Yes, I did finish the only unit I had, it was uploaded allright, and then I reported it. I shut down BOINC, extracted the files. The executable, the pdb file and the app_info are all in the einstein folder.
No such message about anonymous platform.
If the app_info message isn't there, something went wrong with the process. (I had something similar go wrong installing an optimized SETI app about a week ago.) I fixed it by reinstalling the files. I suggest you do the same. If that doesn't work, try redownloading.
Finished my first WU with v4.24 (and it validated). About a 50% improvement over v4.02 on my Intel(R) Pentium(R) M processor 1.80GHz. Also had a bunch of "small x" messages.
[Now I can get some work done here in XP as well as in Linux (Dual-boot)
Would really be courious to know how the 4.24 compares to the 4.17 on the same machine. If I read you results right, there's still a big difference. Is this consistent with wall-clock time or is there a problem with reporting the CPU-time?
BM
Hello
Well I was planning on crunching through the night, but we had a power outage.
Here are benchmarks from each side of this box:
**Windows XP Service Pack 2 with BOINC 5.5.10 - binary files
1329 floating point MIPS (Whetstone) per CPU
2672 integer MIPS (Dhrystone) per CPU
**Gentoo Linux 2.6.17-r4 with BOINC 5.5.6 - BOINC & deps compiled into system - not binary
875 floating point MIPS (Whetstone) per CPU
2530 integer MIPS (Dhrystone) per CPU
Just for the record, using the standard Windows Einstein the WU takes 48,910.25 and has validated, while the 4.17 in Linux takes 28,112.33. I'm going to boot back into WIndows now and complete that WU with the optimised Einstein and will update, but by the looks of things, times are likely to be on par with what I stated earlier.
Gray
Just a small comment - my guesstimation is from time so far and percentage done - not the BOINC guesstimation, which is way out thus far as I only recently restarted BOINC / Einstein in Windows
Gray
PS: in Windows thus far: 3hours 40 minutes and 43.4 per cent done
OK finally done one WU in Windows using 4.24
Time taken for a big WU is 30,173.91
As compared to a well optimised Linux side with 28,112.33
So - quite a substantial improvement here (48,910.25 was previous Windows WU of similar size)
My counterpart for WU=11223988 (the one I did with 2.42) used 4.02 and both came out valid.
A quantum-leap in performance... Akos did it again (and no "Oops!" this time...)!
;-)
A Pentium M (Banias) 1.4 GHz laptop which had been taking about 55,000 seconds to complete large WU results from one major datafile, took 35,000 seconds to complete its first 4.24 WU, which unfortunately is from a different major datafile. It validated, and stderr_txt has a couple of dozen "small x" lines.
If this is the genuine speedup for the Banias, it appears likely to be considerably better than that for my hyperthreaded Gallatin or old Coppermine, though neither of those has a validated result yet. The one completed Gallatin result has perhaps three dozen "small x" lines.
RE: Just installed 4.24,
)
It probably is. All "work" (tasks) that has once been assigned to 4.02 will be finished with that App. New tasks should get assigned to the 4.24.
BM
BM
RE: Yes, I did finish the
)
If the app_info message isn't there, something went wrong with the process. (I had something similar go wrong installing an optimized SETI app about a week ago.) I fixed it by reinstalling the files. I suggest you do the same. If that doesn't work, try redownloading.
Finished my first WU with
)
Finished my first WU with v4.24 (and it validated). About a 50% improvement over v4.02 on my Intel(R) Pentium(R) M processor 1.80GHz. Also had a bunch of "small x" messages.
Bernd, I have an Athlon
)
Bernd,
I have an Athlon 2700+ that I dual boot.
Just finished the first wu under 4.24.
Results so far:
4.24 22,549
4.02 ~33,700
4.17 ~22,500
4.01 ~25,400
Some serious improvement in the windoze times!
Linux, Win98
Norm
RE: RE: RE: RE: [Now
)
OK finally done one WU in Windows using 4.24
Time taken for a big WU is 30,173.91
As compared to a well optimised Linux side with 28,112.33
So - quite a substantial improvement here (48,910.25 was previous Windows WU of similar size)
Cheers - Gray
RE: A new Beta Test App is
)
Intel Pentium D 840 (default 3.2 GHz, overclocked to 3.5 GHz), Windows XP
- using 4.02: ~ 38500-40000
Results
- using 4.24: ~ 25000
WU-ID:11697697
Improvement: ~ 35%
(Apps switched "half way trough the result": Result ID 38322132, 38294835)
(WU 10992793 erroneously aborted by myself)
Sempron 3000+ WinXP -
)
Sempron 3000+ WinXP
- using 4.02: ~ 41,152 - 42,279
- using 4.24: ~ 27,017
4.24/4.02 0,65
Then you're really interested in a subject, there is no way to avoid it. You have to read the Manual.
P4(Prescott, Mod:F, MMX, SSE,
)
P4(Prescott, Mod:F, MMX, SSE, SSE2, SSE3, EM64T)@3,06 GHz, Win2k(SP4)
4.02: 45,136.58 sec.
4.24: 27,735.03 sec.!!!!
My counterpart for WU=11223988 (the one I did with 2.42) used 4.02 and both came out valid.
A quantum-leap in performance... Akos did it again (and no "Oops!" this time...)!
;-)
MfG
L.
Proud member of the Heise OTF-Team.
A Pentium M (Banias) 1.4 GHz
)
A Pentium M (Banias) 1.4 GHz laptop which had been taking about 55,000 seconds to complete large WU results from one major datafile, took 35,000 seconds to complete its first 4.24 WU, which unfortunately is from a different major datafile. It validated, and stderr_txt has a couple of dozen "small x" lines.
If this is the genuine speedup for the Banias, it appears likely to be considerably better than that for my hyperthreaded Gallatin or old Coppermine, though neither of those has a validated result yet. The one completed Gallatin result has perhaps three dozen "small x" lines.
I have a pentium P4 2.8GHz,
)
I have a pentium P4 2.8GHz, SSE SSE2 SSE3, no HT working the long WU 1005.5
WU
using app 4.24..... 29,324 sec
WU
using app 4.02...... 47,185 sec
If this improvement of 17,861 sec ~5 hrs is due to the work of AKOS and Bernd, I really appreciate it