Well, I recode the core of albert, but I cannot test it because of lack of SSE2 capabale machine. Where is my Conroe??? :)
Would you like to test it? S37a
This is a SSE2 version of C37.
I'm also in a lack of Conroe machines :).
But I'll test the functionality with a P4 immediately and with the Yonah within the next days.
Well, I recode the core of albert, but I cannot test it because of lack of SSE2 capabale machine. Where is my Conroe??? :)
Would you like to test it? S37a
This is a SSE2 version of C37.
S37a works on my P4 Northwood 2.4Ghz, but it seems slower than C37 (8% slower in a first and not very accurate partial measure)
Well, I recode the core of albert, but I cannot test it because of lack of SSE2 capabale machine. Where is my Conroe??? :)
Would you like to test it? S37a
This is a SSE2 version of C37.
I've finished the first two WUs on my Athlon 64 X2.
First of all, they are valid!
It took ca. 65-70 sec less to crunch them, that's about -1.5%. :)
So at least there is some benefit.
And again thank you very much for your enthusiastic work.
First, I would like to apologize for unintentionally misspelling your name (as akosh) :-{ on my first response (about 8 or 9 days ago).
I've replaced the A36 version with C37 and I am seeing a further inprovement of ~5% in crunchtime. Our heartfelt thanks to you for your donation of time and genius to the Einstein project.
Athlon XP Mobile 2600 (K7, so only SSE, no SSE2, etc), overclocked to 2.5Ghz
512Mb ddr400 memory
Respects,
Michael R.
microcraft
"The arc of history is long, but it bends toward justice" - MLK
RE: Well, I recode the core
)
I'm also in a lack of Conroe machines :).
But I'll test the functionality with a P4 immediately and with the Yonah within the next days.
RE: C37 is really slower on
)
I'm seeing much better than that on a PIII. C37 is giving me a 20% improvement over A36.
On an old PIII 600 Mhz laptop, with A36 I have 39,811 seconds, whereas C37 is giving me 31,162 seconds.
RE: Well, I recode the
)
S37a works on my P4 Northwood 2.4Ghz, but it seems slower than C37 (8% slower in a first and not very accurate partial measure)
RE: S37a works on my P4
)
Thanks! My hint was -3%...
You also need a Conroe... ;-)
RE: Well, I recode the core
)
I've finished the first two WUs on my Athlon 64 X2.
First of all, they are valid!
It took ca. 65-70 sec less to crunch them, that's about -1.5%. :)
So at least there is some benefit.
And again thank you very much for your enthusiastic work.
cu,
Michael
RE: I've finished the first
)
Wow! Big thanks!
Wow! Was it faster? I never thought that!
I have to get an X2 beside Conroe... ;-)
Akosf, First, I would like
)
Akosf,
First, I would like to apologize for unintentionally misspelling your name (as akosh) :-{ on my first response (about 8 or 9 days ago).
I've replaced the A36 version with C37 and I am seeing a further inprovement of ~5% in crunchtime. Our heartfelt thanks to you for your donation of time and genius to the Einstein project.
Athlon XP Mobile 2600 (K7, so only SSE, no SSE2, etc), overclocked to 2.5Ghz
512Mb ddr400 memory
Respects,
Michael R.
microcraft
"The arc of history is long, but it bends toward justice" - MLK
RE: C37 is really slower on
)
Just to verify, you're saying C37 is compatible with all x86 platforms?
If so then I'll give it a go on my K6-2/500's, maybe even my K6/300 just for laughs. ;-)
Alinator
Hey Alinator! I did C37 for
)
Hey Alinator!
I did C37 for your K6...
Hey Michael!
We will not do problem from a letter...
...'f' is coming from my family name.
I'm glad that good speed of C37. :)
I tested the S37A on My A64
)
I tested the S37A on My A64 3500+ @2500Mhz
1st S37A Result 5316 sec
average 5434 Sec last 6 results C37
=2,2% Faster