akosf C37 performance observations thread

Santas little helper
Santas little helper
Joined: 11 Feb 05
Posts: 36
Credit: 9852826
RAC: 29262

S38 is out. :-) Comparing

S38 is out. :-)
Comparing akosf's C-37 with some S-38 on his 1,86 GHz Pentium-M it's again alot faster, ratio: 0.61
Let's crunch some S-38 :-)

Greetings, Santas little helper

Stef
Stef
Joined: 8 Mar 05
Posts: 206
Credit: 110568193
RAC: 0

S38 is 30% faster than C37 on

S38 is 30% faster than C37 on a P3T and 20% faster on a P3. But it doesn't work on a AMD Thunderbird yet:
***UNHANDLED EXCEPTION****
Reason: Illegal Instruction (0xc000001d) at address 0x0040A84E

anders n
anders n
Joined: 29 Aug 05
Posts: 123
Credit: 1656300
RAC: 0

S38 works fine on the 2

S38 works fine on the 2 computers I have tried it on. :)

Anders n

B52
B52
Joined: 19 Feb 05
Posts: 45
Credit: 273899
RAC: 0

RE: S38 is 30% faster than

Message 25999 in response to message 25997

Quote:
S38 is 30% faster than C37 on a P3T and 20% faster on a P3. But it doesn't work on a AMD Thunderbird yet:
***UNHANDLED EXCEPTION****
Reason: Illegal Instruction (0xc000001d) at address 0x0040A84E

The S38 is SSE optimized, a T-BIRD cannot handle SSE, only MMX. I know this cause I have run the cpu-z app on my own t-bird 1.2 ghz. CPU-z does NOT see the t-bird as SSE capable

Stef
Stef
Joined: 8 Mar 05
Posts: 206
Credit: 110568193
RAC: 0

RE: The S38 is SSE

Quote:
The S38 is SSE optimized, a T-BIRD cannot handle SSE, only MMX. I know this cause I have run the cpu-z app on my own t-bird 1.2 ghz. CPU-z does NOT see the t-bird as SSE capable


Oops, you're right, my fault.
flags: fpu vme de pse tsc msr pae mce cx8 sep mtrr pge mca cmov pat pse36 psn mmxext mmx fxsr 3dnowext 3dnow

pisi78
pisi78
Joined: 22 Jan 05
Posts: 4
Credit: 380361
RAC: 0

RE: RE: The S38 is SSE

Message 26001 in response to message 26000

Quote:
Quote:
The S38 is SSE optimized, a T-BIRD cannot handle SSE, only MMX. I know this cause I have run the cpu-z app on my own t-bird 1.2 ghz. CPU-z does NOT see the t-bird as SSE capable

Oops, you're right, my fault.
flags: fpu vme de pse tsc msr pae mce cx8 sep mtrr pge mca cmov pat pse36 psn mmxext mmx fxsr 3dnowext 3dnow

AMD added SSE with the athlon XP

Alinator
Alinator
Joined: 8 May 05
Posts: 927
Credit: 9352143
RAC: 0

Akosf, Just an update on

Akosf,

Just an update on my K6-2's.

One of them appears to be continuing to return invalid results with C37. What's really baking my noodle on this one is these two machines are pretty similiar from a hardware and basic performance viewpoint!

I had expected if there was going to be a problem it would be with the K6/300, but this has really got me scratching my head. I think the next step is to go back to 387 on the malfunctioning one to make sure it hasn't developed a hardware fault.

Alinator

rbpeake
rbpeake
Joined: 18 Jan 05
Posts: 266
Credit: 1136147797
RAC: 560513

Unbelievable! My time went

Unbelievable!

My time went from 4 hours on the regular science application to 1 hour ll minutes using S38. That is 30% of the time it used to take my AMD 3500+!

That means the 120 day workload could potentially be reduced to about 36 days (for Windows machines, anyway) if everyone used and got the same efficiencies from S38. Not bad! :D

RandyC
RandyC
Joined: 18 Jan 05
Posts: 6629
Credit: 111139797
RAC: 0

RE: AMD added SSE with the

Message 26004 in response to message 26001

Quote:
AMD added SSE with the athlon XP

My Athlon XP 1700+ shows as a T-Bird with CPU-Z. S38 appears not to work on it.

Seti Classic Final Total: 11446 WU.

Akos Fekete
Akos Fekete
Joined: 13 Nov 05
Posts: 561
Credit: 4527270
RAC: 0

RE: My Athlon XP 1700+

Message 26005 in response to message 26004

Quote:
My Athlon XP 1700+ shows as a T-Bird with CPU-Z. S38 appears not to work on it.

AMD introduced XP 1700+ with three core variation (Palomino, Thorougbred-A and Thoroughbred-B), all of them know SSE.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.