F@H: Speed / benchmarks

KifArU

Senior member
Oct 9, 1999
328
0
0
Since there are no simple ways to benchmark systems running F@H, I started to calculate performance on different computers myself. I have 5 machines running: P2-300, P4 cel 1,5, P4 cel 1,7, P3-1000, XP 1800+@2200 MHz.
For these machines, I already have some numbers. But I need more data.

So if you are running F@H and wanna help, please post the following:
(For a list of workunits and credits: http://folding.stanford.edu/psummary.html )

Example:

Cpu-type: P3 tualatin
Cpu-speed: 1000 MHz
FSB: 4x100
Memory type/size: 256 MB
Mem speed/type: 100 MHz DDR
Dedicated: Yes
Benchmark: 3192

Workunit name: p639_L939_K12M_nat
Credits: 70,9
Optimization: None
Time used: 2,3122 days


Quote-formatting:

Cpu-type:
Cpu-speed:
FSB:
Memory size:
Memory speed:
Dedicated:
Benchmark:

Workunit name:
Credits:
Optimization:
Time used:


Added: Please report the optimizations used. SSE seems to be 15% faster than 3DNow on AMD machines.


Or if you want to help, but don't have time / are lazy, please e-mail me with the machine-information and your logfile(s) at kifaru (at) tkau (dot) nl.

Thanks in advance!
 

ShotgunSi

Senior member
Jun 20, 2001
332
0
0
Hope this helps!

Cpu-type: Amd Athlon XP
Cpu-speed: 1666 Mhz (2000+)
FSB: 133 x 2
Memory size: 1 GB
Memory speed: 200 MHZ DDR (PC 3200)
Dedicated: Yes
Benchmark: ?

Workunit name: p639_L939_K12M_nat
Credits: 70.90
Time used: 6 Minutes 6 Seconds per frame, 400 frames.

*********************************************

Cpu-type: Amd Athlon XP
Cpu-speed: 1666 Mhz (2000+)
FSB: 133 x 2
Memory size: 1 GB
Memory speed: 200 MHZ DDR (PC 3200)
Dedicated: Yes
Benchmark: ?

Workunit name: p365_tip5p_pf2_Mg
Credits: 29.40
Time used: 7 Minutes 4 Seconds per frame, 100 frames.

*********************************************

Cpu-type: Amd Athlon XP
Cpu-speed: 1666 Mhz (2000+)
FSB: 133 x 2
Memory size: 1 GB
Memory speed: 200 MHZ DDR (PC 3200)
Dedicated: Yes
Benchmark: ?

Workunit name: p369_tip5p_pf2_Na
Credits: 29.40
Time used: 7 Minutes 4 Seconds per frame, 100 frames.

*********************************************

Cpu-type: Amd Athlon XP
Cpu-speed: 1533 Mhz (1800+)
FSB: 133 x 2
Memory size: 512 GB
Memory speed: 133 MHZ DDR (PC 2100)
Dedicated: Yes
Benchmark: ?

Workunit name: p363_tip4p_pf2_Mg
Credits: 54.60
Time used: 14 Minutes 28 Seconds per frame, 100 frames.

*********************************************

Cpu-type: Intel Pentium 3 (OC'ED)
Cpu-speed: 933 (7*133)
FSB: 133
Memory size: 256 GB
Memory speed: PC - 133
Dedicated: Yes
Benchmark: ?

Workunit name: p363_tip4p_pf2_Mg
Credits: 29.40
Time used: 12 Minutes 38 Seconds per frame, 100 frames.

*********************************************
Cpu-type: Intel Pentium 3 (OC'ED)
Cpu-speed: 933 (7*133)
FSB: 133
Memory size: 256 GB
Memory speed: PC - 133
Dedicated: Yes
Benchmark: ?

Workunit name: p572_L939_K12M
Credits: 47.00
Time used: 22 Minutes 18 Seconds per frame, 100 frames.

*********************************************

Cpu-type: AMD Athlon (OC'ED)
Cpu-speed: 1100 (10*110)
FSB: 133
Memory size: 512 GB
Memory speed: PC - 133
Dedicated: No
Benchmark: ?

Workunit name: p638_L939_K12M_ext
Credits: 50.00
Time used: 9 Minutes 26 Seconds per frame, 400 frames.


 

ProviaFan

Lifer
Mar 17, 2001
14,993
1
0
This sounds like a very good idea, Kifaru. I'll try to collect data for a few days and post in with my results.
 

OverVolt

Lifer
Aug 31, 2002
14,278
89
91
Seems like F@H prefers a few faster CPU's, and a bunch of slower ones. I'm only running...

