Seti@Home optimized science apps and information
Optimized Seti@Home apps => Windows => Topic started by: Simon on 16 Feb 2007, 12:31:43 am
-
Hi folks,
Code Revision 2.2 'Noo? No, Ni!' of the KWSN/Lunatics.at SETI@Home optimized apps has been released.
This new revision supersedes the previously released Rev. 2.0 apps for all systems as well as the 1.41 app for Core-2 based systems.
The people responsible for this new version are, in no particular order, Josef W. Segur, Ben Herndon and Alex Kan. A great big "thank you", guys!
Thanks also to the tireless members of the prerelease test community!
Windows optimized apps (static executables, 32/64-bit compatible)
There are various different optimized apps. Please use a program like CPU-Z (http://www.cpuid.org/cpuz.php) to find out exactly what your system supports.
Choose the first, in order, your system supports among the two lists (AMD/Intel or Intel only).
Intel-only apps (do NOT run these on AMD systems!)
Core 2 SSSE3-optimized app (Conroe, Woodcrest, Clovertown, Kentsfield, Merom, but NOT Core Solo/Duo!)
Pentium M/Core Solo SSE2-optimized app
Only use this app on Pentium M-based systems like Dothan, Banias or Core Solo/Duo (NOT Core 2 Duo!).
Pentium 4/Pentium-D SSE3-optimized app
Pentium 4 SSE2-optimized app
AMD and Intel-compatible apps
SSE2-generic optimized app
SSE-optimized app
MMX-optimized app
As always, use these apps at your own risk. A pre-edited app_info.xml has been included with all files. Please unpack the files to a temporary location using 7-Zip (get it from http://www.7-zip.org), then read Instructions.txt in the folder it creates.
An Auto-Installer for these versions will come tomorrow or the day after, along with versions including graphics.
Happy crunching!
Regards,
Simon
-
What are the improvements of the newest release? I've read something about C2D, so isn't there anything new for older machines ?
-
Thanks to all the programmers for their great efforts! I'm already running the SSE client on my AthlonXP, and a patched Intel "only" SSE2 on my A64.
Looking forward to seeing how much it improves my times, but it seems that the AthlonXP likes the new app. a _lot_.
Also looking forward to seing reports about if the Intel SSE2 or SSE3 is the faster one for my A64.
Furex:
As far as remember there are some rather huge improvements for this release.
They found a memory read that was happening every cycle, and found a way to reduce these reads a lot. This is the biggest improvement, and should help _all_ processors. There is also some new SSE code in the release.
The numbers I've been hearing is improvements of 20-30% and maybe even more in total crunching time. I'm not 100% sure about these numbers, my memory can be a little inconsistent ;D , but time will tell us about the improvements.
-
Seeing about a 10-20% improvement over 2.0 so far ;D ;D
Thanks for the time you have put into this.
BoB
-
I had what appears to be a slight improvement on my Dual Core Opteron by using the patched IntelSSE3 over Generic SSE2 (still gathering stats to be true). - edit: I was talking about 2.0 of course
Is there a thread detailing changes in release 2.2 ? I wasn't able to find one
-
Thanks to Simon and the Coop, as always. Another splendid effort.
One tiny wee bugette in CPU-ID - it thinks my Xeon 53xx 'Clovertown' is a Xeon 51xx 'Woodcrest' - but that's not important: it can be tidied up later, or not, as you see fit.
-
I've also found a tiny error in the output of WU's.
This is the features output for my A64 on the Intel SSE2 client.
Features: MMX, 3DNow!, 3DNow!+, SSE, SSE3, SSE3,
Notice that SSE3 seems to be supported twice, and SSE2 is not :)
-
Thanx for the GOOD work! :)
I have the Intel QX6700.
I use the SSSE3 Core2 app.
The <stderr_txt> isn't correct or?
--------------------------------------------------------
Optimized SETI@Home Enhanced application
Optimizers: Ben Herndon, Josef Segur, Alex Kan, Simon Zadra
Version: Windows SSE3 32-bit based on seti V5.15 'Noo? No - Ni!'
Revision: R-2.2|xT|FFT:IPP_SSE3|Ben-Joe
CPUID: Intel Xeon 51xx 'Woodcrest'
CPUs: 1, cores: 4, threads: 1 cache: L1=32K, L2=4096K, L3=0K
Features: MMX, SSE, SSE3, SSE3, SSSE3
speed: 2666 MHz -- read MB/s: L1=9951, L2=8617, RAM=5579
--------------------------------------------------------
Only for information! :)
EDIT:
I downloaded (I have running (Task Manager)) the KWSN_2.2_SSSE3-C2_Ben-Joe.exe... (with the other files)
Or is the link not right for the SSSE3 app.? And I have the SSE3 app.?
Or the name of the/my app. is not rigtht? SSSE3 but is SSE3?
-
What are the improvements of the newest release? I've read something about C2D, so isn't there anything new for older machines ?
credit for wu: 57.17
Pentium3 1.4GHz@1.63GHz:
v1.3_SSE 21700secs
v2.2_SSE 17775secs
improvement appr. -18%
credit for wu: 60.86
Pentium M 2.0GHz@2.4GHz:
v2.0_SSE2_PM 7821secs
v2.2_SSE2_PM 6348secs
improvement appr. -19% (but first wu was validated INVALID (http://setiathome.berkeley.edu/workunit.php?wuid=113662551))
Thanks to all the optimizers over here.
-
Eek!
Seems I made a typo when changing the stats output; the first "SSE3" should be SSE2, instead.
Guess I'll have to recompile the apps again :) Thanks for noticing. Since it's a cosmetic thing only (it doesn't affect any function choices), it won't be a required upgrade.
What are the improvements of the newest release? I've read something about C2D, so isn't there anything new for older machines ?
As for changes vs. 2.0 -
Improved pulse folding
Improved accuracy (especially on Core 2 systems vs. 1.41)
Benchmarking for the various folding versions
Some extra chirp functions adapted from Alex Kan's code (SSE and SSE2, was only SSE3 before)
Benchmark improvements as far as correct function choices go (the app tests each available function for sub-tasks like chirping, pulse folding, etc. up to the supported SSE level and uses the quickest, but did choose incorrectly sometimes, fixed)
Major efficiency improvement by Joe Segur - Not doing transpose when it's not needed
Doing transpose on 4 FFT chunks at a time rather than 1
and some others I probably forgot. Ben and Joe can complete the list or correct it.
How the apps will perform depends a lot on your host. On my Xeon 3.0 HT system, I saw an unbelievable jump of 60%. On my A64, around 15%, on my PD 805, around 20%, same on PD 9xx hosts. Around 10% quicker than 1.41 on Core 2 systems for most WUs, and around 35% quicker for the dreaded 58.7s, if I remember correctly. These are values I've seen in my benchmarks, so they may not reflect your results.
That said, you have my word you'll be pleasantly surprised because it is quicker than 2.0B and 1.41 on ALL hosts I've tested it on.
HTH,
Simon.
-
CPUID: Intel Xeon 51xx 'Woodcrest'
Features: MMX, SSE, SSE3, SSE3, SSSE3
EDIT:
I downloaded (I have running (Task Manager)) the KWSN_2.2_SSSE3-C2_Ben-Joe.exe... (with the other files)
Or is the link not right for the SSSE3 app.? And I have the SSE3 app.?
Or the name of the/my app. is not rigtht? SSSE3 but is SSE3?
You have the correct app, and the stderr output is also correct (except for that small error others pointed out where SSE2 in the features line becomes SSE3, I've coloured it red above, and our CPUID table needs an update for quad-cores).
We have tested a lot of versions on Core 2 based systems. Try as we might, SSSE3 does not offer any usable functions for SETI@Home crunching, so calling the app SSSE3 really is not true as such. However, you can tell the Intel compiler to optimize the program for Core 2 systems, and this does work.
In the end, the SSE3 functions coupled with the Core 2 optimizations produced the fastest crunch times, so that's why we did it this way.
So that's why it says SSE3 on Core 2, and the app is called SSE3-C2.
All's well ;)
Simon.
-
You have the correct app, and the stderr output is also correct (except for that small error others pointed out where SSE2 in the features line becomes SSE3, I've coloured it red above, and our CPUID table needs an update for quad-cores).
We have tested a lot of versions on Core 2 based systems. Try as we might, SSSE3 does not offer any usable functions for SETI@Home crunching, so calling the app SSSE3 really is not true as such. However, you can tell the Intel compiler to optimize the program for Core 2 systems, and this does work.
In the end, the SSE3 functions coupled with the Core 2 optimizations produced the fastest crunch times, so that's why we did it this way.
So that's why it says SSE3 on Core 2, and the app is called SSE3-C2.
All's well ;)
Simon.
Thanx for explaining for the "ignorant" people, like me! ;)
Small suggestion :) :
In the <stderr_txt>:
instead of:
Version: Windows SSE3 32-bit based on seti V5.15 'Noo? No - Ni!'
that:
Version: Windows SSE3-Core2 32-bit based on seti V5.15 'Noo? No - Ni!'
And the name of the app:
instead of:
KWSN_2.2_SSSE3-C2_Ben-Joe.exe
that:
KWSN_2.2_SSE3-Core2_Ben-Joe.exe
That the people, like me, know that they have the correct app. ... ;)
-
Thing is,
do you put it so it's correct or do you put it so people can more easily identify it? ;) It's always a trade-off.
In any case, an auto-installer package should come up shortly, as well as some recompiles with rectified stderr output.
Tsk, I really need more sleep :)
Regards,
Simon.
-
Simon:
I had an idea it was only an cosmetic error, and found it a little funny that my CPU had 2xSSE3 (actually it has, since its dual core) :-)
Something else for anybody who is interested:
I have done the quite tiresome job of making all the new applications work with the previous auto-installer and tester (a lot of patching and renaming...), to find out which application does best on my A64 (CPUID: AMD Athlon 64 X2 'Toledo'). Its an Socket 939 Athlon64 X2 3800+ clocked at 2475Mhz.
After patching all the Intel "only" versions, and renaming them and the generic version to the correct old application names, I let the program run a medium size test.
The result was this.
Patched Intel "only" SSE3-P4: 214 seconds.
Patched Intel "only" SSE2-P4: 212 seconds.
Patched Intel "only" SSE2-PM: 205 seconds.
Generic SSE2: 212 seconds.
It seems that the fastest version for my A64 this time is the SSE2-PM, and it seems to be quite a lot faster.
More testing has to be done, but until then, I'm running the SSE2-PM version, and letting it stretch its legs.
-
Thanks for those results, Karsten!
