BOINC benchmark: post results

[D_A]

Just as an aside, I've been surfing a few benchmark threads over multiple forums and in general people do not provide enough information for me to be able to give anything useful back. The only exceptions I've found so far are this thread and the threads I've started at BitBenderForums and TheDCZone. While I haven't tracked EVERY possible variable, I think I've got the major/most important stuff covered.

[randomplatypi]

@D_A

I just downloaded the BIONC beta for Vista 64. Do you want my benchmarks again?

[jcool]

So are you gonna make me go through the sausageman thread looking to see what clock you're running those at?

edit: you mentioned you were waiting for SL8WTs, so I take it those are 2.0GHz chips. I'll assume you're running at stock for now, coz you'd be bragging if you weren't.

Sorry, totally forgot that. Running 166x12=2Ghz so yeah, stock.
A lot of people have tried modding, but no success so far
I'd love to run Sammy at FSB 200.

[The Warlock]

Sorry, totally forgot that. Running 166x12=2Ghz so yeah, stock.
A lot of people have tried modding, but no success so far
I'd love to run Sammy at FSB 200.

Auch, have to start looking in the rearview mirror again for you then .People here just gets harder and harder too pass, and coming up faster and faster on my tail, guess i`m getting old
Hope the Nehalem i have inbound will make a good impact!(still far away though, end of november)

[jcool]

And I still haven't got my 2nd Harper back up
You'll really have to work if I ever do

[The Warlock]

And I still haven't got my 2nd Harper back up
You'll really have to work if I ever do

I have`nt been paying attention, you have two of those beasts?
Yeeeez
Jcool, The Beastmaster

[D_A]

@D_A

I just downloaded the BIONC beta for Vista 64. Do you want my benchmarks again?

If it's a different client on the same hardware/OS then yes please.

Single changes like that make the list more useful as it allows direct comparisons.

[randomplatypi]

QX6700@ 3.5 Ghz
Vista 64 Ultimate
BOINC 6.3.17
10/24/2008 5:55:57 PM||Running CPU benchmarks
10/24/2008 5:56:29 PM||Benchmark results:
10/24/2008 5:56:29 PM|| Number of CPUs: 4
10/24/2008 5:56:29 PM|| 3345 floating point MIPS (Whetstone) per CPU
10/24/2008 5:56:29 PM|| 10220 integer MIPS (Dhrystone) per CPU

[D_A]

Thankee A nice direct comparison between versions.

[D_A]

Does XS have crunching teams besides this one that uses the BOINC system? Perhaps they can be convinced to give up their benchmarks as well?

[loonym]

Not sure if this is of any interest or significance to DA but I'm posting it anyway. q9650@450x9=4050 on vista64 ult sp1




10/25/08 00:11:22||Suspending computation - running CPU benchmarks
10/25/08 00:11:53||Benchmark results:
10/25/08 00:11:53|| Number of CPUs: 4
10/25/08 00:11:53|| 4092 floating point MIPS (Whetstone) per CPU
10/25/08 00:11:53|| 11971 integer MIPS (Dhrystone) per CPU
10/25/08 00:11:54||Resuming computation

10/25/2008 12:13:22 AM||Version change (6.3.15 -> 6.3.17)

10/25/2008 12:13:22 AM||Running CPU benchmarks
10/25/2008 12:13:22 AM||Suspending computation - running CPU benchmarks
10/25/2008 12:13:53 AM||Benchmark results:
10/25/2008 12:13:53 AM|| Number of CPUs: 4
10/25/2008 12:13:53 AM|| 4085 floating point MIPS (Whetstone) per CPU
10/25/2008 12:13:53 AM|| 11972 integer MIPS (Dhrystone) per CPU



edit: pretty ho-hum, the only thing interesting to me about it is that it changed itself from a 24hr clock to 12

[jcool]

I have`nt been paying attention, you have two of those beasts?
Yeeeez
Jcool, The Beastmaster

Actually I have 3 pairs of Harpers. Gave one away for a good friend tho.

