Slashdot Mirror
Dual Athlon Motherboards Creep Closer
Quixote writes: "The Inquirer has an article about a dual Athlon MoBo sighted in Taiwan. Running with 1.8GHz Athlons. Sweeeet!. Ahem. Tom's Hardware also has a photo of the MoBo. Can't wait to get my grubby paws on one!" AMD seems to be waiting for the Christmas shopping season or something -- would be nice to see some mid-range dual-AMD motherboards soon, because those dual-NIC, built-in-SCSI ones look a bit out of reach for now.
Then there was the old Macintosh Portable with its 1.5A power supply. That wasn't enough to spin up the hard drive, so you *had* to have a battery in it during boot. (And I never would have debugged that blown fuse if there hadn't been a similar and common problem on the powerbooks . .
hawk
Well, I finally broke down and bought test equipment. A 1.5 P4, i850 motherboard, 640 MB RDRAM and 1.33 Athlon, AMD760 motherboard, 512 MB DDR SDRAM. I built them, installed Linux, installed the CFD software my customers use, and did some benchmarking.
The long and short is this: for this application, the P4 was the winner by a huge margin. The 1.5 P4 with RDRAM was over 60% faster than than 1.33 Althon/DDR rig.
So, this is important to me and to other scientific computing folks for a number of reasons:
- P4 prices are in a freefall
- I can put 1 GB of RDRAM on an i850 board with a 1.7 GHz for $1400 (CPU $400, RDRAM $800, MoBo $200)
- The only DDR motherboards worth anything are based on the AMD chipset and they only have two DDR sockets. So, to get to 1 GB DDR, it costs about $600 each for two DDR DIMMs...$1200 just for RAM!
- Even the "best" DDR implementation is only the slightest improvement over SDR. (about 10% in my tests)
So, I am sure that this will infuriate the lemmings who wander in to moderate. I have been waiting for the dual AMD setup with anticipation, but when it comes, it desparately needs decent DDR support. At least for my application, Intel and RDRAM are doing something very right.So, RAMBUS sucks as a company. Intel isn't much better. For Unreal Tornament and Office Bench-O-Rama 2000, the Athlon might be the easy choice, but I think Intel has a viable platform in the i850 and they may well evolve it into an outstanding dual system. They have the kinks out of their RDRAM implementation. AMD and VIA should take note. Their DDR implementations are worthless.
I don't think that was true, I know it got slammed in initial benchmarks, and then a few months later it was doing OK (but not significantly faster then the much slower AMDs). I think that has more to do with the motherboard chipset and memory systems improving then any compiler changes. Then again I haven't been watching closely, so I could be wrong.
The compiler itself has a fair number of CPU specific bits, but they are mostly enabled by flags. On set for "what exact instruction set should I use", and a second set for "what CPU should I tune for". So you can make sure your code will run on a 486 or better (but maybe not the 386), but have the instruction ordering and cost weightings for a AMD K7, or a Intel P4 (I think that is -mtune=CPU).
Redhat may not want to compile up different RPMs for each arch, but if it is important to you, you can do it yourself. I don't know how, on FreeBSD you can do some unclean things to the master Makefiles and pretty much everything will use the new settings. If RPMs are mostly source you should be able to do the same kind of thing.
That's still not a why :-) I have been told you can use the Intel compiler as a backend for MS Visual C++ if you own it, so I guess either that compiler is too slow for most Windows software authors to want to use, or too expensive for their managers to pay for. I wonder if they can use gcc as a backend...
I expect the Alpha would do well there also. I don't think the Alpha is dead, there are still good people working on it. Just not as many. The next Alpha (which is late already, tape out being a significant fraction of a year late) looks nice. The one after it (which might come on time!) looks like a real killer.
Of corse the other "next gen" CPUs look pretty killer too. The IBM POWER4 actually looks more impressive then the next-next-Alpha. I don't recall that being the case before (the POWER has held the lead, but mostly by having their new CPU coming out a few months ahead of the new Alpha, not a year or more!). Of corse IBM seems to be playing it close to the vest, so we don't know if the POWER4 is on time.
It happens. I've had posts mistaken for personal attacks that weren't.
I doubt it. Most of the compiler related benefit is pretty small. The prefetch instruction for example gave about a 20% boost to the stream benchmark when shoved into an experimental version of gcc (this was the AMD prefetch). At that time it wasn't taken because it made some other benchmarks worse (the compiler wasn't smart enough to know when not to use them). That was about 18 months ago, so I'm not sure if they were improved and put in, or shelved for a post-3.0 release. Most other tweaks are smaller. SSE/3DNow would show a bigger improvement, but so far no compiler has done much with them, that is all hand coding (or on the PowerPC minor compiler assist because Apple modified gcc to have AltiVec datatypes), but you still have to change the C/C++/ObjectaveC yourself).
