Whetstone/Dhrystone problem

Message boards : Number crunching : Whetstone/Dhrystone problem
Message board moderation

To post messages, you must log in.

1 · 2 · 3 · Next

AuthorMessage
Astro
Volunteer tester
Avatar

Send message
Joined: 16 Apr 02
Posts: 8026
Credit: 600,015
RAC: 0
Message 142872 - Posted: 26 Jul 2005, 1:54:53 UTC

Evening all. This is a continuation from my problems found in the thread Optimized windows clients - plz help listing cpu times, it is also the same machine having troubles from the thread Hard drive help- Advanced

The basic problem is that the puter has a bad benchmark and slow WU completion times for the Processor and memory as compared with others of the same breed. I put my puter in fridge to cool it, then rebooted. Only the Taskmanager and Boinc are running during this test. I ran the benchmarks every so often and got the following:

Time Whet Dhry
8:59 1642 2813
9:01 1658 2788
9:03 1548 2815
9:06 935 2348
9:08 1100 2498
9:10 1108 1867
9:13 1112 1805
9:16 919 1628
9:20 739 1960
9:24 923 1558
9:33 736 1724
9:49 955 1637

As you can see, the Whetstone drops off after 4 minutes but the dhrystone drops off at 8 minutes. What beside heat can cause this. The laptop was cleaned when I had it apart last week. It wasn't that filthy at that time.

My Bios doesn't have options for Overclocking or even heat displays with MBM5 or speedfan. It is HOT in this house and the laptop is hot. Is there something hidden in the Bios that is limiting its performance? There are no Options I can find such as "cool and quiet".

Any suggestions?

thanks
tony

ID: 142872 · Report as offensive
Astro
Volunteer tester
Avatar

Send message
Joined: 16 Apr 02
Posts: 8026
Credit: 600,015
RAC: 0
Message 142882 - Posted: 26 Jul 2005, 2:22:58 UTC

from a previous thread HTRAE asked:

htrae
Joined: Apr 3, 1999
Posts: 15
ID: 7841333
Credit: 34,127
RAC: 376
Message 141971 - Posted 24 Jul 2005 14:56:49 UTC

@ mmciastro:

Download this freebie system analysis program and it should be able to tell you which Phoenix Bios version you have. In the left menu choose Motherboard > Bios. The program can also tell some other info such as Ram timings, FSB, Multiplier, etc. Hope this helps.

http://www.lavalys.com/products/overview.php?pid=1&lang=en


My bios is a "PhoenixBios Setup Utility, Bios Rev KAM 1.58

How does this help?


ID: 142882 · Report as offensive
Astro
Volunteer tester
Avatar

Send message
Joined: 16 Apr 02
Posts: 8026
Credit: 600,015
RAC: 0
Message 142883 - Posted: 26 Jul 2005, 2:27:39 UTC
Last modified: 26 Jul 2005, 2:30:31 UTC

from a previous post I2satx asked:

j2satx
Joined: Oct 2, 2002
Posts: 130
ID: 17806
Credit: 120,606
RAC: 85
Message 141960 - Posted 24 Jul 2005 14:43:45 UTC

Pull the battery and run it on AC. Does it improve performance


The answer is 558 whet, 1857 Dhry before, pull plug....922/1522.

Insert AC, remove battery, run benchmark.....934/1669 so I don't see any difference,

thanks though
ID: 142883 · Report as offensive
Astro
Volunteer tester
Avatar

Send message
Joined: 16 Apr 02
Posts: 8026
Credit: 600,015
RAC: 0
Message 142885 - Posted: 26 Jul 2005, 2:33:10 UTC

from a previous thread Fharmon states:

fharmon
Volunteer tester
Joined: Oct 19, 2002
Posts: 326
ID: 93937
Credit: 54,225
RAC: 431
Message 141791 - Posted 24 Jul 2005 6:25:58 UTC

@mmciastro
Is that you laptop, I have my laptop on a cooling platform $20 from Compusa.
If the cpu is running hot most laptops reduce the cpu cycles so it doesn't burn up.


Yes It's a notebook/laptop. It has been elevated for months and from below, it went from fridge temp to reduced performance within 4 minutes after reboot.


ID: 142885 · Report as offensive
Heffed
Volunteer tester

Send message
Joined: 19 Mar 02
Posts: 1856
Credit: 40,736
RAC: 0
United States
Message 142889 - Posted: 26 Jul 2005, 2:41:56 UTC - in response to Message 142872.  

I put my puter in fridge to cool it, then rebooted.

Yikes! Not a good idea at all!

Condensation isn't something you want in your machine. Putting something hot in a cold place causes condensation, and you should always let a cool piece of electronics reach ambient temp before turning it on to avoid condensation. It sounds like you did both!

Hopefully no lasting damage was done.
ID: 142889 · Report as offensive
Astro
Volunteer tester
Avatar

Send message
Joined: 16 Apr 02
Posts: 8026
Credit: 600,015
RAC: 0
Message 142893 - Posted: 26 Jul 2005, 2:46:21 UTC - in response to Message 142889.  

