Re: Mersenne: Problems with ATI graphics
3) The on-board ATI graphics chip is about 2 cm away from the Slot1, directly underneath the CPU's big heat sink. Hmmm, sounds pretty suspicious to me. I would move it down a slot or two... "ON-BOARD" ATI graphics. its a all-in-one motherboard. can't move it. re: emf radiation, I heavily doubt that a CPU could radiate any significant EMF thru a heatsink, which after all is a large mass of aluminum, presumably grounded. I'd be more likely to suspect power transients coming through the +5VDC. 4) The distributed.net client does not cause this behavior. My understanding is that it does not use either FPU or cache. Huh? If you are LL testing you are using FPU and cache. I'm almost as sure that applies if you are factoring as well. he meant the distributed.net client does not use FPU or cache. In fact, it does use some cache, but not nearly as heavily as the LL tests do. The distributed.net client is that RC-5 encryption thing, its all integer math. -jrp
Re: Mersenne: Problems with ATI graphics
On Thu, 28 Jan 1999, Louis Towles wrote: I tried this a few months ago on our E series Gateways (ATI Rage Pro) Prime95 LL test - yes interference Prime95 factoring - no interference Distributed Net Client - no interference FPU benchmark (I think it was ZD winbench's FPU test) - yes interference The no FPU benchmark - no interference It's is in all video modes (although harder to see on some) It start as soon as you start pegging the FPU (can be right after boot with the chip still cold to the touch) Stop as soon as the FPU backs off Louis Towles All this sounds like an excellent time to test the warranty on those motherboards, as they're obviously defective, and fundamentally flawed in their design. Henrik -- Henrik Olsen, Dawn Solutions I/S URL=http://www.iaeste.dk/~henrik/ Get the rest there.
RE: Mersenne: Problems with ATI graphics
POV-Ray is a multi-platform freeware raytracing program that makes heavy use of of the FPU. You can download it from http://www.povray.org/ . The Windows version will actually even render a test scene for you upon installation so you don't have to know anything about using the program. Kent D. Johnson [EMAIL PROTECTED] Jesus is Alive...Elvis Isn't! -Original Message- One good test would be to use any other program that uses the FPU alot and see if the flicker shows up. Can anyone think of another program that would? I suppose any benchmark that tests FLOPS should do the trick.
Re: Mersenne: Problems with ATI graphics
I should have made explicit that "when we examined the processor usage of the mainstream business applications in the Winstone test, we found that they do not use any floating point or MMX instructions. Consequently, CPUmark99 doesn't include those types of instructions either. " per http://www.zdnet.com/pcmag/pclabs/bench/cpumark99.html so this is an elimination benchmark. But Louis Towles' response seems definitive. Leo Leo Feret wrote: Aaron Blosser wrote: Could someone explain this "noisy" cache/RAM? That sounds like it may be the problem. Since the cache on the PII's is built in, saying it's a cache problem might as well be the same as calling it a CPU problem. The cache and integer bits of a PII are used with some regularity even if the machine is just sitting there. One good test would be to use any other program that uses the FPU alot and see if the flicker shows up. Can anyone think of another program that would? I suppose any benchmark that tests FLOPS should do the trick. It might be interesting to run CPUmark 99 from http://www.zdnet.com/zdtv/screensavers/projects/story/0,3656,2175057,00.html Leo
RE: Mersenne: Problems with ATI graphics
Ditto on Aaron's excellent comments. I've added some of my own. From: "Aaron Blosser" [EMAIL PROTECTED] Date: Thu, 28 Jan 1999 14:33:52 -0600 Subject: RE: Mersenne: Problems with ATI graphics ... 2) I don't think heat is the problem. The problems start within 5-10 seconds of starting prime and stop as soon as prime is halted. Then we can say it's related with some certainty. I've actually measured the heat put off by mprime. The above observations are completely consistant with how the CPU heats up. 3) The on-board ATI graphics chip is about 2 cm away from the Slot1, directly underneath the CPU's big heat sink. That's *way* too close, I would think. Good lord, yes. For most cards, they might be OK. But don't forget that some video cards can put off a good deal of heat themselves. ... One good test would be to use any other program that uses the FPU alot and see if the flicker shows up. Can anyone think of another program that would? I suppose any benchmark that tests FLOPS should do the trick. Another good test would be memtest86. I've put a copy of it up on supercomputer.org's download section, for both Windows and Linux users. It will generate about 2C more heat than mprime/prime95 running the torture test. And this doesn't use the FPU. To get the maximum amount of heat, use memtest86 in cache mode. -dwight- === To learn how to build a supercomputer for under $10,000 visit www.supercomputer.org ===
Re: Mersenne: Problems with ATI graphics