Just as importantly gcc sees optmisations for both CPUs (and many others), not just the Intel version.
The Intel compiler (as far as I know) doesn't get AMD optmisations, but it also isn't all that wildly used, despite being a very nice compiler. Most windows code isn't all that optmised, I'm not really sure why.
(note the superscaler changes seem like they would require a lot of compiler help, but ever since the PPro the x86 CPUs have mostly been out-of-order machines and don't need much compiler help in instruction ordering to get pretty close to top speed so unlike the 21064 or Pentimum1 or SuperSPARC rather then getting a 2x to 4x speed up for getting just the right ordering the speed up is tipicaly more like 10% and that is when there are lots of cache hits!)
Actually if you want integer performance Intel pretty much has the SPECInt crown (at least last month comp.arch was abuzz because the Alpha had finally lost it, and was in danger of losing the SPECFP as well, but that's what happens when half your design team is lost and your new CPU gets to be 20% as late as Intel's Merced).
The SPARC isn't a performance leader, and hasn't been for a very long time. It does give you access to some great rackmountable hardware, a ROM monitor that is great for lights out management, and a lot of other things, but raw CPU speed isn't it.
I think the Alpha still wins in SPECFP, but if you can do with reduced accuracy non-IEEE FP the PowerPC or Intel or AMD may beat it. For I/O the S/390s seem like a better bet :-)
Pretty good advice.
If you are CPU bound BSD/OS will do fine. If you have some I/O in there Linux and FreeBSD aren't too bad, but they could be a lot better. They are certainly better then Solaris was after the same number of years of effort. I'm not sure the BeOS kernel is actually any better then those OSes. The userland is better positioned because of the way they designed it, and they promote use of threads quite a bit.
P.S. yeah, I know it was probably a troll, but I had to reply :-)
The article mentions support for Screaming Sindie instructions. That's a good one. Someone must have dictated "Streaming SIMD" (or SSE) over the phone and the reporter obviously didn't have a clue what was meant.
Jeff
Opinions expressed are my own and not necessarily those of my employer.
Have a look at quietpc.com. I've just finished sorting out my 1Ghz Tbird to get it down to an acceptable temperature and noise level. They do silent/quiet PSUs approved by AMD, fans, drive enclosures and other bits. The noisiest thing in my PC now is the 25mm fan in the back of my CDR, which sounds like it's about to fail - Papst have just announced a quiet 25mm fan, but it's not actually available yet.
Also try http://home.swipnet.se/tr/silence.html for more lots of info regarding 'Silent PCs', TCO99, and what manufacturers can help you.
"don't fall into the fallacy of believing that Perl can solve social problems. Maybe Perl 6 can, but that's a ways off"
The inquirer also has an article which breaks down the 460 watt PSU requirement by component. I still want one even if it would up my electricity bill. Random thought: I hope they don't release these first in California...
--
I usually make subsidary boxen out of 'em and their various supporting hardware.
(ie boxen for irc, icq, muds, and light browsing; low-end servers...)
But donating 'em is good too, and I wish my old school had done that more often - they had a literal huge closet full of slightly obsolete PowerMac G3s. Not that I like PMS much, but...that much processing power going to waste irritates me.
-- Veni, vidi, dormivi
I've been waiting for dual-Athlon + DDR for about a year, but I finally realized that they weren't actually going to come out until I spent my money on something else. So I went and bought something else a couple of weeks, to clear the metaphysical logjam.
Think of me when you boot your dualie.
--
Sheesh, evil *and* a jerk. -- Jade
> Of course, it is in Japanese, but the pictures are *great*.
Yeah, I visit that kind of Web site now and then too.
--
Sheesh, evil *and* a jerk. -- Jade
As far as I know, Durons are just as SMP-capable as Athlons, and should work in this motherboards.
The only catch is that Athlons will perform much better due to their bigger amount of cache, and since the motherboards will probably be rather expensive, at least in the beginning, there might not be much point in buying one, and then fitting them with the cheaper but poorer option. Besides, athlons are not very expensive right now.
The parallels are true. Lots of car freaks are geeks. Many geeks are car freaks. Think of how much some of us enjoy hacking, and remember that hacking a car isn't that different from hacking a pc. Philosopically, anyway.
Best Slashdot Co
If you don't care about it looking good, cooling isn't too big a deal. It's easy, even. Either just totally take off the side of the case, or drill lots of holes (if you care about RF), and then blow a $15 super-quiet desk fan at it. I'm running a duron 600 @ 900mhz, plain ol heat sink & fan it came with, @ 47C w/28C ambient. Plus, it cools my overclocked TNT2 much better, and does a good job on my hard drives and power supply too. All for $15cdn..
