Database for k-b-b's: - Page 4

mdettweiler · 2010-08-30, 19:09

Quote:

Originally Posted by 3.14159

I'm unsure whether or not I damaged the CPU or anything related. Checking the computer showed some signs of the fans being a bit too hot (Approx. body temp.)

I doubt you would have damaged the CPU by any of the things you've been running on it over the course of this thread and "PARI's commands"; more likely it already had a minor instability that shows up once the CPU temperature gets over a certain level. That can be due to a number of factors, including room temperature, dust on the CPU, and the kind of application you're running (primality testing vs. sieving).

If the CPU is faulty in the least, Prime95 will tell you--given the frequency of the PFGW roundoff errors you got, if there is a CPU problem it should show up within 12-16 hours of stress testing. Additionally, you may want to check the CPU's temperature with a program designed to do so; google "Core Temp" for one of the most popular ones. ("Real Temp" is also good; they often give different measurements on the same CPU, and some CPUs work more accurately with one than the other. But regardless of what they give, their relative measurements of how far your CPU is from its recommended maximum should be spot-on.)

mdettweiler · 2010-08-30, 19:11

Quote:

Originally Posted by 3.14159

Well, it's been working error free, since b = 440 without the a1 flag.

It happens to do this about every 100 values, as far as I know.

Interesting. That does make me start to wonder if this is really just a program bug instead of a hardware issue...well, I guess the Prime95 stress test will confirm that once and for all.

3.14159 · 2010-08-30, 19:12

As far as I see it, it seems to be nothing more than a minor bug, because the tests still work and I didn't miss anything.

Also: Got every prime up to 1300 digits listed.

3.14159 · 2010-08-30, 19:19

Quote:

Originally Posted by Max

Interesting. That does make me start to wonder if this is really just a program bug instead of a hardware issue...well, I guess the Prime95 stress test will confirm that once and for all.

p95 will be operated at 22:00

If it were a hardware issue: I think I would have gotten blue-screened.

rogue · 2010-08-30, 19:32

Are you using the latest version of PFGW? I added a feature in 3.3.0 that would automatically increase the FFT size if it detected a roundoff/sumout/maxdiff error. You shouldn't need to use -a1 at all.

What I'm really confused by is that you mentioned some problem back in post 23, but there is nothing above it detailing the problem.

Whatever you do, I strongly recommend using PFGW or LLR for the PRP/primality test. PARI will only slow you down.

3.14159 · 2010-08-30, 19:35

Karsten:

File for k-b-b's, k = 1 to 10000; b = 251 to 500:

kbb251to500.txt

I will now work on the file for 501 to 750.

3.14159 · 2010-08-30, 19:39

Rogue: See here

kar_bon · 2010-08-30, 19:46

b<=250 are online, but the next file got another type of presenting the results, so my script won't handle this!

Bring it to the first format and it's no problem to update those data, too!

3.14159 · 2010-08-30, 19:48

Quote:

Originally Posted by Karsten

b<=250 are online, but the next file got another type of presenting the results, so my script won't handle this!

An advanced text editor could have easily helped you out there, but fine. I'll convert 251 to 500 to a more web-friendly format.

kar_bon · 2010-08-30, 19:55

Quote:

Originally Posted by 3.14159

An advanced text editor could have easily helped you out there, but fine. I'll convert 251 to 500 to a more web-friendly format.

Hey, I've done enough for this task and a text-editor (NotePad satisfy) is doable for you, too!

3.14159 · 2010-08-30, 19:58

Quote:

Originally Posted by Karsten

Hey, I've done enough for this task and a text-editor (NotePad satisfy) is doable for you, too!

I spent 330 minutes of my time collecting these. Why not help me out?

2010-08-30, 19:12	#36
3.14159 May 2010 Prime hunting commission. 690₁₆ Posts	As far as I see it, it seems to be nothing more than a minor bug, because the tests still work and I didn't miss anything. Also: Got every prime up to 1300 digits listed. Last fiddled with by 3.14159 on 2010-08-30 at 19:17

2010-08-30, 19:32	#38
rogue "Mark" Apr 2003 Between here and the 1100011001011₂ Posts	Are you using the latest version of PFGW? I added a feature in 3.3.0 that would automatically increase the FFT size if it detected a roundoff/sumout/maxdiff error. You shouldn't need to use -a1 at all. What I'm really confused by is that you mentioned some problem back in post 23, but there is nothing above it detailing the problem. Whatever you do, I strongly recommend using PFGW or LLR for the PRP/primality test. PARI will only slow you down. Last fiddled with by rogue on 2010-08-30 at 19:32

2010-08-30, 19:35	#39
3.14159 May 2010 Prime hunting commission. 2⁴×3×5×7 Posts	Karsten: File for k-b-b's, k = 1 to 10000; b = 251 to 500: kbb251to500.txt I will now work on the file for 501 to 750. Last fiddled with by 3.14159 on 2010-08-30 at 19:36

2010-08-30, 19:39	#40
3.14159 May 2010 Prime hunting commission. 3220₈ Posts	Rogue: See here Last fiddled with by 3.14159 on 2010-08-30 at 19:47

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
CPU Performance Database	JustinGC	Hardware	3	2012-06-22 03:11
Database question	Xyzzy	Forum Feedback	0	2012-04-01 15:37
entire database done up to 2^58	tha	Lone Mersenne Hunters	11	2010-10-17 08:27
NPLB Database	IronBits	No Prime Left Behind	177	2009-10-10 09:00
database.zip	HiddenWarrior	Data	1	2004-03-29 03:53

2010-08-30, 19:46	#41
kar_bon Mar 2006 Germany 2³×3×11² Posts	b<=250 are online, but the next file got another type of presenting the results, so my script won't handle this! Bring it to the first format and it's no problem to update those data, too!