Quite interesting. I'd say that since the Pentium M has a short pipeline like the Athlon64s do, that may be the deciding factor for the speedup you're seeing.
I believe so far people have only patched the P4-SSEx versions. Good idea there.
Let us know how it goes!
Simon.
-
You're welcome Simon.
You're probably right about the pipeline, it could be significant, but then it should have shown in the previous versions also. I know I tested the SSE2-PM with rev. 2.0, and it wasn't faster at that time (but not much slower either).
But many factors have changed, and even the new code-changes, (who as I understand it, put less pressure on the L2 cache and memory), could make the PM-version perform better on A64.
It could also be a fluke, but a 7 seconds difference on 212 seconds run-times is a big variation. I'm going to run the long test tomorrow, to see if the results are the same.
While I'm here I will just mention, that on my old AthlonXP Thoroughbread at 1936Mhz, with 256Kb cache, I'm seeing a 25+ % improvement on 62 points WU's. An extremely impressive result!
-
Thanks to Ben, Joe and Alex, it is indeed impressive :)
The next app revision will probably deal with the new 5.18 code, and will offer different challenges.
Regards,
Simon.
-
As for changes vs. 2.0 - (list follows)
Thank you Simon! :)
I'm testing GenSSE2 and it seems faster than patched iSSE3. I'm doing my tests on a batch of long WUs (~62.4) I've retrieved some time ago. After the interesting findings by KarVi I'll probably end up doing more tests on patched R2.2 apps to see whether the gains on the short benchmark units show up in real world crunching, too.
Hope this release does something also for some classes of shorter units which turned out to be much slower (up to 40-50%) than the longer ones.
-
New results.
Running KWSN Test & Benchmark tool, with patched and renamed Rev. 2.2 applications, in long test mode, on my Athlon64.
Patched Intel "only" SSE3-P4 Rev. 2.2: 397 seconds.
Patched Intel "only" SSE2-P4 Rev. 2.2: 387 seconds.
Patched Intel "only" SSE2-PM Rev. 2.2: 381 seconds.
Generic SSE2 Rev. 2.2: 395 seconds.
The results seem to be conclusive:
For my processor, the patched SSE2-PM is the fastest client.
-
New results.
Running KWSN Test & Benchmark tool, with patched and renamed Rev. 2.2 applications, in long test mode, on my Athlon64.
Patched Intel "only" SSE3-P4 Rev. 2.2: 397 seconds.
Patched Intel "only" SSE2-P4 Rev. 2.2: 387 seconds.
Patched Intel "only" SSE2-PM Rev. 2.2: 381 seconds.
Generic SSE2 Rev. 2.2: 395 seconds.
The results seem to be conclusive:
For my processor, the patched SSE2-PM is the fastest client.
KarVi,
Could you possibly PM me the patched clients to test on my FX60 rig?
-
I don't think its possible to PM files to each other? I haven't found a way to do it.
Instead i will send the files to the e-mail address listed in your profile.
-
Those intrepid testers have found the truth of it. The Pentium-M compiled & patched version will be the fastest on AMD X2 chips. How far into the past this applies I do not know (versions of AMD chips). Pentium M has SSE2 instructions, so it certainly wouldn't work on an SSE only AMD chip.
Simon, can you confirm if, when using the 2.2 version, on a system where the "-bench" command line shows that one of the AK chirping routines is chosen...that it in fact does use 32Meg less memory when running (like it was written to do)?
Regarding the CPUID table, I may have enough info to update that, but really I should make it an external table. Maybe I will do that this week. Then we can post the table for those who want to download it to their systems.
-
Hi Ben,
I'll get that info for you, stat ;)
In fact, while looking at the task manager, I saw that it was using a ridiculously low amount of memory, around 27MB I think. My first speculation was that since the system only had 256MB RAM, not enough may be free, but it had ~90M free mem (Win2K). So I'd think it uses less RAM. This was on an Athlon XP.
On my PD, both running tasks use about 30MB. So yes, for me, when AK chirps are used, it runs with a significantly smaller memory footprint.
Good idea on the external CPUID table. By the way, I've been trying to convince people on the BOINC dev list to incorporate CPUID, but Rom's against it; he's complaining about exactly that, keeping the CPUID table current.
In my opinion, that's a bit weak, since new CPU revisions don't exactly come out every week; once a year, sometimes twice per manufacturer is what it amounts to.
I'm going to pursue this further, if only to finally get the cache size detection in BOINC on Windows to work correctly. He's got to be amenable to fixing that ;) I've looked at the code, it seems that the cache size detection alone could easily be ported on its own, right?
The current (BOINC 5.8.x) system of listing all the flags in []'s isn't the bee's knees, as far as I'm concerned, I'd prefer the CPUID computed string as per your version, but that's just me. Also, on Windows it only goes up to SSE2, even if higher SIMD levels are supported.
Regards,
Simon.
-
As for changes vs. 2.0 -
Improved pulse folding
Improved accuracy (especially on Core 2 systems vs. 1.41)
Benchmarking for the various folding versions
Some extra chirp functions adapted from Alex Kan's code (SSE and SSE2, was only SSE3 before)
Benchmark improvements as far as correct function choices go (the app tests each available function for sub-tasks like chirping, pulse folding, etc. up to the supported SSE level and uses the quickest, but did choose incorrectly sometimes, fixed)
Major efficiency improvement by Joe Segur - Not doing transpose when it's not needed
Doing transpose on 4 FFT chunks at a time rather than 1
and some others I probably forgot. Ben and Joe can complete the list or correct it.
...and so it happened, Ben added one thing I forgot: when any of Alex Kan's chirp routines are used, the app now uses significantly less memory. On my systems, between 27 and 29 MB per running process, compared to around 65-70 before.
Rather impressive, considering that it's also quicker ;)
To make further development and especially my task of re-synching it simpler, I'm posting a source archive of my current sources used to compile the 2.2 apps. Please use this as a base for further work, you'll make my life that much easier guys!
Regards,
Simon.
-
Hi, everyone
Call me what you will, and I like looking at the pretty pictures (I find it soothing). And, compared to every machine I've ever owned previously, I have CPU capacity to "burn", anyway.
The thing is, this leaves me left out, as far as the 2.2 app goes, as I don't see a graphics version. No big deal, really - 2.0 is probably a significant improvement over the standard version.
A couple of things I've read recently intrigued me. One was that modifications were "easier" on the cache (or the like). My A64 3700+ "San Diego" has a 512k L2 cache. Is that really needed, or is my ignorance showing again? Yes, I do know that the L1 cache is kind of small (64k + 64k), and that there are a number of CPUs with much greater cache capacity than mine.
Anyway, the other intriguing thing I saw was that A64 x2 CPUs are architecturally closest to Pentium M (short pipeline). I've seen it theorized in more than one place on the web that, given the similarities to dual core A64 CPUs, that perhaps the "San Diego" cores really ARE dual core chips, with one of the cores disabled because it failed QC. Does this have any implications on which variation might work best on my CPU?
Now, I've done some very low level coding in my time (including hand-assembling 8085 and Z80 code, to date myself), and the chances of me finding the time to do any of my own SETI patching/compiling any time soon are basically nil.
That being said, I'm willing to take the time to do some testing/benchmarking on my particular configuration, especially if that means I might end up with something faster that still does graphics. If someone can compile something that might work better than 2.0 on my system, I'm willing to try it out and see what happens.
Otherwise I'll just stick with what I have.
I also wonder if the performance hit caused by running graphics varies depending on your video card. While mine is nothing to write home about, it is reasonably fast.
One more thing - the auto-configure setup failed on my system. If anyone is interested in knowing exactly how, I can certainly run it again so I can report the specific error (I can even provide the data from CPUZ, if that might be helpful). I didn't see an AMD-friendly SSE3 (even though I'm pretty sure my CPU has it), so I opted to manually install the 2.0 SSE2 version instead (which is running just fine, though I don't know how to determine what the performance increase, if any, was).
-
Hi Lloyd,
as noted in the announcement, versions with graphics are in the works. In fact, I've been compiling them for the past 2 hours or so. Halfway done, and they will get released when they're done "cooking" ;)
Regards,
Simon.
-
Hi Optimizers,
what is the rate of invalid results with rev.2.2?
I had 2 in two days. That feels quite not alright if you had maybe 1 in 1000 with rev.1.3 before. I keep an eye on the further results and hope the best. The improvement is great. I remember having nearly the same times with an optimized pre-enhanced application.
-
I don't know if my results are representative, but out of the 44 results my two machines have reported back after the switch to rev. 2.2, none (zero) have been marked invalid.
-
I don't know if my results are representative, but out of the 44 results my two machines have reported back after the switch to rev. 2.2, none (zero) have been marked invalid.
Hi KarVi,
maybe i had just bad luck and the next 2000 wus are valid.
-
Just did a full benchmark run using patched apps on my FX60. Confirmed what I think you guys already figured out. The SSE2-PM patched app is the winner.
Full test results attached if they are of any interest.
[attachment deleted by admin]
-
that it in fact does use 32Meg less memory when running (like it was written to do)?
http://www.progettosi.dyndns.org/forum/Smileys/smiley_alive/worshiper.gif
-
I see the 2.2l with graphics app is up now. Thanks for letting me know it was in the works.
In the meanwhile, I found the benchmarks (in the download section - duh!) I'll just leave 2.0 on there for now and run them, then repeat the process with 2.2
You aren't kidding that it's a slow process to run them all! My first try resulted in a bunch of error messages that the app couldn't find the SAH file (?!). I'm trying a couple of things differently (with a fresh download), so I hope it will work this time.
-
Here is my first computation error with rev.2.2: a 0xc0000005 (http://setiathome.berkeley.edu/result.php?resultid=478310853) !
Just one thing: the KWSN_2.2_SSE_Ben-Joe.pdb file that came with the executable is not used in the Windows Runtime Debugger. What to do to have it used next time ?
[attachment deleted by admin]
-
With the (SSSE3-Core2) Rev 1.41 it was sometimes that I had invalid results...
(When I remember correct, this were only with "normal/good" crunching WUs)
Since I have the (SSSE3-Core2) Rev 2.2 running I have two invalid results...
But "only":
SETI@Home Informational message -9 result_overflow
NOTE: The number of results detected exceeds the storage space allocated.
...results.
O.K., I got a few "overflow"- WUs, so around 30.
So it's "only" ~7 % mistakes... ;)
So the mistakes like with Rev 1.41 are away that sometimes the good crunched WUs don't get Credits?
But the new app have now maybe sometimes problems with the "overflow"- WUs?