P4 @ 3.15Ghz 800fsb w/ HT (counts as 2 CPU's)
P4 @ 2.4Ghz 533fsb
Athlon XP Palimino 1.75ghz @ 170FSB

I think F@H likes the memory bandwidth. Anyways, thats all i'm running and i know i'm putting some decent stats up on the board. And i know some people are running like 3 P3700's ect ect, and aren't getting good results up there.

I'd send you my log file, but it's hard to interpret my results from mt HT enabled CPU.

Can ya get a link to your e-mail i might get around to sending the log.
 

KifArU

Senior member
Oct 9, 1999
328
0
0
Originally posted by: OverVolt
Seems like F@H prefers a few faster CPU's, and a bunch of slower ones. I'm only running... P4 @ 3.15Ghz 800fsb w/ HT (counts as 2 CPU's) P4 @ 2.4Ghz 533fsb Athlon XP Palimino 1.75ghz @ 170FSB I think F@H likes the memory bandwidth. Anyways, thats all i'm running and i know i'm putting some decent stats up on the board. And i know some people are running like 3 P3700's ect ect, and aren't getting good results up there. I'd send you my log file, but it's hard to interpret my results from mt HT enabled CPU. Can ya get a link to your e-mail i might get around to sending the log.

I'll give it a go.
My e-mail is in the first post.
[edit] And my profile is now visible. [/edit]
 

KifArU

Senior member
Oct 9, 1999
328
0
0
After some research in several forums, it seems there are (too) many factors in estimating the speed.
Most important are: Cpu-type (AMD/Intel), OS and optimizations (3DNow+ or SSE) and client-version.
For Tinkers, there's little more than pure speed, as this doesn't use optimizations.
For Gromacs, it's not very clear. FSB doesn't seem to have much influence. But memory bandwidth, cpu-cache can have an effect, but the results from tests are too far apart to draw a conclusion.

For Tinkers, there has been a drastic change in speed. On forum.folding-community.org, there's a thread where everybody has posted the speed for 1 specific protein.
The frame-times for AMD-processors are between 2:10 and 2:20, or within a range of 13-15 minutes.
(All frame-times are scaled to a "per GHz" speed)
So I guess, this is the result of a new client-version.

The average "frametimes per GHz" from this topic:
http://forum.folding-community.org/viewtopic.php?t=1276 (only the p180_enghomeoGB protein)
* means too few results to be trustworthy

AMD: 2:12:48 or 0:13:56 (no matter what AMD-model, duron, T-bird or XP)
Intel P3: 3:30:00 or 0:18:54
Intel P4-M: 4:45:00 or 0:28:32 *
Intel P4: 3:42:55 or ? *


My own testing: The p638_L939_K12M_ext protein (also a Tinker)
(average frametime per GHz, Console client 3.24 with -forceasm flag)

AMD XP T-bred B: 0:10:56
Intel P4 celeron: 0:17:54
Intel P3 tualatin: 0:15:32


Conclusion (for now): AMD are the fastest for tinkers, P3 = 71% of AMD, P4 celeron = 62% of AMD.
If you use the rating for AMD XP instead of true GHz, P3 = 97% of AMD, P4 celeron = 84% of AMD.

If you want to test this specific protein on your cpu, please e-mail me and I'll e-mail you the .dat-file.
 

Slider76

Senior member
Jan 27, 2003
656
7
81
Kifaru: i have times on many proteins folded on my pc (Athlon classic @ 1540mhz=11x140)
with tinker FSB makes no difference i can run 10x140 and i get the same results as if i run 10.5x133, thats to the second from a fresh boot.

The p638_L939_K12M_ext protein (note my times are always based on the last frame of each protein, i have never known them them to fold any quicker or slower than any other frames)
per frame: 6:04
so per ghz thats about 9:21
as long as we dont do a comparison at gormacs, then the old t-bird gets hammerd :brokenheart:

if you want any other frame times i also have them for
p180, p340, p341, p342, p344, p356, p533, p538, p539, p540, p542, p543, p544, p547, p553, p571, P612, P615, P626, P632, P633, p639, P648, P649, p651, p659, p663, p664, p669, p677, p720, p909, p910
that should be it.
 

KifArU

Senior member
Oct 9, 1999
328
0
0
Slider76,

I would like to have those numbers. Could you send them to me or post them here?

Thanks!
 

Slider76

Senior member
Jan 27, 2003
656
7
81
No problems, i will send them to you later today. As it is 1:30am atm and I am sleepy. I have to format all the results properly so it could take some time but I should be able to get them to you some time in the next 10-18 hours.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |