of kings,
The truth about Michael
imagine a beowulf cluster of these!!
omega_rob
Ever taken a look inside a laptop? There's not much open space. This causes two issues when designing chips: form factor and heat.
To reduce the form factor means cutting down on the # of transistors on the chip (hence computational ability) or moving to a smaller process.
The heat issues are probably surmountable, though, but difficult.
This implies to me that if you want a dual-processor laptop, you'll have to have two CPU's with less power (each) than a single-CPU laptop. Considering how few applications are enhanced by multiprocessor boxes, you would in all probability have performance that was either equivalent to or not quite better than, a top-of-the line single-CPU laptop. Factor in the higher cost associated with the twin CPU approach, and there's almost no reason to have an SMP laptop.
Of course, IANA chip designer, so this could very easily be wrong!
I'm serious!
Methinks the post you're replying to is intended as humorous. Just because it hasn't been modded that way yet doesn't mean that it's not funny.
This may or may not be a useful piece of info, but ISA I believe was developed by IBM in 1984, as the original 8 bit standard... Just a useless little fact...
--
Join
Actually, in the telecom industry, it's not
uncommon to have 24 layer boards. Quite a few are
power and ground planes for noise problems, etc,
but the signal layers account for more than 3/4
of the board. Yes, they do get thick. You can
beat someone to death with some of the backplanes
I've seen.
OTOH, the thickest PC mobo I've heard of was about
8 layers.
Promise FasTrak 100 controllers running RAID 10 are very fast, IBM 75 GIG drives net a 150 GIG of data and is thousands less than a SCSI system. The performance is VERY GOOD.
Your Average Joe
Anyone found something on planned release dates?
this is a copy of a post i made at benews. go here to read the follow-up posts.
For a long time I have been unable to use BeOS. My computer would mysteriously and randomly completely lock up. I say randomly because I started to keep a log, but there wasn't any pattern at all.
I looked all over the web for help, and since I had a graphics card that some people had reported to have problems, I replaced it. This did not solve my problem.
Next, I thought something might be the matter with BeOS. This seemed to be true for a while because it got to the point where I needed to reinstall and the reinstall kept crashing. Linux was still working at this point.
But it got worse. Linux would crash periodically and sometimes my ext2 partition would be too mangled for fsck and I would try to reinstall. It got to the point where my distribution would not install (Gentus). I tried a few other distributions and Debian Installed but kept giving me these IDE time-out and restart warnings.
I actually like Debian and would have never tried it out if it hadn't been for this problem, but my problem got worse. Debian started having the same problems as the rest of the distributions.
A friend suggested we try installing Windows 2000 just to eliminate the possibility that software was the problem. We tried a couple of times and Win2k's best was 76% completed and then crashed.
Since I had gotten some error messages with Debian about my IDE I thought it might be my IDE controller. I found a few people on the web who said that they also had problems with their Abit BP6 motherboard's IDE controllers. So I started the process of choosing another motherboard.
I finally selected an EPoX KP6-BS with slotkets. (I have two 500Mhz Celerons).
I am glad that I finally found the proplem. I have not had a single crash since I installed the new motherboard and I have done everything that made it crash before. Installed BeOS and Debian GNU/Linux, compiled my kernel 3 times, forced an fsck, Used Net+ (Net+ crashed once, but BeOS caught it), etc.
I really enjoy using BeOS and really missed it during the several months it took me to finally find the problem.
I am posting this here for anyone else who may be having similar problems. I hope this helps.
one thing to keep in mind. as a single piece of hardware becomes more complex, the chances of a single manufacturing error rendering the part useless go up.
The Highpoint 66 controller is definitely a problem. It is a major pain to get them configured in any OS except their Linux distribution (Gentus).
I played around with the them for a while, but because I wanted to use BeOS, I switched to the 33. This was before I was having any serious problems.
Isn't ISA superior to PCI?
