I also have gotten a speed increase from D41.14 on my Athlons. BUT on my K6-2 500 (oc'd to 550) there was a slow down from D41.12. ~34 k sec to 40 k sec.
I am now running D41.14 on the Tbirds and D41.12 on the K6-2
I also have gotten a speed increase from D41.14 on my Athlons. BUT on my K6-2 500 (oc'd to 550) there was a slow down from D41.12. ~34 k sec to 40 k sec.
I am now running D41.14 on the Tbirds and D41.12 on the K6-2
On my K6-2 400 I noticed a slowdown from D40 to D41.14. D41.12 seems to be the fastest version, but it produced only one result without an error (http://einsteinathome.org/task/27206675). Maybe I will try it again.
What I find strange though is that my K6-2 (WU times ranging from about 15500 to 19500 seconds) is a lot faster than yours (about 32000 to 46800), although mine is running at a lower clock speed.
What I find strange though is that my K6-2 (WU times ranging from about 15500 to 19500 seconds) is a lot faster than yours (about 32000 to 46800), although mine is running at a lower clock speed.
Maybe you have faster RAM and a bigger or a Burst-L2-cache?
What I find strange though is that my K6-2 (WU times ranging from about 15500 to 19500 seconds) is a lot faster than yours (about 32000 to 46800), although mine is running at a lower clock speed.
Maybe you have faster RAM and a bigger or a Burst-L2-cache?
RAM Timings are 2-2-2-6@100Mhz, CPU-Z says it has no L2 cache and 32+32KB L1 cache.
What I find strange though is that my K6-2 (WU times ranging from about 15500 to 19500 seconds) is a lot faster than yours (about 32000 to 46800), although mine is running at a lower clock speed.
My K6-2 speed differences are from different Akos app versions. I seem to only have one WU size on this computer so far, but other computers of mine have TruX estimated times of 18 minutes to 1hour 30 minutes due to different WU sizes. I expect the time difference betweens us is from different size WUs, though I do not have fast memory or a particular great mb.
I also have gotten a speed
)
I also have gotten a speed increase from D41.14 on my Athlons. BUT on my K6-2 500 (oc'd to 550) there was a slow down from D41.12. ~34 k sec to 40 k sec.
I am now running D41.14 on the Tbirds and D41.12 on the K6-2
Good work Akos!
Joe B
I can't use the U41 app with
)
I can't use the U41 app with my 427427 system. Which is faster the D41 or S41 app?
RE: I can't use the U41 app
)
S41
My a1500 seems to be
)
My a1500 seems to be breakeven over D41.12. Still 70-74m per WU.
RE: I also have gotten a
)
On my K6-2 400 I noticed a slowdown from D40 to D41.14. D41.12 seems to be the fastest version, but it produced only one result without an error (http://einsteinathome.org/task/27206675). Maybe I will try it again.
What I find strange though is that my K6-2 (WU times ranging from about 15500 to 19500 seconds) is a lot faster than yours (about 32000 to 46800), although mine is running at a lower clock speed.
RE: What I find strange
)
Maybe you have faster RAM and a bigger or a Burst-L2-cache?
Aloha, Uli
RE: RE: What I find
)
RAM Timings are 2-2-2-6@100Mhz, CPU-Z says it has no L2 cache and 32+32KB L1 cache.
RE: What I find strange
)
My K6-2 speed differences are from different Akos app versions. I seem to only have one WU size on this computer so far, but other computers of mine have TruX estimated times of 18 minutes to 1hour 30 minutes due to different WU sizes. I expect the time difference betweens us is from different size WUs, though I do not have fast memory or a particular great mb.
Joe B
RE: I can't use the U41 app
)
452748 Computer ID
AMD Athlon(tm) 64 Processor 3200+, MMX,SSE,SSE2,3DNOW,x86-64, L1:128kB, L2:512kB
result-id name CPU time min:sec albert_4.37_windows_
27983413 z1_1154.5__643_S4R2a_0 3.578,7 59:38,7 D40
27987544 z1_1154.5__648_S4R2a_3 3.578,3 59:38,3 D40
27993032 z1_1154.5__641_S4R2a_2 3.579,8 59:39,8 D40
27996804 z1_1154.5__640_S4R2a_2 3.578,7 59:38,7 D40
28006014 z1_1154.5__639_S4R2a_1 3.101,0 51:41,0 D40/D4114
28013047 z1_1154.5__638_S4R2a_1 2.687,1 44:47,1 D4114
28016302 z1_1154.5__637_S4R2a_2 2.687,2 44:47,2 D4114
28023975 z1_1154.5__636_S4R2a_1 2.687,3 44:47,3 D4114
28026494 z1_1154.5__635_S4R2a_1 2.686,7 44:46,7 D4114
28026500 z1_1154.5__634_S4R2a_1 2.700,6 45:00,6 D4114
28039328 z1_1154.5__633_S4R2a_0 2.687,7 44:47,7 D4114
28041752 z1_1154.5__951_S4R2a_3 2.695,5 44:55,5 D4114
28044714 z1_1154.5__632_S4R2a_2 2.701,1 45:01,1 D4114
28044716 z1_1154.5__631_S4R2a_1 2.686,9 44:46,9 D4114
28048214 z1_1154.5__630_S4R2a_2 2.688,4 44:48,4 D4114
28071267 z1_1154.5__628_S4R2a_1 2.269,4 37:49,4 D4114/S4107
28071270 z1_1154.5__627_S4R2a_1 2.267,9 37:47,9 S4107
28071364 z1_1154.5__626_S4R2a_0 2.268,0 37:48,0 S4107
the crunching time
D4114 / D40 is 75%
S4107 / D40 is 63,5%
Thanks to all who responded.
)
Thanks to all who responded. You have saved me a lot of time testing. I will stick with the S41 for now.