I put my puter in fridge to cool it, then rebooted.

Yikes! Not a good idea at all!

Condensation isn't something you want in your machine. Putting something hot in a cold place causes condensation, and you should always let a cool piece of electronics reach ambient temp before turning it on to avoid condensation. It sounds like you did both!

Hopefully no lasting damage was done.

I fully understood this going in. That's why after the diagnosis from the "Advanced Hard drive" help thread, I'm typing this on my new AMD mobile Athlon64 3700+ laptop. I'm just trying to figure out what might be in common. Might both problems be Bios, or are they both Hardware (i.e. motherboard). Thanks for your concern though. The next stop for that laptop is the trash so I was unconcerned.
ID: 142893 · Report as offensive
Profile The Gas Giant
Volunteer tester
Avatar

Send message
Joined: 22 Nov 01
Posts: 1904
Credit: 2,646,654
RAC: 0
Australia
Message 142894 - Posted: 26 Jul 2005, 2:46:43 UTC - in response to Message 142889.  
Last modified: 26 Jul 2005, 2:47:44 UTC

I put my puter in fridge to cool it, then rebooted.

Yikes! Not a good idea at all!

Condensation isn't something you want in your machine. Putting something hot in a cold place causes condensation, and you should always let a cool piece of electronics reach ambient temp before turning it on to avoid condensation. It sounds like you did both!

Hopefully no lasting damage was done.


erm...shouldn't that be something cold in a hot place produces condensation....like cold water into a glass gets condensation on the outside of the glass.
ID: 142894 · Report as offensive
Heffed
Volunteer tester

Send message
Joined: 19 Mar 02
Posts: 1856
Credit: 40,736
RAC: 0
United States
Message 142896 - Posted: 26 Jul 2005, 2:51:20 UTC - in response to Message 142894.  

erm...shouldn't that be something cold in a hot place produces condensation....like cold water into a glass gets condensation on the outside of the glass.

Try this...

Get a hot bowl of something. Cover with plastic wrap. Place in fridge. What happens?
ID: 142896 · Report as offensive
W-K 666 Project Donor
Volunteer tester

Send message
Joined: 18 May 99
Posts: 19012
Credit: 40,757,560
RAC: 67
United Kingdom
Message 142898 - Posted: 26 Jul 2005, 2:55:21 UTC

Hi Tony,

If I rememeber correctly, AMD fitted a thermal diode to their CPU's to interrupt the clock, which effectively slows down the CPU. And if my memory is working correctly, there must also be overheat protection device fitted to the motherboard which switches off the PSU if the temperature rises beyond the temp of the internal CPU diode.

So I guess what you are seeing is 'correct operation'. I'm not into notebooks so don't know if there is anything can be done on the cooling front except run it in the fridge ;-)

Andy
ID: 142898 · Report as offensive
Astro
Volunteer tester
Avatar

Send message
Joined: 16 Apr 02
Posts: 8026
Credit: 600,015
RAC: 0
Message 142899 - Posted: 26 Jul 2005, 2:55:38 UTC

The extreme is what I was shooting for. I took the benchmarks as frequently as I did to see if the performance loss was digital or analog. It didn't degrade little by little, but rather digital, like a switch was thrown. It also happened within minutes of activation, so I don't think it could have been hot enough to trip a function temp sensor. I'm assuming I have a bad temp sensor, but I'm just guessing.
ID: 142899 · Report as offensive
W-K 666 Project Donor
Volunteer tester

Send message
Joined: 18 May 99
Posts: 19012
Credit: 40,757,560
RAC: 67
United Kingdom
Message 142901 - Posted: 26 Jul 2005, 3:01:43 UTC - in response to Message 142899.  

The extreme is what I was shooting for. I took the benchmarks as frequently as I did to see if the performance loss was digital or analog. It didn't degrade little by little, but rather digital, like a switch was thrown. It also happened within minutes of activation, so I don't think it could have been hot enough to trip a function temp sensor. I'm assuming I have a bad temp sensor, but I'm just guessing.


I occasionally monitor temps using everest if computer manufacturer doesn't supply and the CPU temp on this machine rises from 45C approx to mid 60s in less than 2 sec when switching from idle to running Boinc/seti.

Andy
ID: 142901 · Report as offensive
Astro
Volunteer tester
Avatar

Send message
Joined: 16 Apr 02
Posts: 8026
Credit: 600,015
RAC: 0
Message 142902 - Posted: 26 Jul 2005, 3:01:48 UTC - in response to Message 142898.  

Hi Tony,

If I rememeber correctly, AMD fitted a thermal diode to their CPU's to interrupt the clock, which effectively slows down the CPU. And if my memory is working correctly, there must also be overheat protection device fitted to the motherboard which switches off the PSU if the temperature rises beyond the temp of the internal CPU diode.