If not, why in the world does it cost more?
Ben Schumin :-)
I agree. Definitely not IDE raid. Why would they put some low-end Raid-0 or Raid-1 on board for a SERVER motherboard? If they want raid, they'll get a dedicated controller and do it right. They don't need some half-assed raid solution on board.
Ben Schumin :-)
The SuperMicro board linked to is for Intel CPUs (either Coppermine or Tualatin), NOT AMD. It's based on the VIA Apollo Pro 266 chipset, which provides support for DDR SDRAM, as well as regular old SDRAM (hence the "3+2" DIMM setup, though it will not support both kinds of memory simultaneously.) As far as dual Athlons goes: right now, the only chipset that supports dual Socket As is the AMD 760MP (which isn't even out on the market yet). The board pictured on the Japanese-language site is of Tyan's version of the AMD reference design, which was shown at Comdex in Vegas in November. It is "Extended" ATX form factor and has a 6-layer design, thereby taking it out of the price range of any BP6 owners who may be drooling over it. Dual Athlons (indeed, dual Durons) are definitely on their way, but we will probably have to wait until the VIA KT266 chipset (AMD processors + DDR) for a practical solution.
Not to be completely obnoxious, but please take a look at the damn picture before you go spouting off about something you know nothing about. If you care to look at the picture (the url is: http://www.watch.impress.co.jp/pc/docs/article/200 10124/17.jpg
, in case you never felt the need to go there), there are four IDE connectors on the board, the two at the top are most likely connected to the VIA Apollo Pro 266 chipset (which is in fact ATA/100), and the other two are strangely positioned next to a label that says "IDE RAID" if you look closely. So, not only is there IDE RAID, but I am quite certain there is no ATA/33 or ATA/66 on this board.
And what does this looks like?
* Dual Athlons
* 2 Ethernets (look near the parallel port)
* 2 SCSI connectors (in the extreme left side, the picture quality sucks)
* 4 64-bit PCI slots
So what does this looks to you? a typical workstation?
Hetz (Heunique)
Or maybe years ago....The AMD K6-III (that was years ago, wasn't it?) had a L1 and L2 cache. It went into existing K6 motherbords that had an existing L2 cache, and it worked with the existing cache, making that an L3 cache.
Not that it was all that popular.
This really is what's going on, and everyone is doing it. Memory speeds are advancing much more slowly than CPU speeds. To fight this, we see more or faster cache, more levels of cache (expect L3 to start showing up soon on IA32) cleverer cache controllers and cleverer compilers. Real benchmarks (see for instance http://www.specbench.org) show improvements from memory OR CPU speed-ups, so neither is totally bottle-necking the other at this stage.
Thanks! Good point. The KT series are often advertised as having ECC, but, looking at VIA's spec sheets, the KX actually has it.
My blog: http://www.seebs.net/log/ --- My iPhone/iPad app: http://www.seebs.net/seebsfrac/
A lot of 133Mhz motherboards based on VIA claim to "support ECC"... but don't actually do any error checking or correction. I'm not sure where this came from; the VIA specs don't list ECC as a feature, but the ABit VP6 specs *say* the board supports ECC. It doesn't.
So, if you want a nice, stable system, be very wary.
My blog: http://www.seebs.net/log/ --- My iPhone/iPad app: http://www.seebs.net/seebsfrac/
This implies to me that if you want a dual-processor laptop, you'll have to have two CPU's with less power (each) than a single-CPU laptop.
Transmeta's chips (and the G4) both use much less power than Intel/AMD offerings. The G4 is also smaller (don't know about Crusoe). That's why I think AMD would need some new technology to pull it off. There's also all the cool shit IBM has (0.01 micron process in the lab, SOI, etc.).
Apple has demonstrated the possibilities of G4 laptops (with no fans, I believe), so a dual G4 or Crusoe laptop should be possible with clever engineering. Phase change heat pipes might be useful, for instance...
...but can you imagine a Beowulf Cluster of these? [Ducks for cover]
Those are very cool (figuratively), but I bet they put out tons of heat. Someone mentioned that the board appeared designed to fit in a 1U case. What would it take to properly cool it in such a small enclosure?
What would be even cooler is if AMD bought Transmeta and made dual-CPU laptops (or at least chips suitable for use in said laptops). Then there might be something out there cooler than Apple's Titanium PowerBook G4. Multiprocessor laptops... Drool.... For those occasions when you need to perform advanced simulations of nuclear weapons tests while on that long redeye flight. (Can anyone else actually think of a good use for that much power?)
Please excuse my ignorance, I haven't followed hardware developments very closely for some time. ... could somebody enlighten me what news there are other than the raw clock speed? (and no, I'm not trolling, I really want to know. otherwise I would have checked 'post anonymously' ...)
So those Thunderbirds have a multiplication factor of 11.5 over the system clock? And over the RAM, if it's SDRAM133 (or whatever that is called). Even with DDR that still makes a factor of about 6, which imho is downright ridiculous. How well do these things score in real applications, not just cycle counting in first level cache (aka MIPS and MFLOPS)? I can't imagine you get much over a 800 MHz system
EagerEyes.org: Visualization and Visual Communication
The only difference I can see is that mine has 4 DIMM slots as opposed to 5. I can't tell from the picture in this article whether it's a socket for PIII or Athlon though. Considering it's got a VIA chipset that looks just like mine, I'd say this motherboard is just a dual PIII board.
However, it totally rocks. :) To sum up a few of its features:
- Dual IDE buses (yes, it does have 4 IDE ports), one at ATA 66, one at ATA 100.
- Firewire and USB onboard
- Nice clocking facilities
As well as the usual neat bits, and onboard sound. Recommended if you need a PIII board, specially dual processor.Fross
I'd have to agree. You can clearly see that the two dimm slots nearest to the processors are different than the other three. The two closest to the processor appear to be used for DDR SDRAM, due to the single notch, whereas the other three appear to be for standard SDRAM, due to the two notches.
Not likley, I dont know about the pc motherboard market specifically but in other fields of electronics manufacturing it isn't terribly uncommon to have 6, 8, or more layers to a pcb. It may be a little more tricky for somthing the size of a full ATX mobo but it's not uncommon. It shouldn't be teribly expensive in production quantities either (~$20 per board in ten-thousand quantities at a guess).
"Prefiero morir de pie que vivir siempre arrodillado!"
Ok, I would like to say that no one thought of using the babel fish japanese to english translator. That site was not entirely about a dual socket A motherboard. The supermicro board you saw was the 370ddi board, a new dual p3/celeron board that has 5 SDRAM slots, not DIMM slots. The actual dual socket A part was very small and only mentioned the Tyan board. Please translate before you go posting something like this, because it just creates a lot of hype for things that are false.
When I worked for Corel (in Ireland on Linux Support, not Canada on code, remember that let's you know the sort of people I could get too and hence how much the following is about personal opinions rather thatn corporate policy, it's just what I think), I tried (in the limited ways that I could) to get them to do just this! I suggested that if they can give away free to use WP8 than surely they can dig back somewhere and give away a Free version (say the first draw+wp versions released for unix). Alas, while a few people thought perhaps, they generally couldn't grasp any benefits to them. As for GPL or even Open Source WP9/10 ... I don't think they had the stomach for that AND I don't think they thought anyone else would have a clue how to work with the code ... I'm not even sure if they knew if they could work with the code. When I heard talk of a re-write of some of the programs I always suggested it again (if anyone who might be worth saying it to and I knew wasn't against the idea was around) because it could offer someone like them the opportunity to ditch being toolkit writers and let them focus on the important stuff.
Never underestimate the dark side of the Source
IDE RAID on the MOBO has already been done.