Hey! It's not a big problem! :)
I wanted only to inform you! ;)
-
Hmm...maybe Joe's fixes increased the accuracy too much :o
No..really...if Joe's fixes give more accurate results and more precision on some sin/cos & angle results, then it might cause mis-compares with the standard application. The more accurate reusult would be better science, but would fail in validating against other clients.
-
To make further development and especially my task of re-synching it simpler, I'm posting a source archive of my current sources used to compile the 2.2 apps. Please use this as a base for further work, you'll make my life that much easier guys!
Simon,
I was looking around on the downloads area (both pre-release and regular) and didn't find any 2.2 source package.
-
Hi.
I have a problem with this new app, that I didn't have before with the 2.0. It takes too much resources, making my computer slow, and worse, I can't play a video or a videogame without having huge slowings. It seems that I'm not the only one (I'm from the hardware.fr forum and the Alliance Francophone). I tried to set my programs in realtime priority but it didn't change a thing.
I'm going back to 2.0
My configuration :
Intel E6600@3.4GHz watercooled
2*512Mo GEIL PC6400 C4
Asus P5B
Windows XP 32bit SP2
BOINC 5.4.11
Thanks for your comments.
-
Asmodeus,
when the 2.2 apps ran, you said they used too much of your system resources.
Which exactly do you mean? On your system, the 2.2 apps should use quite a lot less RAM (~30M vs. ~65M) per running task. As far as slowing the rest of your computer down, that hasn't happened to me yet. Still, I don't crunch while gaming (not at all on my laptop), so your experience may differ.
All my systems running 2.2 are as responsive as they were before, no difference for me.
Please, some more detail!
Regards,
Simon.
-
To make further development and especially my task of re-synching it simpler, I'm posting a source archive of my current sources used to compile the 2.2 apps. Please use this as a base for further work, you'll make my life that much easier guys!
Simon,
I was looking around on the downloads area (both pre-release and regular) and didn't find any 2.2 source package.
Hi Ben,
that would be because there isn't one online until now. Problem being, I'm in Vienna, and my compile system, the PD (having the current sources) isn't accessible anymore. Must have bluescreened on me or whatever. So it'll take me til the weekend to come up with a current source archive.
HTH,
Simon.
-
Hmm...maybe Joe's fixes increased the accuracy too much :o
No..really...if Joe's fixes give more accurate results and more precision on some sin/cos & angle results, then it might cause mis-compares with the standard application. The more accurate reusult would be better science, but would fail in validating against other clients.
Now I have with the "normal/good" WUs invalid results like with the "overflow"- WUs too... :(
If I understand you right, then it will be better to have an application what make not so correct results... but then I have better chances for valid results (Credits)?
Where I can get them? ;)
(If I remember right, in the time I used Crunch3rs apps I didn't had invalid results...
Or I didn't find them? ;)
Maybe, what he had done in other way?)
-
Simon,
you changed the application to Rev. 2.2A...?!
...you didn't said it! ;)
You changed the Features (yellow).
If you rename/extend the Version-name for the SSE3-Core2 app. and update the CPUID for the QX6700 (the red entries) too, I will be endless happy! ;)
When you have time! :)
-------------------------------------------------------------
Optimized SETI@Home Enhanced application
Optimizers: Ben Herndon, Josef Segur, Alex Kan, Simon Zadra
Version: Windows SSE3-Core2 32-bit based on seti V5.15 'Noo? No - Ni!'
Revision: R-2.2|xT|FFT:IPP_SSE3|Ben-Joe
CPUID: Intel Xeon 51xx 'Woodcrest'
CPUs: 1, cores: 4, threads: 1 cache: L1=32K, L2=4096K, L3=0K
Features: MMX, SSE, SSE2, SSE3, SSSE3
speed: 2666 MHz -- read MB/s: L1=9939, L2=8501, RAM=5525
-------------------------------------------------------------
-
Asmodeus,
when the 2.2 apps ran, you said they used too much of your system resources.
Which exactly do you mean? On your system, the 2.2 apps should use quite a lot less RAM (~30M vs. ~65M) per running task. As far as slowing the rest of your computer down, that hasn't happened to me yet. Still, I don't crunch while gaming (not at all on my laptop), so your experience may differ.
All my systems running 2.2 are as responsive as they were before, no difference for me.
Please, some more detail!
Regards,
Simon.
It's not measurable (by definition BOINC use all the CPU time available), and it's not a problem of memory used. It is more like the priority isn't managed as it would be normally. Now there are more people with the same problem on my forum, they're affected no matter the program they use, even winamp. An other example that I can describe is when I launch BOINC after having stopped it, to watch a movie for example : for a few seconds my mouse is very very slow, it jumps from a point to another without the intermediate positions.
-
Simon:
I have to report I have made the same observation as Asmodeus on my system.
Specifically when a new WU starts crunching, my mouse slows down, and moves jerkingly (can one say that) over the screen. If i listen to music with e.g. Winamp, or watch a movie with any movieplayer (BSplayer, VLC, WMP), the movie is interrupted for a few seconds, or it skips frames. ???
Observing this, I have found that it matches exactly with BOINC starting a new WU. For the first few seconds of the WU everything is slowed (even typing in text) down, and then it normalizes. I guess its during the pre-WU benchmark that is done each time a WU starts.
-
KarVi,
that sounds logical. However, 2.0B already did this as well, so I'm surprised it's causing more of a problem now than it was before. Still, something to look into.
As far as increasing rev number to 2.2A, yes, I did say I'd fix it and put updated archives in. I'm just not about to make a big fuss about cosmetical errors, and writing announcement threads etc...gets tiring ;)
I can't fix the CPUID table, that's something Ben will do probably. Since I can't give him a source archive til the weekend (check a few posts above), that may also still take some time.
Regards,
Simon.
-
Simon:
I didn't experience the problem with 2.0 apps., but its very noticeable now.
I wonder if any changes where made to the benchmark code? I recall something about something being changed, to make benchmarks more precise, so that it would pick the right crunching-code more often.
Its something that needs looked into, I think, since it gets annoying very fast.
I'm sticking with 2.2 for now, would hate to loose the speedup ;)
-
There are more routines added to the benchmarking suite for 2.2 vs 2.0. Notably timing & choosing which folding routine and choosing which transpose routine will be used.
When the program is running these tests it tries to make sure it gets accurate numbers on how well each of the routines will perform when crunching an entire WU. It only has access to a counter to tell it how long each version required, however in modern operating systems, because of multi-tasking, the O/S might freeze Seti and then begin running a media player - word processor or whatever else needs to use some CPU time. If an interruption occurs the timing of a given routine will be completely wrong (time = time of rotuine + time for O/S switch + other program running + coming back).
Therefore, the benchmark routines, just before they start timing a function version, raise the task priority very high, and just after the routine timing it restores task priority to where it was (usually the IDLE_PRIORITY level).
On most higher end systems (pentium D and beyond X2, etc), the benchmark testing should only require around 3-4 seconds of time, with a quad core, perhaps more because of memory contention.
During this time the computer might seem unresponsive. The priority setting and benchmarking occurred in the 2.0 version also, however a few more routines are timed in 2.2.
Note: Each individual pass/test turns priority high/low, and the the slowest test is about 1/50th of a second on an AMD 3800+ X2. Every low priority time dip between tests allows the O/S to do whatever switching is needed.
-
Well that explains it :)
IMHO its something that needs fixing, since its has become very apparent in the newest version.
Don't really know how it could be done though, as i understand the reasoning behind raising the priority.
Running the test at just normal priority would remove the problem on most systems I think, but would also give the risk of wrong measurements.
Its a trade-off: Does one want absolutely accurate measurements, or an app. that doesn't interfere with normal operation.
I vote for the last, but I will let it be up to you guys to decide.
-
Being already used to see lower memory requirements of R2.2, this one caught my eye:
(http://img254.imageshack.us/img254/5484/kwsn22memusageip6.png)
-
That would mean that for this particular WU, it chose chirping functions other than Alex Kan's.
Strange, since if supported those functions should almost always be used (and definitely always if they've been chosen before).
Is this host being used exclusively for BOINC, or do you do work alongside?
Regards,
Simon.
-
Is this host being used exclusively for BOINC, or do you do work alongside?
It only crunches in its idle time; do I sense an explanation coming? :)
-
No,
just trying to lower the amount of variables.
-
Well that explains it :)
IMHO its something that needs fixing, since its has become very apparent in the newest version.
Don't really know how it could be done though, as i understand the reasoning behind raising the priority.
Running the test at just normal priority would remove the problem on most systems I think, but would also give the risk of wrong measurements.
Its a trade-off: Does one want absolutely accurate measurements, or an app. that doesn't interfere with normal operation.
I vote for the last, but I will let it be up to you guys to decide.
I agree. If I understand it correctly, doing what you suggest will slow the crunching of a WU from time to time, it does not seem a big deal. I just had a serie of around 20 WU crunshed very quickly (5-6 seconds each) and it made my computer unusable for as long as a minute.
-
Now I have with the "normal/good" WUs invalid results like with the "overflow"- WUs too... :(
If I understand you right, then it will be better to have an application what make not so correct results... but then I have better chances for valid results (Credits)?
Where I can get them? ;)
(If I remember right, in the time I used Crunch3rs apps I didn't had invalid results...
Or I didn't find them? ;)
Maybe, what he had done in other way?)
I found my coding mistake a few hours ago and apologize for any anxiety it has caused. I had seen some indication of the problem before release but didn't expect it to have much impact, so that bad judgement compounded the original mistake.
Joe
-
I found my coding mistake a few hours ago and apologize for any anxiety it has caused. I had seen some indication of the problem before release but didn't expect it to have much impact, so that bad judgement compounded the original mistake.
Joe
Don't worry, at least you found your mistake.
Here are some more error reports attached occurring to my dual P3 yesterday. (0xc0000005) Hopefully that flaw can be found someday.
[attachment deleted by admin]
-
...
Here are some more error reports attached occurring to my dual P3 yesterday. (0xc0000005) Hopefully that flaw can be found someday.
That's a puzzle. For these 3, the callstack indicates it's happening while the app is putting its identifying information in stderr.txt, specifically the line in red:
fprintf( stderr, "Optimized SETI@Home Enhanced application\n\n" );
fprintf( stderr, "%9s: Ben Herndon, Josef Segur, Alex Kan, Simon Zadra\n", "Optimizers" );
fprintf( stderr, "%9s: %s %s %s-bit based on seti V%d.%2d 'Ni!'\n", "Version"
,_OS_, _fft_simd_, _bits_
, gmajor_version, gminor_version );
fprintf( stderr, "%9s: (R-%s|%s)\n", "Rev", _release_, _compOps_ );
How it got to trying to execute non-existent code at 0x0055B8E3 I don't know.
The other one you posted is even more a puzzle, the callstack indicated it was executing a tail section which should only be used if the WU has a num_samples not evenly divisible by the fft length.
[Edit: The above is wrong, I managed to look at the wrong disassembly listing.]
Joe
-
Don't worry Joe, Urs is probably super over clocking his Pentium III. ::)
(just kidding for those whose sarcasm sensors don't function)
-
That's a puzzle. For these 3, the callstack indicates it's happening while the app is putting its identifying information in stderr.txt, specifically the line in red:
fprintf( stderr, "Optimized SETI@Home Enhanced application\n\n" );
fprintf( stderr, "%9s: Ben Herndon, Josef Segur, Alex Kan, Simon Zadra\n", "Optimizers" );
fprintf( stderr, "%9s: %s %s %s-bit based on seti V%d.%2d 'Ni!'\n", "Version"
,_OS_, _fft_simd_, _bits_
, gmajor_version, gminor_version );
fprintf( stderr, "%9s: (R-%s|%s)\n", "Rev", _release_, _compOps_ );
How it got to trying to execute non-existent code at 0x0055B8E3 I don't know.
The other one you posted is even more a puzzle, the callstack indicated it was executing a tail section which should only be used if the WU has a num_samples not evenly divisible by the fft length.
Joe
Hi Joe,
great you were able to spot the related source-places.
Maybe the fprintf C-function is not safe (stack buffer overrun is possible?). That is one of the other things i did change in my private build at seti beta using instead streaming to stderr, e.g.:
std::cerr<<"Some text"<<std::endl;
Don't worry Joe, Urs is probably super over clocking his Pentium III.
(just kidding for those whose sarcasm sensors don't function)
Hi Ben,
sarcasm noted. That machine has NO overclocking possibilities! (-> Gigabyte 6BXDU)
P.S. Attached another one.
[attachment deleted by admin]
-
...
Here are some more error reports attached occurring to my dual P3 yesterday. (0xc0000005) Hopefully that flaw can be found someday.
For my previous reply I was looking at a disassembly of a test version rather than the fresh disassembly I did of SSE 2.2. Just clumsy, I guess.
With the right file, all but the first error are in the setup for testing pulse folding, this code:
// At some angle ranges the total number of test folds may be too few to get good
// timing measurements. Scale up so there are at least 8K test folds.
while (NumPlans < 8192) {
for (iL = 0; iL < 32; iL++) {
FFTtbl[iL][1] *= 2;
FFTtbl[iL][2] *= 2;
FFTtbl[iL][3] *= 2;
}
NumPlans *= 2;
}
The Intel compiler unrolls the for loop completely, the FFTtbl[32][5] array is integers and on the stack, so the generated code loads a value from the stack into esi, adds esi to itself, and stores it back on the stack. The specific one at 0x0055B8E3 is "mov [ebp-44h], esi". But the dump info says "write attempt to address 0x0013D20B" and "ebp=0012e6f8". I have no theories on what's really happening, nor why it's only affecting your dual PIII system. But it should only happen at high angle ranges where that scaling is needed (the two which remained in your results list an hour ago had ar=1.948616 and ar=7.924980 from other host reports).
For the first "Access Violation (0xc0000005) at address 0x0056A89F write attempt to address 0x02239040" you posted, the instruction is "movntps [esi],xmm3". That's the first store for a 4x4 transpose function which writes to the transposed PowerSpectrum array. Transposing is used heavily, so it seems like a one-time glitch unless the system has been choosing one of the other transpose routines since then.
If you would put a copy of the app and a WU renamed work_unit.sah in a temporary directory and execute the app with a -bench command line parameter, it would produce a stderr.txt showing the testing of routines and which ones are chosen. It then exits without actually crunching the WU, so should take less than a minute. I'll attach a README.txt which has all the command line options.
Joe
[attachment deleted by admin]
-
If you would put a copy of the app and a WU renamed work_unit.sah in a temporary directory and execute the app with a -bench command line parameter, it would produce a stderr.txt showing the testing of routines and which ones are chosen...
Joe
Hi Joe,
here you are. Took just a few seconds. Hope it helps somehow.
[attachment deleted by admin]
-
Hi folks,
I'm currently compiling Rev-2.2B apps with Joe's fix. In the meantime, I've uploaded a current source archive (http://lunatics.at/index.php?module=Downloads;sa=dlview;id=71) (including this fix).
Regards,
Simon.
-
Thanks Simon, hope you had a safe trip, wish I could have gone...LOL
-
Well, an empty Autobahn at night is fun, I usually make the ~190 mile trip in about 2 hours. (~150 km/h average, about 94 mp/h or so.)
Don't tell the fuzz, now ;) Only Germany has no speed limits on some parts, here it's 130 km/h ~(81 mp/h).
-
I had a 67 VW Bug when I lived in Germany...It went slow uphill and a bit faster downhill...Made it as far as Innsbrook and just loved it. I think 190KMH was the fastest I ever went in a freinds mercedes. Where I live the freeways are always clogged, even at night. Seattle area.
-
Picked up a 545i at the delivery center in Munich in 05' and enjoyed a little time on the autobahn on my way to visit friends in Neu Ulm. At it's governed max speed of 155, it was smooth as silk and stable as hell. In my younger and less responsible days, I drove more than a few classic muscle cars at 120-130-140 mph that were on the verge of shaking apart, not to mention the VERY unstable steering and downright scary braking (drum brakes anyone?)
The 545i was (is) a most civilized experience. Only a couple of minutes into my "cruise", I saw "blinking" headlights appear quite rapidly behind me. Moving aside to the right, it was barely in time as a new 911 Turbo blew past me like I was standing still. Maybe 185-200ish? Porsche appears to be exempt from the vehicle speed governing ;)
Always owned GM vehicles in the past, but I'm "sold" on German now. Amazing engineering and experience, IMO.
Ok.. Back to topic.
-
Lol. I'd have to agree though...german cars are built with the Autobahn in mind; and yes, going at 155 mph doesn't mean that you'll not get overtaken like you were standing still. I've reached ~195 mph with a motorcycle (myself) and ~180 with a car (though as a passenger) according to their speedometers. I'll admit it, I'm a bit of a speed freak ;) Well not just top speed, but racing, really.
Back to topic ;) New apps with Joe's fix have been compiled, I'll upload them today.
Happy crunching!
Simon.
-
... going at 155 mph doesn't mean that you'll not get overtaken like you were standing still. I've reached ~195 mph with a motorcycle (myself) and ~180 with a car (though as a passenger) according to their speedometers...
Simon, you Speedy Gonsales, did you really mean mph (pretty rare rating on German Autobahns, but my hat down if really) or actually km/h?
Peter
-
Yes, I really meant mph...I did mention I was a bit of a speed freak ;)
195 mp/h would be ~312 km/h, which is what the speedo said when I finally bothered to check it. Didn't feel that quick til I knew how fast I was going. This was on a souped-up 900 ccm racer with ~190 hp, good brakes, too...
As for in a car, a Ferrari F355 of a friend (which he since sadly totalled) on the way to Munich did 290 by the speedo.
Never really can trust them though, they're usually a bit on the quick side.
When I figured out I wanted to be a racer, I was too old to really go anywhere big without investing truckloads of money; so it remains a hobby. Beat my Need For Speed (computer racing game, any version) times though, I dare ya ;)
Regards,
Simon.
-
Your not supposed to say 'SPEED FREAK' You were using the available equipment to it's full potentail! That was my excuse for 180 mph in a Konig Ferrari when they found me a few days later. Mind you the overall journey was a bit on the slow side with 4 fuel stops in 520 miles Ahhhhhhh!
Barry.
-
Hi!
Simon and/or the "optimizer team" can you look here please:
"Probs with Simons S@H V5.15 Rev.2.2A and B"
http://setiathome.berkeley.edu/forum_thread.php?id=37974
Friendly greets!
-
Hi!
Simon and/or the "optimizer team" can you look here please:
"Probs with Simons S@H V5.15 Rev.2.2A and B"
http://setiathome.berkeley.edu/forum_thread.php?id=37974
Friendly greets!
I'm not really competent to comment on C2 systems, but my guess is that when you can get good temperature readings they'll be rather high. The improvements in 2.2 mean there are fewer places where the app has to wait for memory IO, so the CPU doesn't have those stalls to cool down.
If you were running BOINC 5.8.x, you could try the CPU throttling feature. Perhaps slowing down by 5% or so would be enough to stabilize the system.
Joe
-
Yes, I really meant mph...I did mention I was a bit of a speed freak ;)
195 mp/h would be ~312 km/h, which is what the speedo said when I finally bothered to check it. Didn't feel that quick til I knew how fast I was going. This was on a souped-up 900 ccm racer with ~190 hp, good brakes, too...
As for in a car, a Ferrari F355 of a friend (which he since sadly totalled) on the way to Munich did 290 by the speedo.
Never really can trust them though, they're usually a bit on the quick side.
When I figured out I wanted to be a racer, I was too old to really go anywhere big without investing truckloads of money; so it remains a hobby. Beat my Need For Speed (computer racing game, any version) times though, I dare ya ;)
Regards,
Simon.
in Slovakia there are speed limits so there is only one place where i obey speed limits : CPU and GPU clocks speeds ...
-
Hi!
Simon and/or the "optimizer team" can you look here please:
"Probs with Simons S@H V5.15 Rev.2.2A and B"
http://setiathome.berkeley.edu/forum_thread.php?id=37974
Friendly greets!
I'm not really competent to comment on C2 systems, but my guess is that when you can get good temperature readings they'll be rather high. The improvements in 2.2 mean there are fewer places where the app has to wait for memory IO, so the CPU doesn't have those stalls to cool down.
If you were running BOINC 5.8.x, you could try the CPU throttling feature. Perhaps slowing down by 5% or so would be enough to stabilize the system.
Joe
I don't have a problem with the temp. ;-)
This was "only" a question at the side... :-)
(BTW. The probs (bug) with Rev.1.41 (SSSE3-app) and Rev.2.2(A) were the same?) (invalid results)
I posted ("main"):
I have problems with my QX6700 and the Revs. 2.2A & B.
With Rev. 1.41 and 2.2 I didn't had this problems...
I use WinXP Home.
And Boinc V5.3.12.tx36
After ~11h Boinc is frozen. (I had two times this problem after this time!)
The icon is in the "Task border"/"Info. range" shown.
But when I want to click with "right-Mouse" nothing is happen.
In the Task-Manager:
S@H is not running.
boinc.exe use 25% of the whole CPU.
(this mean that 1 complete thread is 100% in use and the 3 other threads are in idle (Quad Core! :-) )
The only thing what I can do is to reboot the system.
Then two messages are shown:
----------------------------------
Close program: boincmgr.exe
The program don't react
(Finish now) (Abort)
----------------------------------
----------------------------------
Close program: boinc.exe
The program don't react
(Finish now) (Abort)
----------------------------------
Have other people the same probs?
What is going wrong?
THANX !
After a short time I posted this:
Now I got the prob after ~3h... :-(
Now I went back to Rev. 1.41... :-(
O.K., with Rev. 1.41 I get sometimes invalid results (and it's little bit slower) , but not so often like with Rev. 2.0...
Now I'm little bit confused...
I have probs now with my new QX6700...? But he isn't OC!
-
Maybe it's because I didn't updated my WinXP Home SP2 over the Microsoft homepage...
This can be the reason?
Or I need .NET Framework?
-
CF,
Deine Situation dürfte ein Zusammenspiel aus mehreren unabhängigen Faktoren sein.
Also, als allererstes empfehle ich Dir mal, die neueste BOINC Version (5.8.15) zu downloaden und zu installieren. Die 5.8er Version gibt's schließlich noch nicht lange, und leider hatte sie schon so einige Probleme. Versionen vor .15 sind definitiv buggy, .15 hat zwar auch noch kleinere Macken, aber funktioniert bei mir fein.
Das sollte mal eine Variable ausschließen und verhindern, daß Du auf einmal nur mehr einen Task laufen hast oder zu viele.
Als nächstes - Deine Temperaturen sind viiiiiiiiel zu hoch. Nicht nur ein bißchen, sondern wirklich viel. Das ist nicht mehr im Bereich "wird schon passen" sondern im Bereich "wird irgendwann bald eingehen" - Dauertemperaturen über 70° machen Deine CPU sicher bald tot.
Was hast Du für 'nen Kühler? Standard-Intel? Hast Du ihn selbst montiert? Die Temperaturen sollten auch damit definitiv nicht über ca. 60° bei Volllast hinausgehen. Mein Rat: besorg Dir z.B. nen ThermalTake Big Typhoon, der kostet nicht die Welt und kühlt hervorragend. Meine wirklich schwer übertaktete CPU, die 3mal so viel Strom frißt (und damit Hitze abgibt) wie Deine CPU, läuft damit ca. 53° unter Volllast.
Wenn Du das nicht tun willst, dann rate ich Dir, Deinen Kühler mal zu demontieren, danach zu reinigen (Isopropyl-Alkohol z.B.) und mit einer DÜNNEN Schicht Kühlpaste zu bestreichen, wo er in Kontakt mit der CPU kommt. Danach neu montieren, und schauen, ob's nicht besser wird.
lG,
Simon.
-
Hallo!
Danke für Antwort!
Wgn. Boinc...
Hab "<cpu_affinity/>" (Trux .tx36) laufen, somit sollte jeder Thread/Core nur ein WU rechnen, oder?
Sollten nicht alle Dual/Quad - PCs diesen client nutzen. Der "original" kann doch nicht diese Funktion, oder?
Ich habe mir den PC beim Handler geholt.
Somit wurde der (boxed) original Kühler mit Lüfter "vom Fachmann" (ich hoffe!) verbaut.
An der Unterseite des Kühlers waren 4 kleine Stellen mit "irgendwas" beklebt.
(Verkäufer meinte: Paste wird nicht benötigt)
Der originale Lüfter läuft von ca. 1.000 und ca. 6.000 UpM. (max. Wert hab ich so im Hinterkopf (hatte ich noch nicht)) ;-)
Bei mir läuft er immer so um 2.000 und 2.500 um die komische "Temperaturanzeige" auf 22°C zu halten.
Ich denke mal das dies die Temperaturspanne bis zur max. erreichbaren Temperatur ist.
Wenn der PC im Leerlauf läuft, ist die "komische" Temp. auf ca. 40°C - 45°C.
(Dann wird die CPU von 2.667 auf 1.6 GHz runtergeregelt)
Hab das neuste BIOS draufgemacht. Vorher war ein Minus vor den Temp.anzeigen...
Ich habe ein Intel D975XBX2- Board.
Wenn die Temperatur zu hoch wäre, oder sogar die max. Temp. erreicht wurde (nahe dran), würde doch der Lüfter auf max. Umdrehungen laufen, oder?
Hast mir jetzt ganz schön viel Angst gemacht! ;-)
Kann es nicht sein, das der QX6700 bis 100°C geht? O.K. dann 100% Tod, aber bis vielleicht bis 90°C noch so lala...?
In CoreTemp steht ja auch als Wert bei "Tjunction" (was auch immer das heißen mag) 100°C.
Der Thermalright SI-128 liegt schon bereit.
Im Handbuch vom Board steht: Der Bereich um die CPU muss auch durch den CPU- Kühler/Lüfter gekühlt werden.
So nach den Tests die ich gesehen hab, ist das dann doch die beste Wahl, oder?
Wie sieht das denn mit der Paste aus.
Die Beste ist nach Tests die Coollaboratory Paste oder das Pad.
Die Handhabung ist mir aber zu unsicher.
Die zweitbeste Lösung ist die "Arctic Silver 5", nach den gelesenen Tests.
Kann eine Paste wirklich so gravierende Unterschiede hervorrufen?
Welche benutzt Du?
Werd mal im Internet noch mal die Tests suchen, wenn mich nicht alles täuscht hab ich da nämlich mal ´n Beitrag gelesen mit den Temps.
Die waren glaub´ ich alle so wie ich sie hab´...
MfG
PS. Jetzt läuft Rev.2.2B seit ca. 14 Std. ohne probs :-)
-
Hab einen Test gefunden:
http://hardware.thgweb.de/2006/11/01/intel_core_2_extreme_qx6700/page14.html
Zalman CNPS-9700 und min. Drehzahl:
Cores bei 80°C.
"Die Kerntemperatur erreicht hierbei 80 °C und ist weit von der Throttling-Temperatur entfernt."
Also ist 80°C für einen QX6700 "normal"?!
Ab wann wird die CPU instabil, bzw. und ist kaputt?
Also ich denk´mal, die CPU wird mit BOXed Kühler/Lüfter auf max. 80°C gehalten.
(Wenn sie heißer wird, wird der Lüfter schneller drehen)
Also ich hab´momentan max. 2.500 UpM gehabt (22°C Raum).
-
Öööh...Tom's hab ich vor etwa 4 Jahren als seriöse Quelle ausgeschlossen. Meine Meinung: die schreiben, was der Sponsor zahlt, nicht was die Daten sagen. Die Aussage, daß 80° weit von der Throttling-Temperatur weg sind, halte ich für fragwürdig, ebenso die Meßmethoden. Das Bild, das dort gepostet ist ist von RightMark CPU Clock Utility, das man sehr wohl dazu anweisen kann, nie zu throttlen (ich verwende das Ding auf meinem Laptop). Also: nur weil da steht, daß das Ding die ganze Zeit mit der gleichen Geschwindigkeit läuft, heißt das ganz genau gar nichts. Der "viele" Text ist auch ein Hinweis ;)
Ich würde Dir empfehlen, selbst einen Vergleich anzustellen. Bau einfach den Thermalright ein.
ALLE von mir gelesenen Reviews, die von Overclockern geschrieben wurden weisen explizit darauf hin, daß Volllast-Temperaturen von jenseits der 60° der Stabilität und Lebensdauer abträglich sind. Jenseits von 70 ist, wie gesagt, weit außerhalb des empfehlenswerten Rahmens.
Hab "<cpu_affinity/>" (Trux .tx36) laufen, somit sollte jeder Thread/Core nur ein WU rechnen, oder?
Sollten nicht alle Dual/Quad - PCs diesen client nutzen. Der "original" kann doch nicht diese Funktion, oder?
Richtig; CPU affinity können nur inoffizielle BOINC-clients bis jetzt. Sollte sich aber ändern.
Ich habe mir den PC beim Handler geholt.
Somit wurde der (boxed) original Kühler mit Lüfter "vom Fachmann" (ich hoffe!) verbaut.
An der Unterseite des Kühlers waren 4 kleine Stellen mit "irgendwas" beklebt.
(Verkäufer meinte: Paste wird nicht benötigt)
Der Verkäufer war kein Fachmann. Man kann natürlich mit den sogenannten "Thermal Pads" (das waren die kleinen beklebten Stellen) arbeiten; allerdings: das vermindert die Oberfläche, an der thermischer Transfer stattfinden kann beträchtlich. Noch dazu: die Thermal Pads sind meist zu dick und haben verglichen mit z.B. Arctic Silver eine wesentlich höhere "thermal resistance", d.h. sie transferieren einfach nicht so viel Hitze.
Also: Du hast mit einem schlechteren Transfermaterial bei einer kleineren Transferfläche angefangen und bist daher derzeit nicht optimal gekühlt. Das mit den 70 Grad Dauerbetrieb war durchaus ernst gemeint; klar verträgt die CPU kurzfristig bis zu 100°, aber Dauerbetrieb ist einfach etwas anderes.
Außerdem: TJunction ist höher als die angezeigte CPU-Temperatur; also wenn Du schon bei 70+ bist wird die TJunction noch 5-12 Grad drüber liegen.
Der originale Lüfter läuft von ca. 1.000 und ca. 6.000 UpM. (max. Wert hab ich so im Hinterkopf (hatte ich noch nicht)) ;-)
Bei mir läuft er immer so um 2.000 und 2.500 um die komische "Temperaturanzeige" auf 22°C zu halten.
Das kann auch andere Gründe haben. Ich kenne den Core2-Standard-Kühler nicht; vermute aber, daß der im Kühlkörper einen Temperatursensor hat und damit die Drehzahl regelt, und nicht über's Board. Nachdem du Thermal Pads benutzt, wird nicht genug Hitze in den Kühler transferiert, der glaubt, es ist eh nicht so heiß und regelt nicht rauf. Nachdem Du eh einen besseren Kühler hast, verbau den.
Zur Schichtdicke der Kühlpaste - am besten trägst Du sowas mit einer Rasierklinge oder etwas ähnlich dünnem auf. Die Schicht sollte so dünn wie möglich sein, ohne Löcher zu bekommen. Also ca. 0.1-0.2mm, auf keinen Fall über 0.5mm.
Ich habe ein Intel D975XBX2- Board.