[D_A]

Not sure if this is of any interest or significance to DA but I'm posting it anyway. q9650@450x9=4050 on vista64 ult sp1




10/25/08 00:11:22||Suspending computation - running CPU benchmarks
10/25/08 00:11:53||Benchmark results:
10/25/08 00:11:53|| Number of CPUs: 4
10/25/08 00:11:53|| 4092 floating point MIPS (Whetstone) per CPU
10/25/08 00:11:53|| 11971 integer MIPS (Dhrystone) per CPU
10/25/08 00:11:54||Resuming computation

10/25/2008 12:13:22 AM||Version change (6.3.15 -> 6.3.17)

10/25/2008 12:13:22 AM||Running CPU benchmarks
10/25/2008 12:13:22 AM||Suspending computation - running CPU benchmarks
10/25/2008 12:13:53 AM||Benchmark results:
10/25/2008 12:13:53 AM|| Number of CPUs: 4
10/25/2008 12:13:53 AM|| 4085 floating point MIPS (Whetstone) per CPU
10/25/2008 12:13:53 AM|| 11972 integer MIPS (Dhrystone) per CPU



edit: pretty ho-hum, the only thing interesting to me about it is that it changed itself from a 24hr clock to 12

Loonym, this is EXACTLY the kind of thing I'm interested in. Being able to compare one instance to another with only one change is great! By only changing one variable at a time we can get a better idea of what changes make what difference.

[[XC] serlv]

CPU Type 2x QuadCore Intel Xeon X5460, 3800 MHz (9.5 x 400)

10/25/2008 4:29:32 AM|| 3889 floating point MIPS (Whetstone) per CPU
10/25/2008 4:29:32 AM|| 11312 integer MIPS (Dhrystone) per CPU

~~~~
Trying to iron things out though. WUs are taking a rediculously large amount of time. Something isn't right.

[D_A]

No power/thermal throttling turned on?

... and what OS and BOINC version are you running there, Scott?

[[XC] serlv]

XP 64-bit and BOINC v6.2.19 for Windows 64-bit.

C1E and Speedstep is turned off in BIOS.

[jcool]

Scott,

if this is on the Z7S, make sure you have "SF Bypass" on "enable" in Bios. Which is the default setting anyway. Snoop filter is bugged and makes this mobo run slow as hell.

For reference:

HCC completion times should be around or below 3h for 8 WUs.
430x9 took around 2:52, 451x9 takes around 2:45. High FSB helps a lot tho.

[D_A]

XP 64-bit and BOINC v6.2.19 for Windows 64-bit.

C1E and Speedstep is turned off in BIOS.

Sheet updated

[jcool]

D_A,

can you maybe PM the OP and ask him if he'd link your boinc benchmark list in post #1?
So I don't have to search all the time when it's updated

[D_A]

Might do that.

... and since I forgot to mention it, it was this thread that inspired me to start the list in the first place.

[[XC] serlv]

D_A,

can you maybe PM the OP and ask him if he'd link your boinc benchmark list in post #1?
So I don't have to search all the time when it's updated

Where is this spreadsheet/list? Is there a URL for it, maybe?
Can it be put on the Google app? Would that do us any good?

[jcool]

This is what we are talking about.

D_A, you wrote XP home every time XP x64 comes up, might wanna fix that
Since there is only XP x64 Professional

Also, what is up with that Desktop/Server classification? Any Xeon CPU is a server CPU, Sossaman as well as the Clovers, Harpers and Opterons. However, Q6600 should be Desktop.

[[XC] serlv]

This is what we are talking about.

Ahh. TY, jcool.

Here is another result...

Q9550 ~ XP 64-bit SP2 ~ 3926.1 MHz (8.5 x 461.9 MHz)

10/25/2008 5:39:30 PM|| 3988 floating point MIPS (Whetstone) per CPU
10/25/2008 5:39:30 PM|| 11540 integer MIPS (Dhrystone) per CPU

[jcool]

So how are completion times on the Harper now?

[[XC] serlv]

another result...

X5350 @ 2.66GHz ~ Windows Server 2003, Enterprise Edition, 64-bit - BOINC 64-bit v6.2.18

10/25/2008 5:49:30 PM|| 2535 floating point MIPS (Whetstone) per CPU
10/25/2008 5:49:30 PM|| 8255 integer MIPS (Dhrystone) per CPU