:)
:(
Check it out here: Abit KT7A-RAID. It's got the on-board IDE RAID you dreamed of, AND it's been acclaimed on Anandtech... their mobo pick for many of their high-end systems. Ultra ATA-100, 3 DIMMs, 1 AGP, 6 PCI and you even get to keep your one ISA card
Sadly, no dual processor support yet
Download your mp3s any way you want, and support the artist via FairTunes
I used to talk directly to the head of the tech support and he was also in charge of overseeing development. We had some customers want some really particular bios settings and tech support said they would burn us custom bios if we really needed it. Have you actually tried updating the bios? I built a Lightwave animation box with a P6DBU and a 3d Labs Oxygen RPM video card. There was a conflict with the bios. I jumped on their site and found the updated bios right away. Their tech site is good with updated bios, pdf manuals, etc.
Actually all those Abit boards (& Epox & IWill, etc) that have an extra IDE controller, so that they support 2x4 IDE drives (instead of the normal 2x2 IDE drives) suport RAID (0,1 & 0+1) on the extra 2 IDE connectors.
Abit & Epox used a Highpoint controller, IWill uses an AMI controller, while Asus uses Promise (which on some Asus boards has to be bios hacked for RAID). All 3 brands of controller have their good & bad points.
Looks like it has 5 x 64bit PCI slots, and the four DIMM sockets are slanted, perfect for a 1U case.
What I am interested in is some info on some older multi processor boards, just so that I can get the last bit of milage from some older cpus I have kicking around.
But then I remember advice I used to give some retail customers, half in jest:
Remember, if you can buy it here at (mass market store), technically it is obsolete already.
But then, some people love living on the trailing edge of technology.
"It is a greater offense to steal men's labor, than their clothes"
Take a closer look. The first two slots closest to the processor are slightly larger. The keying notches are also in different places. Most likely DDR slots, but the board still has SDRAM slots for those who don't want DDR; now that's a nice feature.
Now, all of this stuff is well and good (dualie Athlons), but AMD needs server-class boards. That means things like built in SCSI, 64 bit PCI, etc. AMD has wanted the server market since they originally introduced the Athlon (and everyone wanted it since then) so it should be here soon.
..not faster cpu's that hide the flaws. Case in point "Wordperfect 5 (or 4.2)".
--
Jon - TheSpork
There this 760MP based board from Tyan shown in the article. Only four DIMM's though...
All these boards look fine to me. But what is more apperent now is that SMP is truly a dead end when it comes to multi processor systems. Intel has bled it to the last drop and others such has Sun and IBM have taken SMP to new bounds that are still not matched in the PC industry.
But, right now we are heading in the direction of massively parallel comuputing done using non SMP meothods. Good clustering practices has made SMP nearly obsolete. Mix that with the non ability for non intel vendors to come up with a motherboards for non-intel CPU's on the PC market, I feel that the end of SMP has been well written in concreate and hardended.
Mix that with good software such as Mosix which lets you parallel process any software without specfically having to compile the software into a traditonal beowulf type messging system (PVM/MPI), we are now on the verge of having home grown Cray system in every basement.
Enjoy
Trust the source!
I know there's a lot of hype behind supermicro, I know they have really cool feature lists.
But I know too many people who have paid a premium to own a supermicro product only to find that they were slightly flaky, that the super nifty features didn't work properly, and that the boards were quickly orphaned (support discontinued) when newer boards were released.
I won't buy 'em, and don't recommend 'em.
This is just like television, only you can see much further.