2) I don't think heat is the problem. The problems start within 5-10 seconds of starting prime and stop as soon as prime is halted. I never said anything about heat. I said electromagnetic radiation. It emenates (sp?) from anything that is electrical. The more electricity, themore emenations. It works the same way as a radio transmitter, and unfortunately your graphics card appears to be the receiver so to speak... 3) The on-board ATI graphics chip is about 2 cm away from the Slot1, directly underneath the CPU's big heat sink. Hmmm, sounds pretty suspicious to me. I would move it down a slot or two... 4) The distributed.net client does not cause this behavior. My understanding is that it does not use either FPU or cache. Huh? If you are LL testing you are using FPU and cache. I'm almost as sure that applies if you are factoring as well. -- ~~~ : WWW: http://www.silverlink.net/poke : Boycott Microsot: : E-Mail: [EMAIL PROTECTED] : http://www.vcnet.com/bms: ~~~
Re: Mersenne: Problems with ATI graphics
I have the same problem with a 400MHz PII Compaq, I think it has a ATI graphics card. James Escamilla From: novarese @ job.cba.ua.edu on 01/27/99 04:11 PM To: mersenne @ base.com cc:(bcc: James Escamilla/TXN) Subject: Mersenne: Problems with ATI graphics Has anyone experienced probelems with ATI graphics cards? I've been running prime95 on some evaluation machines here and the display gets fuzzy when prime is running. -- PGP Fingerprint 413D 8BEE 88B3 087B 630D 30A4 951A A435 3D01 AFF4 [EMAIL PROTECTED]
Re: Mersenne: Problems with ATI graphics
I think we talked about this a while ago, but it manifested itself in noisy RAM (or was it cache chips???). The conclusion was that it was the extensive heat/EMF given off by a very active processor. Perhaps ATI video cards are more susceptible to interference. Try moving your video card a slot or two away from the CPU. On Wed, 27 Jan 1999, Paul Victor Novarese wrote: Has anyone experienced probelems with ATI graphics cards? I've been running prime95 on some evaluation machines here and the display gets fuzzy when prime is running. -- PGP Fingerprint 413D 8BEE 88B3 087B 630D 30A4 951A A435 3D01 AFF4 [EMAIL PROTECTED] -- ~~~ : WWW: http://www.silverlink.net/poke : Boycott Microsot: : E-Mail: [EMAIL PROTECTED] : http://www.vcnet.com/bms: ~~~
Re: Mersenne: Problems with ATI graphics
On Thu, Jan 28, 1999 at 10:55:39AM -0600, Paul Victor Novarese wrote: Could someone explain this "noisy" cache/RAM? That sounds like it may be the problem. As I recall, someone reported a similar symptom to what you're reporting: i.e. a "noisy" video signal. One possible cause that was advanced on the mailing list was that Mersenne's heavy usage of the FPU or cache might be generating enough interference to cause the problem. More facts for you detectives out there: 1) This is a Dell Optiplex. All three models (slimline, desktop, midtower) exhibit this behavior. They probably all share the same motherboard design. 4) The distributed.net client does not cause this behavior. My understanding is that it does not use either FPU or cache. That's certainly not true. Distributed.net's program _does_ use memory (and hence, the cache). It may use it _less_ than mersenne, but less is not "does not use". That difference could conceivably be enough to cause the behavior you're seeing. Depending on which core you're using, Distributed.net might use MMX (which use the FPU registers, but perhaps not the FPU itself) or solely integer instructions. Do you see the same behavior in all video modes? If you switch to text mode (DOS mode), do you still see it? Could you take a picture of the interference or try to describe it? I'm very curious about this... Here's some speculation: since you're not reporting screen corruption (just picture instability), I'll speculate that the corruption is occurring either between the frame buffer memory (VRAM) and the RAMDAC, or in the analog video signal after it leaves the RAMDAC. The second makes the most sense, since that's an analog signal and is more susceptible to interference than the digital signals from the VRAM. The question in my mind is how the CPU can generate a strong enough signal to interfere with the video signal... -andy -- Andy Isaacson [EMAIL PROTECTED] [EMAIL PROTECTED]Fight Spam, join CAUCE: http://www.csl.mtu.edu/~adisaacs/ http://www.cauce.org/
RE: Mersenne: Problems with ATI graphics
Could someone explain this "noisy" cache/RAM? That sounds like it may be the problem. Since the cache on the PII's is built in, saying it's a cache problem might as well be the same as calling it a CPU problem. More facts for you detectives out there: 1) This is a Dell Optiplex. All three models (slimline, desktop, midtower) exhibit this behavior. I assume the video card is similarly close to the CPU in all those models? 2) I don't think heat is the problem. The problems start within 5-10 seconds of starting prime and stop as soon as prime is halted. Then we can say it's related with some certainty. 3) The on-board ATI graphics chip is about 2 cm away from the Slot1, directly underneath the CPU's big heat sink. That's *way* too close, I would think. 4) The distributed.net client does not cause this behavior. My understanding is that it does not use either FPU or cache. Since you don't see the problem normally, my gut instinct tells me it's related to FPU activity. The cache and integer bits of a PII are used with some regularity even if the machine is just sitting there. One good test would be to use any other program that uses the FPU alot and see if the flicker shows up. Can anyone think of another program that would? I suppose any benchmark that tests FLOPS should do the trick.