So I guess what you are seeing is 'correct operation'. I'm not into notebooks so don't know if there is anything can be done on the cooling front except run it in the fridge ;-)

Andy

I fear you might be right. Would this CPU diode cause the Whetstone to drop independently from the Dhrystone? There was a 4 minute seperation. Also, If you look up the hard/drive (IDE controller section) problem, you'll also notice it's heat exacerbated as well.
ID: 142902 · Report as offensive
W-K 666 Project Donor
Volunteer tester

Send message
Joined: 18 May 99
Posts: 19012
Credit: 40,757,560
RAC: 67
United Kingdom
Message 142903 - Posted: 26 Jul 2005, 3:11:10 UTC - in response to Message 142902.  

I fear you might be right. Would this CPU diode cause the Whetstone to drop independently from the Dhrystone? There was a 4 minute seperation. Also, If you look up the hard/drive (IDE controller section) problem, you'll also notice it's heat exacerbated as well.


I know that intel also fit several thermal diodes in critical places , like the cpu core, number cruncher and cache areas, to their CPU's. I can only assume AMD do likewise, and that the cpu core is heating first, interger critical and that the bit slicers etc in the arithmetic unit, which will do the double precision maths of the Whetstone part, are heating up at a slower rate and then kicking in, producing the stepped effect seen.

Andy

ID: 142903 · Report as offensive
Profile Jim Baize
Volunteer tester

Send message
Joined: 6 May 00
Posts: 758
Credit: 149,536
RAC: 0
United States
Message 142908 - Posted: 26 Jul 2005, 3:36:07 UTC - in response to Message 142896.  

Because the moisture is coming from the substance and condensing on the cold plastic wrap. The moisture is also condensing in the cold air trapped inside the plastic wrap.

now, if the warm moist "outside air" was still trapped inside the laptop and unable to escape then yes, it would condense on the parts of the laptop that cooled the fastest.

Warm air holds more moisture than does cold air.

Try this...

Get a hot bowl of something. Cover with plastic wrap. Place in fridge. What happens?


ID: 142908 · Report as offensive
W-K 666 Project Donor
Volunteer tester

Send message
Joined: 18 May 99
Posts: 19012
Credit: 40,757,560
RAC: 67
United Kingdom
Message 142912 - Posted: 26 Jul 2005, 3:43:34 UTC

Tony

When you had the notebook apart did you clean and change the heatsink compound?

And I knew I had seen something somewhere, so have been off looking and found this Notebook Coolers .

Andy
ID: 142912 · Report as offensive
Profile Jord
Volunteer tester
Avatar

Send message
Joined: 9 Jun 99
Posts: 15184
Credit: 4,362,181
RAC: 3
Netherlands
Message 142925 - Posted: 26 Jul 2005, 4:48:21 UTC

Tony, if you don't mind me asking. What make & model is this laptop? I checked a couple of the other threads, but never saw you mention it in detail.
ID: 142925 · Report as offensive
Astro
Volunteer tester
Avatar

Send message
Joined: 16 Apr 02
Posts: 8026
Credit: 600,015
RAC: 0
Message 142929 - Posted: 26 Jul 2005, 4:52:50 UTC - in response to Message 142925.  

Tony, if you don't mind me asking. What make & model is this laptop? I checked a couple of the other threads, but never saw you mention it in detail.

Compaq Presario 2140US model 954A
mobile AMD AthlonXP2200+
512M ram
ID: 142929 · Report as offensive
Professor Desty Nova
Avatar

Send message
Joined: 17 May 99
Posts: 59
Credit: 579,918
RAC: 0
Portugal
Message 142964 - Posted: 26 Jul 2005, 6:58:57 UTC

Do you know CPU-Z?
You can use it to see how the processor frequency changes in real time.
SETI@home classic workunits: 1,985 CPU time: 24,567 hours



Professor Desty Nova
Researching Karma the Hard Way
ID: 142964 · Report as offensive
Kokie

Send message
Joined: 18 Nov 03
Posts: 12
Credit: 8,294,384
RAC: 0
Germany
Message 142982 - Posted: 26 Jul 2005, 8:45:51 UTC - in response to Message 142898.  

If I rememeber correctly, AMD fitted a thermal diode to their CPU's to interrupt the clock, which effectively slows down the CPU.


The Athlon XP has a thermal diode, but no protection circuitry to throttle the CPU. This was introduced at first with the Athlon 64.
ID: 142982 · Report as offensive
Profile Jord
Volunteer tester
Avatar

Send message
Joined: 9 Jun 99
Posts: 15184
Credit: 4,362,181
RAC: 3
Netherlands
Message 142993 - Posted: 26 Jul 2005, 9:46:25 UTC - in response to Message 142929.  

Tony, if you don't mind me asking. What make & model is this laptop? I checked a couple of the other threads, but never saw you mention it in detail.

Compaq Presario 2140US model 954A
mobile AMD AthlonXP2200+
512M ram

Ok.

One HP/Compaq update page for you. :)
ID: 142993 · Report as offensive
1 · 2 · 3 · Next

Message boards : Number crunching : Whetstone/Dhrystone problem


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.