Wenn die Temperatur zu hoch wäre, oder sogar die max. Temp. erreicht wurde (nahe dran), würde doch der Lüfter auf max. Umdrehungen laufen, oder?
Siehe oben
Hast mir jetzt ganz schön viel Angst gemacht! ;-)
Kann es nicht sein, das der QX6700 bis 100°C geht? O.K. dann 100% Tod, aber bis vielleicht bis 90°C noch so lala...?
In CoreTemp steht ja auch als Wert bei "Tjunction" (was auch immer das heißen mag) 100°C.
Auch siehe oben, nicht ohne Grund!
[...]
Die zweitbeste Lösung ist die "Arctic Silver 5", nach den gelesenen Tests.
Kann eine Paste wirklich so gravierende Unterschiede hervorrufen?
Welche benutzt Du?
Arctic Silver oder ähnliche Produkte. Ich bin nicht der Ansicht, daß die Paste riesige Unterschiede macht (sondern geringe); die Applikation ist entscheidender (siehe oben).
Werd mal im Internet noch mal die Tests suchen, wenn mich nicht alles täuscht hab ich da nämlich mal ´n Beitrag gelesen mit den Temps.
Die waren glaub´ ich alle so wie ich sie hab´...
MfG
PS. Jetzt läuft Rev.2.2B seit ca. 14 Std. ohne probs :-)
Wie gesagt, die Temperaturen von 70+ Grad unter Last sind definitiv weit oberhalb des normalen. Weiß nicht, welche Reviews Du da hast, bitte um URLs ;)
mfg,
Simon
-
Da haben sich unsere posts etwas überschnitten... ;)
Im BIOS kann ich CPU-Fan-Control de-/ u. aktivieren.
Wenn an, passiert das was ich schrieb.
Wenn aus, läuft er (soll er!) auf max. speed (hab´ ich bis jetzt noch nicht versucht, da ich Angst habe, (nach dem oben gelinkten Test) das das Board aus dem Gehäuse fliegt!) ;)
Vielen Dank für Arbeit & Zeit!
Muss mir jetzt nur noch die Paste kaufen gehen... :)
EDIT:
Kurz noch am Rande...
Also wenn ich den Boinc V5.3.12.tx36 laufen lasse (oder gibts einen anderen Grund zum wechseln?) (wenn Du CPU-affinity benötigen würdest, würdest Du ihn nicht auch nehem? ;) )und den Kühler verbaut habe für bessere Kühlung, sollte ich keine Probs mehr haben?
Also Hauptproblem ist die Kühlung bei mir?!
-
Da haben sich unsere posts etwas überschnitten... ;)
Im BIOS kann ich CPU-Fan-Control de-/ u. aktivieren.
Wenn an, passiert das was ich schrieb.
Wenn aus, läuft er (soll er!) auf max. speed (hab´ ich bis jetzt noch nicht versucht, da ich Angst habe, (nach dem oben gelinkten Test) das das Board aus dem Gehäuse fliegt!) ;)
Vielen Dank für Arbeit & Zeit!
Muss mir jetzt nur noch die Paste kaufen gehen... :)
Hi Captain Future,
ich habe eure Diskussion Deines Problems hier verfolgt und habe da einen Hinweis (http://www.intel.com/support/motherboards/desktop/sb/CS-012552.htm) für dich bei Intel gefunden. Demnach sollte Dein Prozessor nicht über 72°C bis 75°C bei voller Leistung betrieben werden. Laut obiger Quelle (die zugegeben noch etwas allgemein gehalten ist) sollte bei diesen Temperaturen in den beiden Mainboardzonen die Drosselung (Throttle) der CPU einsetzen. Vielleicht kannst Du diese Temperatur im BIOS Deiner Hauptplatine irgendwo niedriger einstellen, da bei Dieser Temperatur Deine Lüfter bereits auf vollen Touren laufen sollten.
-
Habe jetzt mal die CPU- Steuerung ausgemacht...
Lüfter dreht jetzt bei ~ 3.780 UpM und verursacht erheblichen Lärm...
(Hoffe die Nachbarn werden nicht gleich vorbeischauen ;) )
CoreTemp:
61°C - 67°C
Intel Desktop Utilities:
30°C
Da denkt man, man kauft ´nen PC vom Händler und dann das...
Da hätte ich ja fast gleich 1.000,- EUR (nur die CPU!) ausm´ Fenster werfen können... >:(
*$!&/($§"&/§"/#?!§"&%$§"!*
Danke nochmals!
-
Gern :)
Deswegen war ich auch so direkt in meiner Aussage, weil das ansonsten teuer wird. Muß ja nicht sein, nä? Danke auch Urs für die Referenz, war zu faul zum selber suchen.
mfg,
Simon
-
Ich wollte eigentlich den Papst 4412 F/2GP auf den Thermalright SI-128 verbauen.
Der hat, wie der original Lüfter, einen 4-pin Board- Anschluss.
Somit wird die Drehzahl nach CPU- Temp. gewählt.
Dann wird die CPU aber auch wieder erst bei 80°C mehr gekühlt, oder? Eben wie gehabt...
(Oder auf max. ~ 80°C gehalten)
Da liegt dann wohl eher ein BIOS- Fehler bei der Lüftersteuerung vor, oder?
Bin auch langsam etwas müde... ;)
-
Das mit der Lüftersteuerung ist etwas mißverständlich; im BIOS solltest Du sie ausschalten. Der Papst Lüfter ist so oder so wesentlich leiser als das Standardmodell, und sollte auch immer laufen.
Das Motherboard wird dann dennoch bei höherer Temperatur den Lüfter schneller ansteuern. Er wird eben auch ohne Last schneller drehen als sonst.
Die Lüftersteuerung ist eher für den Standard-Kühler gedacht, weil der sonst eben doch recht lautstark ist. Mit dem Thermalright sollte das kein Problem sein.
Mfg,
Simon.
-
Hmm... also wenn ich die CPU- Lüftersteuerung ausschalte...
...dann läuft doch der Papst- wie der original- Lüfter auf maximum, oder?
Bei dem Papst heißt das dann: ~170 m3/Std. und >40 dBA
Würd Dir gern´ ´n Link vom Papst- Lüfter geben, momentan ist der aber überall ausverkauft und aus dem Programm aller Händler genommen worden.
=br '>Und das Board wird ihn dann doch mehr bei höherer Temp. ansteuern (bei ausgeschalteter Steuerung?) ?
-
...einfach probieren :) die meisten Papst Lüfter sind recht leise, was Du beschreibst klingt allerdings ziemlich laut.
Wenn die Temperaturen mit dem Thermalright und der Lüftersteuerung an trotzdem passen, dann kannst Du's ja so lassen, sonst auschalten.
Gut's Nächtle ;)
-
Also wie schon einmal gesagt...
Vielen Dank!
Muss jetzt wirklich in die Heia... ;)
Du bist aber auch ´ne "Nachteule", oder?
Dich "sehe&quou; ich meistens Abends/Nachts/Morgens
Du wohnst doch in Österreich und nicht in den USA, oder? ;)
-
Hallo!
Hab´ mal bei Intel angerufen...
Die max. Temp. für den QX6700 ist mit 65°C angegeben! :o
http://processorfinder.intel.com/details.aspx?sS@ec=SL9UL#
Er hat eher vermutet, dass CoreTemp nicht richtig den QX6700 "auslesen" tut...
Wenn 22°C angezeigt wird (Intel Deskop Utilities) ist mit der CPU alles in Ordnung.
Er sagt... 65°C - 22°C = 43°C CPU (Lüfter (-Steuerung) bei 2.000 UpM)
Soll ich das jetzt glauben?
Wenn ich die CPU-Lüfter-Steuerung ausschalte, läuft der Lüfter bei ~3.780 UpM und die Anzeige ist dann bdi 30°C.
D.h. 65 - 30 = 35°C?
Dann wären die Temps ja in Ordnung...
(Alle Werte bei Volllast)
Mal zu meinem "Fachmann" zurück... bei dem ich den PC gekauft hab´...
Seit dem ich den PC hab´, hab´ich folgender Fehler:
Manchmal wenn der PC (meistens) im Leerlauf läuft verschwindet die Farbe im Monitor...
D.h. kurzzeitig ist es schwarz/weis oder wie ein alter "Grün-Monitor" und dann kommt die Farbe wieder.
Mein Händler hat gesagt, kein Problem schicke neue Grafikkarte zu, kann den PC ruhig laufen lassen. Keine Beeinträchtigung des Systems!
Kann dieser Fehler auch den ganzen Schlamassel verursachen?
Nächstes Mal werd ich meinen PC wieder selbst zusammenbauen! >:(
PS. Hatte heute wieder Boinc eingefroren vorgefunden... Bei 30°C (Intel Desktop Utilities)
PC ist jetzt erst mal aus...! :'(
*$&)"§!=)?§+#;$!"!(&%#§"#)$*
EDIT:
Wenn die gemessene Temp. bei CoreTemp wirklich gestimmt hat (71 - 79°C , Variationen zwischen den Cores) wär die CPU doch eh schon "durchgebrannt", oder? Der PC lief ja fast 2,5 Monate bei dieser Temp.! :o
-
Daß er durchbrennt muß nicht sein, hat aber 'ne gute Chance dazu wenn Du das länger machst.
Zu CoreTemp: dem würde ich schon vertrauen. Es gibt eben mehrere Temperaturen, die auslesbar sind pro CPU. Die eine ist die, die das Intel Utility anzeigt, die ist aber nicht direkt auf den Cores gemessen sondern am IHS (Heat Spreader, das ist die Metall-Abdeckplatte).
Deswegen ist die auch geringer. CoreTemp ist eben dazu da, die direkten Hardware-Sensoren des Chips auszulesen, die für jeden Core einzeln die wirkliche aktuelle Temperatur zeigen. Die haben zwar eben eine höhere Maximaltemperatur als 65, aber das tut ihnen trotzdem nicht gut.
Ich empfehle Dir weiterhin, zumindest den Kühler zu tauschen und besonders mit der Kühlpaste anstatt den Pads zu arbeiten.
Mfg,
Simon
-
Hallo Simon!
Hab´ mir Kühlpaste bestellt, sollte morgen bei mir sein...
Neue Grafikkarte auch...
Wenn Karte und Kühler montiert, gibt ´s natürlich einen Bericht! :)
Also bei einem QX6700 sollten die Temps bei CoreTemp pro Core nicht über 65°C gehen...?
Welche Temp. würdest Du halten/erreichen wollen mit dem Kühler/Lüfter, bei Volllast?
45°C?
A small question to the thermal compound...
Some say, a small piece to the heatsink and a small piece to the CPU.