RE: Mersenne: Problems with ATI graphics
As I recall, someone reported a similar symptom to what you're reporting: i.e. a "noisy" video signal. One possible cause that was advanced on the mailing list was that Mersenne's heavy usage of the FPU or cache might be generating enough interference to cause the problem. And besides, most anything running on the machine will cause the CPU to use it's cache. That's just unavoidable. I'm very curious about this... Here's some speculation: since you're not reporting screen corruption (just picture instability), I'll speculate that the corruption is occurring either between the frame buffer memory (VRAM) and the RAMDAC, or in the analog video signal after it leaves the RAMDAC. The second makes the most sense, since that's an analog signal and is more susceptible to interference than the digital signals from the VRAM. The question in my mind is how the CPU can generate a strong enough signal to interfere with the video signal... Your basic PII probably consumes 18W or so. That's a guess by the way, don't quote me on that. Some of that gets radiated as heat, some will invariably be EMF. I would hazard a guess that Intel never reckoned on a video card being so darn close to the CPU. Either that, or the ATI graphics card is especially sensitive to EMF. One other consideration might be a slightly higher voltage applied to the CPU. PII's core voltage is somewhere in the neighborhood of 2.1V I think, but again, I'm guessing (I actually *do* have the PII specs, just not right now). If the applied voltage were a bit higher, that could result in an unusually high amount of EMF. Either way...EMF from the CPU definitely appears to be the culprit. One possibility which I emailed the original person was to place some braided copper between the CPU and video card, and tie that to a good ground. That's not practical though. :-) We don't all have copper braiding lying around. Another option is moving the card to another slot which is the BEST idea if it's a PCI card. If it's an AGP card, you can't move it. Oh drat. Remember those old hobby computers from back in the 70's that generated so much EMF you could make a TV flicker from the other side of your house? Or how you could place a radio nearby and run a program that generated so much noise in such a way that you could actually try to play a tune? Oh, for fun... (Okay, I'm not quite old enough to have used a hobby computer in the 70's...but close, so close).
Re: Mersenne: Problems with ATI graphics
I tried this a few months ago on our E series Gateways (ATI Rage Pro) Prime95 LL test - yes interference Prime95 factoring - no interference Distributed Net Client - no interference FPU benchmark (I think it was ZD winbench's FPU test) - yes interference The no FPU benchmark - no interference It's is in all video modes (although harder to see on some) It start as soon as you start pegging the FPU (can be right after boot with the chip still cold to the touch) Stop as soon as the FPU backs off Louis Towles - Original Message - From: Aaron Blosser [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Thursday, January 28, 1999 3:33 PM Subject: RE: Mersenne: Problems with ATI graphics One good test would be to use any other program that uses the FPU alot and see if the flicker shows up. Can anyone think of another program that would? I suppose any benchmark that tests FLOPS should do the trick.
Re: Mersenne: Problems with ATI graphics
Aaron Blosser wrote: Could someone explain this "noisy" cache/RAM? That sounds like it may be the problem. Since the cache on the PII's is built in, saying it's a cache problem might as well be the same as calling it a CPU problem. The cache and integer bits of a PII are used with some regularity even if the machine is just sitting there. One good test would be to use any other program that uses the FPU alot and see if the flicker shows up. Can anyone think of another program that would? I suppose any benchmark that tests FLOPS should do the trick. It might be interesting to run CPUmark 99 from http://www.zdnet.com/zdtv/screensavers/projects/story/0,3656,2175057,00.html Leo