I found this over at amdzone I agree:
2CPU has what they are calling Dual 1.53GHz Athlon scores on a Tyan board. Well, there are a couple of problems with that which make me very, very unsure that these are legit. First of the Tyan 760MP board does not have overclocking features. I know, I held one in my hands at Comdex. So you think that they might add clock multiplier features in the meantime? Nope, what they had was the final revision of the board which only needs a final chipset from AMD to be complete. Second, it is very difficult to get a 1.2GHz Athlon to even 1.4GHz, much less 1.53GHz, and then you are telling me they got two to go that high? Lastly there are no details about this system at all except that it is using the Tyan board. Who knows, maybe it could be right, and I'm not saying 2CPU is making it up, but there are not enough details and not enough evidence, and there is too much logic keeping me from believing it. And for the fools that will say that I am jealous because I don't have dual scores to post here, don't even bother e-mailing me. That is ridiculous.
Pretty much says it all regarding the benchmarks we "saw".
Try to hack my 31337 firewall!
Comment removed based on user account deletion
________
Does anyone actually have a Java program designed to control air traffic, or for the operation of a nuclear facility?
- Installing the latest RPM or DEB packages, doing a bit of config work via rsh, and being able to scale to 64 way by putting 64 boxes in the rackmount unit, or
- Designing a motherboard that integrates together 64 CPUs?
SMP certainly provides the benefit of very fast communication between CPUs, so if processing is strongly dependent on that, SMP wins.But it is most certainly a lot more expensive to scale it up. The vendors may sell 2-way and 4-way SMP motherboards for not overly princely sums, but moving on to higher multiples guarantees pretty monstrous prices, because you're simultaneously mandating:
- Small production runs, compared to the teeming hordes of single CPU motherboards, and
- Sitting at the bleeding edge, because nobody wants a souped up version of a CPU that was "state of the art," LAST YEAR.
Kernel compiles aren't liable to benefit all that much from parallelism, moving forward; consider that with a Pentium III or Athlon, it doesn't take long to recompile Linux even from scratch. It's just getting to be less and less an issue.As for benefiting from other forms of parallelism, it is entirely likely that the toolsets surrounding Beowulf and PVM will improve over time to make it easier to manage doing "clustered tasks" in much the same way that we have progressed from having rather primitive "package management" tools to having stuff like AutoRPM, apt-get , and BSD Ports.
If you're not part of the solution, you're part of the precipitate.
The 5 DIMM slots are almost certainly a combination of 2/3 with 2 being one type, and 3 of another (between SDRAM and DDR-SDRAM).
The 760MP has been more than a passing interest to me lately, and I've been digging up information/rumors about it daily for the past week. This is what is sounds like:
One EV6 bus requires a hefty chunk of PCB. Two will require even more (of course). This means that DP Athlon boards will require as many as 8 layers to fit into an ATX form factor. IIRC, the industry standard is 4 or 6, so this would be a new (and likely expensive) manufacturing process that may require new tooling to produce in bulk.
Athlon boards are already more expensive than P3 boards, and I think the overall DP price/performance comparison isn't going to be that bad for Intel when/if 760MP ships.
--
Maybe peyote juice?
But then again we must remember that this story has been promoted to front page material by the same group that brought us nanopants.
That's not a grain of salt the editor mentions, that's a rock of crack ...
But I digress ...
Why do I strongly suspect this is non-authentic? Does anyone else remember the photograph of the modified Duron that was supposedly being produced by AMD to thwart overclockers. A pin was physically "removed" from the pin interface.
The picture was posted all over the place. Everyone was all up in arms. It was the end of all things.
And then someone noticed that this "modified" processor had the same serial number as an unaltered promotional processor photo from another web site.
NEW FLASH: Overzealous Tech Sites Taken in by Paintshop Pro Forgery ...
Many Japanese tech sites are notorious for posting outrageously altered faux benchmark screens. This would appear to be another one of those posts ...
Multiprocessor Thunderbirds will rock, I am certain. But this is as ridiculous as the recently reported SMP Thunderbird Linux kernel compile that supposedly demonstrated a greater than two times increase in speed between one processor and two ... Yup.
Why couldn't I find people who believe things like this when I was selling electronics? I'd have made a fortune in commission. :)
Get off my virtual lawn, you damned virtual kids!