How much torque test is needed?

E_tron · 2004-01-04, 19:14

I am actively overclocking an Athlon XP machine.

I am wondering how much torque test in p95 will deem a machine 99.999999999999999...% stable?

1hour?
2hours?
5hours?
5years?

I want to complete work faster, however i want to complete work with 100% accuracy.

outlnder · 2004-01-04, 20:16

A 24 hour torture test with a 24 hour Memtest should do it.

dsouza123 · 2004-01-04, 20:52

Should the tests be done sequentially or in parallel ?

Xyzzy · 2004-01-04, 22:49

Why bother with a synthetic torture test when you can run a few double checks and know for sure, and in the process have some useful work completed?

I'd still run the one hour default torture test, but anything beyond that I'd try some actual work...

garo · 2004-01-05, 18:27

Doublechecks are the way to go. Run 7-10 days of doublechecks. You will get credit for your tests and get the ultimate torture test.

E_tron · 2004-01-07, 01:47

thanks for the advice

This machine passed 24hours of P95 test

. Now it can do double checks and LLs 38% faster

.

for the people that want to know... This is a AMD Athlon XP 2500+ Barton based system with an Abit NF7-S mainboard. The CPU is the prodigy of all the 2500+s i got my hands on.

I have the CPU running at 2383.5mhz with a 1.825vcore and 227mhz FSB speed. It is air cooled with the RAM running at 430DDR'ish.

227mhz FSB x 10.5 CPU multiplier = 2383.5mhz'ish

according to AMD's rating system formulas, this is a "3500+," but I don’t think it could out perform a 3.5ghz P4.

...now if i could get a P4 to do this... i could be crunching some numbers.

Xyzzy · 2004-01-07, 03:48

Quote:

Originally posted by E_tron
...now if i could get a P4 to do this... i could be crunching some numbers.

Getting a P4 to run at 2.4GHz should be pretty easy...

E_tron · 2004-01-07, 04:20

Quote:

Originally posted by Xyzzy
Getting a P4 to run at 2.4GHz should be pretty easy...

you are right. it would be easy

.

i was getting at the percentage performance increace

(not that a P4 can't get the same ratio)

2004-01-04, 19:14	#1
E_tron Sep 2002 Austin, TX 3×11×17 Posts	How much torque test is needed? I am actively overclocking an Athlon XP machine. I am wondering how much torque test in p95 will deem a machine 99.999999999999999...% stable? 1hour? 2hours? 5hours? 5years? I want to complete work faster, however i want to complete work with 100% accuracy.

2004-01-04, 20:16	#2
outlnder Aug 2002 2×3×53 Posts	A 24 hour torture test with a 24 hour Memtest should do it. Last fiddled with by outlnder on 2004-01-04 at 20:16

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Modifying the Lucas Lehmer Primality Test into a fast test of nothing	Trilo	Miscellaneous Math	25	2018-03-11 23:20
Adraktas papers - 'the Torque is strong in this one'.	ewmayer	Miscellaneous Math	24	2016-02-03 03:02
Help needed to test Athlon64 code	geoff	Programming	7	2006-08-18 12:16
Predicting the Amount of CPU Power Needed to LL Test Exponent p	jinydu	Math	19	2005-02-07 02:52
Help needed for stress test of Albatron motherboard	phrater phil	Hardware	6	2004-10-21 11:20

2004-01-04, 20:52	#3
dsouza123 Sep 2002 662₁₀ Posts	Should the tests be done sequentially or in parallel ?

2004-01-04, 22:49	#4
Xyzzy "Mike" Aug 2002 17545₈ Posts	Why bother with a synthetic torture test when you can run a few double checks and know for sure, and in the process have some useful work completed? I'd still run the one hour default torture test, but anything beyond that I'd try some actual work...

2004-01-05, 18:27	#5
garo Aug 2002 Termonfeckin, IE 3²×307 Posts	Doublechecks are the way to go. Run 7-10 days of doublechecks. You will get credit for your tests and get the ultimate torture test.

2004-01-07, 01:47	#6
E_tron Sep 2002 Austin, TX 1000110001₂ Posts	thanks for the advice This machine passed 24hours of P95 test . Now it can do double checks and LLs 38% faster . for the people that want to know... This is a AMD Athlon XP 2500+ Barton based system with an Abit NF7-S mainboard. The CPU is the prodigy of all the 2500+s i got my hands on. I have the CPU running at 2383.5mhz with a 1.825vcore and 227mhz FSB speed. It is air cooled with the RAM running at 430DDR'ish. 227mhz FSB x 10.5 CPU multiplier = 2383.5mhz'ish according to AMD's rating system formulas, this is a "3500+," but I don’t think it could out perform a 3.5ghz P4. ...now if i could get a P4 to do this... i could be crunching some numbers.