Same setup would work really well with a dual configuration, and it's whisper quiet, too. Could always do it with ductwork, I guess, but this works great, and it's $15cdn. Can't beat that. Heat concerns amuse me when they have to put a 1lb heatsink on the P4, and that's to act more as a heat transfer buffer than an effective sink, imho.
..don't panic
Yea, right...
The UP1000 would technically accept a Slot-A Athlon in the Slot-B slot. (Slot-B being a
superset of Slot-A) Why you'd want to do that when you can get a Slot-A motherboard cheaper is beyond me.
The UP2000 has Slot-B slots but the chipset, the Tsunami, doesn't support Athlons.
There's no such board as a UP1200.
FWIW, I work at API.
What's my Karma Mr. Burns? "Excellent"
IHMO I don't see the big deal about running out to get the `latest` high speed (Mhz) chip to run from any vendor be it Intel, AMD, Transmeta, etc., especially something that hasn't been tested in real world condition so my question are as follows...
[1]
When the 300's and 400's came out the same hooplah surrounded them and everyone hyped them much the same way as the comments on this story. If marginal fractions of people brought those chips, then carried on to other higher speeds, what is everyone doing with their other chips... Is there a chip cementary I should know about or are people doing cool shit like donating older chips to non profit companies, or edu's or something? What the heck happens to all those outgoing chips?
[2]
What about recycling some of the 300's and 400's and making a Quad motherboard running say a PIII 400 wouldn't it be cheaper, and faster for certain tasks. Has anyone tried this?
I'm not hardcore on hardware since I don't do any high tech scientific computing, or distributed computing, hardcore gaming, etc., so I always wonder who in the `real world` buys these chips and for what...
Want Root?
create a an overlying case for the PC with 4 slots cut out, left side, right side, top, and back. On the side place some fans blowing inbound, let the back blow outbound. With the top cut out you could get a hose similar to something a clothing dryer would use, and hook it up to a "Penguino" floor standing air condition to keep cool air circulating into it.
Actually you could make a nice little desktop set up with something like that and keep most of your PC's in there to keep them cooled. Just don't forget to have AC filters on the insides of the carve outs to keep the PC's dust free. Or you could dish out for a single freon cooled case.
Want Root?
It is now at:
http://www.tomshardware.com/technews/technews-2001 0510.html
There is also another article with decent info here
They also have updated info on the specs on the front page.
Check out the Vinny the Vampire comic strip
"It is a greater offense to steal men's labor, than their clothes"
http://www.tomshardware.com/technews/technews-2001 0510.html
http://www.tomshardware.com/technews/technews-2001 0510.html
I don't usually flame people but I can't resist this one.
This is so full of crap I almost don't know where to begin. The possible performance increase you might see in the future from using the P4 is offset by the fact that it costs more know and doesn't perform as well. In the future that you are looking at the Sledgehammer will make a much nicer desktop processor than either of those chips. Get the best performance you can afford know and upgrade as necessary. If you really need the best performance you use what is best at the time of purchase and upgrade when there is a significant improvement by changing. You don't buy the slower now and pray for miracles.
As far as OS's go, that is what truly gave away the fact that you are a troll. SMP support in Linux and W2k is more than adequate for any purpose your limited brain can come up with. Especially since SMP performance is determined as much by the app as by the OS and the current crop of apps could all stand a little work in that area. There is also the fact that the Palimino chip cans use 3Dnow and Screamin Sindie instructions if I remember correctly.
You buy your P4. I'll spend the same amount on an Athlon system and spend the difference in price on RAM. My system will smoke yours now and even at the end of there useful lifespans there won't be an appreciable difference in performance over the range of apps most people use. I'll still beat you on everything except games and graphic apps.
"If there is nothing you are willing to die for, then you are not really alive." Myself
You have restored my faith. My comments about compiler based performance increases are drawn from benchmakring done prior to the P4 release. I've utterly forgotten the details, except that P4 was slammed utterly in the initial benchmark, but after severe hand made adjustments from the intel techs, it blew the competition away. Evidently the compiler wasn't quite up to scratch at that point.
Obviously designing a compiler like gcc requires making some trade-offs, given the way it is used. Redhat can't afford to compile rpms exclusively for x86, and they can't start seperate distros for AMD and intel, realistically. As a result, compiler optimisations are a compromise.
I can tell you why the windows compiler doesn't optimise well: Apparently they haven't written code to optimise for P2 yet. Hopefully, they'll skip a generation or two and jump straight to P4.
Drag about alpha. I'd still recommend sparc for high-end over most other things, due to excellent bus speeds and huge MP support. Bus speeds count for a hell of a let in some fields.
Finally, it wasn't meant to be a troll. Seriously. I don't know what happened there.
Denial isn't just a river in Italy