Some say only to the heatsink.
Some say only to the CPU.
I will use Arctic Silver 5 (80% silver) how I must use these?
For example Coollaboratory compound (liquid metal) say, heatsink and CPU...
Friendly greetings!
BTW.
I found an interesting side about compound and life...
http://www.bigbruin.com/2006/chillfactor_1
----->What is of interest to me is that a life expectancy is listed in the specifications. I have believed that paste should be re-applied after a certain amount of time, especially after taking a system apart and seeing the condition of the paste I may have installed just 6-12 months prior. Although some may prefer to believe their paste is good forever since the packaging may not say otherwise, I like that the 12 month life expectancy is provided as an official statement from Thermalright.
A feature that may be seen as a positive or negative by different people is the published life expectancy of 12 months. Many pastes do not include such a date, and people may be under the false impression that the stuff will last forever. Having the date posted on the packaging lets people know that they should be prepared to re-apply the paste in a year. I am fine with this as I rarely have a system stay together for a few months let alone a whole year, but others may find 12 months to be limiting.<-----
So Thermalright say you can use the compound 12 months, or you must take every 12 months new compound to the CPU and heatsink?
It's only now with the new (Thermalright) compound...
Or how you do it with your PCs?
-
I suppose it might depend on the compound. Some apparently can dry out. Some apparently don't reach peak efficiency till have been in use for a few hundred hours.
I probably would swap my thermal compound after a year or so just to be on the safe side and it'd be part of my occasional overall system de-dusting etc.
I don't see any problem though as long as temps are not starting to slowly increase (indicating new compound needed).
Have to clean off the old compound residue first. For that you need a cleaning solution. Couple of the compound makers sell a solution for this. Someone once told me you can use glasses cleaning wipes as long as your careful but I've never wanted to try that. Or even a tiny drop of LCD screen cleaner fluid on a cotton bud but again I've never wanted to try that either.
I've always gone for the small drop of compound on the CPU and spread it out thinly and evenly, then press the heatsink down into place and give it a little bit of movement (move around a tiny bit) just to help the compound spread even thinner and evenly and get into all the tiny surface imperfections. Then clamp the heatsink down.
I prefer putting the compound on the CPU because then your sure it is all going exactly where its suppose to be and you're sure you've got the right amount.
Maybe someone could translate this, and add their better wisdom to it too so more helpful?
-
I will use Arctic Silver 5 (80% silver) how I must use these?
Every TIM must be applied according to the specific instructions supplied by the manufacturer (or the marketer) - For AS5, you can rely on their own detailed instructions (http://www.arcticsilver.com/instructions.htm).
-
Hello!
Now I'm little confused...
With the BOXed heatsink and fan I had temp. with Intel Desktop Utilities 22 °C (Processor Thermal Margin).
Is here somebody who know the usage of the Intel Desktop Utilities?
Because what does this mean= 22°C (Processor Thermal Margin)?
22°C to go to the max. temp.? So 65 °C - 22 °C = 43 °C?
OnBoard FanControl: 2.000 RPM.
CoreTemp: 71- 79 °C
[room= 22 °C]
Now with the Thermalright SI-128 + Papst 4412 F/2GP and Arctic Silver 5:
Desktop Utilities: 25 °C
OnBoard FanControl: 1.500 RPM
CoreTemp: 69- 73 °C
[room= 20,5 °C]
(FanControl OFF: Fan on 2.470 RPM but only ~ 2 °C less temp.)
Now maybe 4 °C less temp. with the SI-128... (CoreTemp)
This can be right?
I used the thermal compound right?
The Motherboard, Intel D975XBX2 have now a "roundness" around the CPU-Socket (if you look from the side to the board), because the holder from the heatsink pulls a lot on the motherboard. It's looking like the CPU is now more far away from the heatsink... (I had explained it well? ;) )
I found pics from a test:
http://pics.computerbase.de/1/5/0/9/8/3.jpg
http://pics.computerbase.de/1/5/0/9/8/2.jpg
It's looking like this.
If you look from the under side (I didn't found a pic), it's looking there is a small mountain direct under the socket.
So I have the fear, that the heatsink are not correct on the CPU (without the correct pressure)(without the direct correct contact)
Friendly greetings!
-
Wow! It looks like the heatsink it bowing the motherboard! I'd be scared to mount it upright in a case!
Have you tried Intel's Thermal Analysis Tool? I am using that with my Bad Axe 2. I downloaded it from Intel's site. I have temps around 50 deg C with the stock heat sink and a little overclocking.
Regards,
Pam
-
Wow! It looks like the heatsink it bowing the motherboard! I'd be scared to mount it upright in a case!
Have you tried Intel's Thermal Analysis Tool? I am using that with my Bad Axe 2. I downloaded it from Intel's site. I have temps around 50 deg C with the stock heat sink and a little overclocking.
Regards,
Pam
Hello!
You have REALLY ~ 50 °C under full load?
I used the Intel's Thermal Analysis Tool and the program shows me the same temps like that CoreTemp program.
Room= 19,5 °C
67- 72 °C the 4 Cores.
You must use CoreTemp too, then you see the 4 cores! ;)
You overclock and the stock heatsink & fan and you have ONLY 50 °C? :o
You use/ed the Intel Desktop Utilities, too?
Which temp you have "Processor Thermal Margin"?
Please can you check this?
I have there 25- 26 °C and the fan is running 1.200 RPM (~ 50 %) (max. speed 2.470 RPM)
(Thermalright SI-128 + Papst 4412 F/2GP)
In the past with BOXed heatsink & fan I had room: 22 °C and the Cores: 71- 77 °C (FanCortrol ON)
FanControl OFF: ~ 10 °C less
Maybe I have a damaged Motherboard or CPU...
Friendly greetings!
-
Sorry, I only have the E6700. I used the Intel Desktop utilities for a while but it seemed to freeze up on me. But with three different temp monitors, we have to believe it is really that hot. That Thermalright fan looks about like a Big Typhoon, I surprised it can't cool it down any. And after seeing your temperature troubles with the QX6700, I don't think I want one.
Before you resort to water cooling, maybe you could try dropping the cpu core voltage back a bit. I have done that to lower my temperatures before. What is the BIOS version on your board? Perhaps updating to the most current version might help?
-
Sorry, I only have the E6700. I used the Intel Desktop utilities for a while but it seemed to freeze up on me. But with three different temp monitors, we have to believe it is really that hot. That Thermalright fan looks about like a Big Typhoon, I surprised it can't cool it down any. And after seeing your temperature troubles with the QX6700, I don't think I want one.
Before you resort to water cooling, maybe you could try dropping the cpu core voltage back a bit. I have done that to lower my temperatures before. What is the BIOS version on your board? Perhaps updating to the most current version might help?
Hello!
Thanx for reply!
I don't overclock, it's @ 2.666 GHz...
I have the newest BIOS.
Your 50 °C on the CPU- heatspreader (which program you use) or with "Intel Thermal Analysis Tool" direct the Cores?
I started a thread at S@H about the temps...
and it's looking like this that the high temperatur is normal...
"Intel Core2 Extreme QX6700 and temps"
http://setiathome.berkeley.edu/forum_thread.php?id=38069
The Quad have around 20 °C more temp. than the Dual´s...
Friendly greetings!
-
I found this here
http://www.pureoverclock.com/download.php?det=6
"TAT (Intel Thermal Analysis Tool) accurately monitors all of the individual core temperatures for Intel processors like the Core 2 Duo and Quad range.
Note* Reported temps will be much higher in TAT than speedfan or motherboard bios readings, they are read from internal diodes inside each core and are not comparible to non-core readings, expect temps of 60c+ even on water!"
I've been watching your thread with interest, it sounds encouraging that there is nothing wrong with your rig.
Happy Crunching,
Pam
-
I found this here
http://www.pureoverclock.com/download.php?det=6
"TAT (Intel Thermal Analysis Tool) accurately monitors all of the individual core temperatures for Intel processors like the Core 2 Duo and Quad range.
Note* Reported temps will be much higher in TAT than speedfan or motherboard bios readings, they are read from internal diodes inside each core and are not comparible to non-core readings, expect temps of 60c+ even on water!"
I've been watching your thread with interest, it sounds encouraging that there is nothing wrong with your rig.
Happy Crunching,
Pam
Yes, yes, I used the Intel Thermal Analysis Tool... :)
A few posts before I said it... ;)
The prog show me only the Core # 0 and # 1...
There is no "button" to shop me the others... or I didn't found him...? ;)
The cores # 0 and # 1 had the same temp like with CoreTemp ...
You have the Intel D975XBX or the D975XBX2?
Which prog you use, to look to the "heat-spreader-temp" from the CPU?
I used MotherboardMonitor, SpeedFan, AIDA32 and EVEREST... no temp from the CPU (heat-spreader)...
Now I have the probs again, why I started a few threads...
"I have probs with Simons S@H V5.15 Rev.2.2A & B"
http://setiathome.berkeley.edu/forum_thread.php?id=37974
Friendly greetings!
-
Yes, I have the Bad Axe 2, so I can maybe upgrade to the QX6700 sometime. That is disappointing to hear TAT doesn't show all four cores.
If BOINC is freezing, that is a kind of crash- did you ever run memtest to check your memory? Some crashes I have had with other builds were with memory the board didn't like. If I couldn't get through a full pass of memtest, I'd switch to some other memory that could and that seemed to solve my problems. I had two sticks of Corsair Value Select 512 Mb and the only board they worked in was a VIA chipset board, they hated my NForce 3 boards. The NForce 3 liked 256Mb sticks but not 512 Mb of the same brand.
I use Crucial Ballistix 2X512 Mb on my Bad Axe 2, ddr2 800 or PC 6400. But I am overclocking and running it at 533 MHz, or 1:1 ratio with memory/fsb. I am at 322 fsb and 1.35 V vcore today with 55-50 deg C according to TAT, which I think measures core temp.
Sorry you are having these troubles, that should be a really good cruncher. I guess I wonder about the version of BOINC, I am using 5.4.11 and am happy with those results and 2.2B Seti app. I have not tried BOINC clients with cpu affinity. I use a trial version of Wndows XP Pro 64 and do updates when it has them.
Regards,
Pam
-
Yes, I have the Bad Axe 2, so I can maybe upgrade to the QX6700 sometime. That is disappointing to hear TAT doesn't show all four cores.
If BOINC is freezing, that is a kind of crash- did you ever run memtest to check your memory? Some crashes I have had with other builds were with memory the board didn't like. If I couldn't get through a full pass of memtest, I'd switch to some other memory that could and that seemed to solve my problems. I had two sticks of Corsair Value Select 512 Mb and the only board they worked in was a VIA chipset board, they hated my NForce 3 boards. The NForce 3 liked 256Mb sticks but not 512 Mb of the same brand.
I use Crucial Ballistix 2X512 Mb on my Bad Axe 2, ddr2 800 or PC 6400. But I am overclocking and running it at 533 MHz, or 1:1 ratio with memory/fsb. I am at 322 fsb and 1.35 V vcore today with 55-50 deg C according to TAT, which I think measures core temp.
Sorry you are having these troubles, that should be a really good cruncher. I guess I wonder about the version of BOINC, I am using 5.4.11 and am happy with those results and 2.2B Seti app. I have not tried BOINC clients with cpu affinity. I use a trial version of Wndows XP Pro 64 and do updates when it has them.
Regards,
Pam
You had probs with Corsair?
Uhh... I have Corsair CM2x1024-6400C4, they running now with 5-5-5-15 @ 800 MHz(normal)
But they are tested (permission) with 4-4-4-12 @ 2.1V & @ 800 MHz too (that's the reason why I bought them)
I have no experiences with overclocking...
With which x-x-x-x you let run your RAMs?
The seller, where I bought the PC, wanted to let run CL4, but it wasn't running (because of the BIOS, I think it was the 2333 11/27/2006... the old...)
I had updated to the newest... but I didn't had the courage to test it now again... ;)
Now I let run the PC without the Intel Desktop Utilities prog, because you said that you had probs (freezing) with this (with BOINC too?)...
Maybe that's the reason for my probs...
If I have probs again... then I make the memtest :)
But I'm little confused, because I took (like I said) AIDA32, EVEREST, SpeedFan and MotherboardMonitor and they don't worked with the D975XBX2.
With which prog you look to the complete CPU temp? (not the cores, the "heat-spreader-temp")
Intel Desktop Utilities show me (you too?) only the "temperature range" ("processor thermal margin", that's the heat-spreader?) to the max. temp 65°C? (..on the middle point of the CPU (heat-spreader) QX6700))
(I think, because the man from Intel I called, couldn't tell me what this temp ("processor thermal margin") is ). My experiences, so hotter the CPU so smaller the indicate (idle=50°C, full load=25°C (FanControl ON))
Friendly greetings!
But I'm little bit confused, because with the SSSE3-Core2 Rev. 1.41 I didn't had the probs with freezing BOINC...
-
...
Here are some more error reports attached occurring to my dual P3 yesterday. (0xc0000005) Hopefully that flaw can be found someday.
Here is my first one of these since i started using 2.2B over a week ago.
[attachment deleted by admin]
-
You had probs with Corsair?
That was just some ddr400. My XMS ddr2 800 works great.
I am running 5-5-5-15 while I am overclocking, maybe later I will try to tighten the timings. If you are not overclocking, you should be able to run 4-4-4-12 with no problems. I did.
Now I let run the PC without the Intel Desktop Utilities prog, because you said that you had probs (freezing) with this (with BOINC too?)...
Maybe that's the reason for my probs...
I just didn't like the way the program acted, I was worried it would steal cpu cycles, sitting there in my systray. It seemed to take forever to respond. I don't think it made BOINC crash, or anything.
But I'm little confused, because I took (like I said) AIDA32, EVEREST, SpeedFan and MotherboardMonitor and they don't worked with the D975XBX2.
With which prog you look to the complete CPU temp? (not the cores, the "heat-spreader-temp")
Intel Desktop Utilities show me (you too?) only the "temperature range" ("processor thermal margin", that's the heat-spreader?) to the max. temp 65°C? (..on the middle point of the CPU (heat-spreader) QX6700))
(I think, because the man from Intel I called, couldn't tell me what this temp ("processor thermal margin") is ). My experiences, so hotter the CPU so smaller the indicate (idle=50°C, full load=25°C (FanControl ON))
Friendly greetings!
But I'm little bit confused, because with the SSSE3-Core2 Rev. 1.41 I didn't had the probs with freezing BOINC...
I guess I am satisfied if it shows me the core temperature. As long as the core isnt' too hot, I am happy. Do they put sensors on the heatspreader? I didn't use Desktop Utilites that long and don't remember "temperature range" or "thermal margin".
When I googled "processor thermal margin", your post here was the top result. I think it must be some kind of engineering term, and your Intel man wasn't an engineer. I don't think it is a temperature, it is a range of temperatures your processor can stand.
If your trouble started when you changed from 1.41 to 2.0 or 2.2, I guess that is what we should look at. From what everyone has said, your processor is not too hot. Maybe you just got a corrupted download of the app. Maybe you could download it again and replace it with a fresh copy. I think the 2.2B apps are doing great, my RAC is still climbing. I don't see why it should be any different for you and it should be better with the QX6700.
Regards,
Pam
-
Hello!
Thanx for your support! :)
I wanted to let run the 4-4-4-12 with the RAMs, but this wasn't function...
After the Power-On Self-Test (POST) I got the message that the actually settings are not good to let run the PC...
I couldn't let him run... he don't wanted... ;)
I musted go back to the "normal" settings (CL5)
When I bought the PC, I said the seller can make the CL4 settings, but to this time with the old BIOS it wasn't possible to let run too... so he called to Intel and the people said that the BIOS must be improved...
I have the newest BIOS... ;)
How you had let it run?
In the BIOS I can select AUTOMATIC - USER - MODE 1 (or like this ;) )
If I select MODE 1 that's the Corsair 4-4-4-12 settings... with 2.1 V
When I let run Intel Desktop Utilities the Windows Task-Manager said that the prog take no CPU-cycles.
If I remember right there is a temp-sensor direct in the middle of the heat-spreader from the CPU.
I let run him all the time with the first copy of Simons app and it's running good now since I terminated the Intel Desktop Utilities...
I don't know, but I will see what will be... :)
At the weekend I changed the heatsink and in the compound manual they say that the first 200 hours the compound need a few circles of cool- hot temps... ON/OFF the PC... so I can let run him 22 hours without probs 2 hours to cool down and then again 22 hours... I will see what will be if I let run him again 24/7... hope he will do the work successful again... :)
@ Simon and/or the optimizer-team
This can be the reason for my probs?
I used:
Intel Desktop Utilities Version 3.0.6.10 and 3.0.10.15
http://www.intel.com/design/motherbd/software/idu/index.htm
I'm not a programmer ;) so maybe if you have time you can look to this software?
-
I am not really amused... :(
*§"%&/(/)/$%"$$"§%*
I had Boinc frozen again... :(
Now I took a new copy of the app and Microsoft .NET V2 and the new Intel Desktop Utilities V3.0.11.16 because I had probs without this prog too...
So... if I have again the probs... then I think I must delete the HDD and make a fresh WinXP to the drive...
Now I don't have updates from the Microsoft homepage, I had only the CD-Rom (WinXP Home SP2) in the drive...
So maybe I must take all the updates from Microsoft to have a good working PC?
(he is running only for S@H, so I don't make safety-relevant work with him...)
-
Bear with me please - I ain't too bright. When I examine my pc I get this: Genuine Intel Intel(R) Core(TM)2 CPU T5600 @ 1.83GHz [x86 Family 6 Model 15 Stepping 6] [fpu tsc pae nx sse sse2 mmx]
I see optimized versions for single cpu sse2 or dual core sse3 but I assume I have a dual core sse2. Why is there not a dual core sse2 - it would probably help if I knew what all that means but heck that would require work on my part!
-
I have a T5200 and already asked...You need the Windows SSE3-Core 2
-
Don't know if this has been reported yet, but my athlon 64 3800+ x2 is returning a CPUID of: CPUID: AMD Sempron 'xxx'.
@ KarVi: Could you Email me your latest version of the 'Patched Intel "only" SSE2-PM'?
-
....
With the BOXed heatsink and fan I had temp. with Intel Desktop Utilities 22 °C (Processor Thermal Margin).
Is here somebody who know the usage of the Intel Desktop Utilities?
Because what does this mean= 22°C (Processor Thermal Margin)?
22°C to go to the max. temp.? So 65 °C - 22 °C = 43 °C?
OnBoard FanControl: 2.000 RPM.
CoreTemp: 71- 79 °C
[room= 22 °C]
....
Hi, I thought this might help with those struggling with thermal margin (I was). This guide was for core 2 duo's but the figures for the quads should be easily looked up from intel once you understand them. http://forumz.tomshardware.com/hardware/Core-Duo-Temperature-Guide-ftopict221745.html (http://forumz.tomshardware.com/hardware/Core-Duo-Temperature-Guide-ftopict221745.html)
It was a link I found on seti forums (I think),a short while ago. I think it is in simple language and should make anyone a wizard on "Thermal Margin etc...". As I understand it ( perhaps wrongly ) thermal margin is the difference between the core temp and cpucase temp , 15C for a C2d (.... and 22C for a c2Q ?)
So by my understanding in your original situation you say Tjunction(core) = 71 to 79 C, Tcase(heatspeader) = 65C .... so that makes a thermal margin under load of 14C ... which is close to the spec of 15C for a c2d, but if it's a quad then it would probably mean the offset needs tweaking (in Speedfan) I guess to correct for innaccurate mobo/bios/chipset reading. ( I think the new offset adjusts the tcase reading down - from 65(innacurrtae bios reading) down 8 degrees to 57, which fixes the numbers!!! yay, 57+22 = 79C )
for the quad cores I would imagine higher cpucase(heatspreader), and higher core temperatures would be specified. for a c2d 65C(case) and 71-79 C would be considered "Hot", though I have heard of people driving them much harder than that. :o
I would love to know what the "intel specified" values for quads are if you find out.
[ Later: Found something...
QX6700 = 130 Watts , Max tcase at fullpower = 64.5C ( at full power, but we are really only 58!)
Heres something "In the event of a catastrophic cooling failure, the processor will
automatically shut down when the silicon has reached a
temperature approximately 20 °C above the maximum TC" So I guess we can TRY and invent a new table for the quad
Loaded
tcase tcores mode
62.5 84.5 thermal shutdown
60 82 Hot
55 77 warm
50 72 normal
45 67 normal
40 62 normal
etc,,,
So If yours is a quad then I think your adjusted tcase of 58C, and core temps of 79, were warm warm, but not quite hot.
and your upgraded cooling giving tcase=50 and tcore=72 sounds much nicer is now reporting spot on 22 degrees so offset won't be needed in speedfan ( if those temps didn't use an offset already )
Can anybody confirm or deny they really run like this ? I mean 82C cores perhaps you could use it to cook eggs somehow.
]
Jason