NEW CORE for gpu client Thread. Updated regularly.

[[XC] riptide]

New Core 14. Report findings.

[p2501]

Stuttering. Outlandish points and fluctuations.

Projekt : 5900
Core : Unknown
Frames : 100
Punkte : 420


-- 295-G0 --

Minimale Frame-Dauer : 37s - 9807.57 ppd
Gemittelte Frame-Dauer : 44s - 8247.27 ppd
Aktuelle Frame-Dauer : 51s - 7115.29 ppd
L3F-Frame-Dauer : 49s - 7405.71 ppd
Effektive Frame-Dauer : 48s - 7560.00 ppd


-- 295-G1 --

Minimale Frame-Dauer : 25s - 14515.20 ppd
Gemittelte Frame-Dauer : 32s - 11340.00 ppd
Aktuelle Frame-Dauer : 25s - 14515.20 ppd
L3F-Frame-Dauer : 35s - 10368.00 ppd
Effektive Frame-Dauer : 38s - 9549.47 ppd

Near 15k PPD from one core.

[Sparky]

I set -advmethods on one 8800GS on a dedicated cruncher. Will see if it gets a 5900 and core 14 and see how it does.

[MikeB12]

I cleared benchmarks.dat so these are on one right now. and they are picking them up with no -advmethods now..

Project : 5900
Core : Unknown
Frames : 100
Credit : 420


-- X4-GPU2-8800GT-1782 --

Min. Time / Frame : 1mn 33s - 3901.94 ppd
Avg. Time / Frame : 1mn 33s - 3901.94 ppd
Cur. Time / Frame : 1mn 33s - 3901.94 ppd
R3F. Time / Frame : 1mn 33s - 3901.94 ppd
Eff. Time / Frame : 1mn 32s - 3944.35 ppd


-- X3-GPU1-9600GSO-1728 --

Min. Time / Frame : 1mn 36s - 3780.00 ppd
Avg. Time / Frame : 1mn 36s - 3780.00 ppd
Cur. Time / Frame : 1mn 36s - 3780.00 ppd
R3F. Time / Frame : 1mn 36s - 3780.00 ppd
Eff. Time / Frame : 1mn 38s - 3702.86 ppd


-- X5-GPU2-9800GX2-1783 --

Min. Time / Frame : 1mn 14s - 4903.78 ppd
Avg. Time / Frame : 1mn 14s - 4903.78 ppd
Cur. Time / Frame : 1mn 14s - 4903.78 ppd
R3F. Time / Frame : 1mn 14s - 4903.78 ppd
Eff. Time / Frame : 1mn 14s - 4903.78 ppd

[DAK1640]

Like the avatar Mike...Well, what's the verdict on this new core???

[OldChap]

Since these started I've had a couple of instances of my rig re-starting but I'm not sure that this is related.

My numbers....

Project : 5900
 Core    : Unknown
 Frames  : 100
 Credit  : 420


 -- 260-1 --

 Min. Time / Frame : 56s  - 6480.00 ppd
 Avg. Time / Frame : 57s  - 6366.32 ppd
 Cur. Time / Frame : 1mn 02s  - 5852.90 ppd
 R3F. Time / Frame : 1mn 02s  - 5852.90 ppd
 Eff. Time / Frame : 1mn 08s  - 5336.47 ppd


 -- 260-2 --

 Min. Time / Frame : 55s  - 6597.82 ppd
 Avg. Time / Frame : 59s  - 6150.51 ppd
 Cur. Time / Frame : 1mn 06s  - 5498.18 ppd
 R3F. Time / Frame : 1mn 06s  - 5498.18 ppd
 Eff. Time / Frame : 1mn 33s  - 3901.94 ppd


 -- gx2-1 1 --

 Min. Time / Frame : 56s  - 6480.00 ppd
 Avg. Time / Frame : 1mn 01s  - 5948.85 ppd
 No Cur. Time / Frame
 No R3F. Time / Frame
 No Eff. Time / Frame


 -- gx2-1 2 --

 Min. Time / Frame : 53s  - 6846.79 ppd
 Avg. Time / Frame : 1mn 06s  - 5498.18 ppd
 No Cur. Time / Frame
 No R3F. Time / Frame
 No Eff. Time / Frame


 -- gx2-2 1 --

 Min. Time / Frame : 1mn 00s  - 6048.00 ppd
 Avg. Time / Frame : 1mn 06s  - 5498.18 ppd
 Cur. Time / Frame : 1mn 08s  - 5336.47 ppd
 R3F. Time / Frame : 1mn 08s  - 5336.47 ppd
 Eff. Time / Frame : 1mn 09s  - 5259.13 ppd


 -- gtx --

 Min. Time / Frame : 1mn 04s  - 5670.00 ppd
 Avg. Time / Frame : 1mn 09s  - 5259.13 ppd
 No Cur. Time / Frame
 No R3F. Time / Frame
 No Eff. Time / Frame


 -- GX2-2 2 --

 Min. Time / Frame : 1mn 07s  - 5416.12 ppd
 Avg. Time / Frame : 1mn 10s  - 5184.00 ppd
 No Cur. Time / Frame
 No R3F. Time / Frame
 No Eff. Time / Frame

[Sparky]

8800GS is 3500 PPD right now.

[WFO]

Almost 24hrs since I put the -advmethods flag in all my gpu clients. Fahmon is showing an average of about 16k pd increase for 14 clients. I can only hope that's accurate! I think it is. I'm already at 85K with 2 updates. Hopefully 115K + for today.

 Project : 5900
 Core    : Unknown
 Frames  : 100
 Credit  : 420


 -- 9450-1 gpu 0 --

 Min. Time / Frame : 1mn 24s  - 4320.00 ppd
 Avg. Time / Frame : 1mn 25s  - 4269.18 ppd
 Cur. Time / Frame : 1mn 47s  - 3391.40 ppd
 R3F. Time / Frame : 1mn 51s  - 3269.19 ppd
 Eff. Time / Frame : 1mn 46s  - 3423.40 ppd


 -- 9450-1 gpu 1 --

 Min. Time / Frame : 1mn 14s  - 4903.78 ppd
 Avg. Time / Frame : 1mn 25s  - 4269.18 ppd
 Cur. Time / Frame : 1mn 32s  - 3944.35 ppd
 R3F. Time / Frame : 1mn 32s  - 3944.35 ppd
 Eff. Time / Frame : 1mn 35s  - 3819.79 ppd


 -- 9450-2 gpu 0 --

 Min. Time / Frame : 56s  - 6480.00 ppd
 Avg. Time / Frame : 1mn 00s  - 6048.00 ppd
 Cur. Time / Frame : 1mn 11s  - 5110.99 ppd
 R3F. Time / Frame : 1mn 10s  - 5184.00 ppd
 Eff. Time / Frame : 1mn 13s  - 4970.96 ppd


 -- 9450-2 gpu 1 --

 Min. Time / Frame : 42s  - 8640.00 ppd
 Avg. Time / Frame : 51s  - 7115.29 ppd
 Cur. Time / Frame : 1mn 15s  - 4838.40 ppd
 R3F. Time / Frame : 1mn 15s  - 4838.40 ppd
 Eff. Time / Frame : 1mn 15s  - 4838.40 ppd


 -- 9450-2 gpu-2 --

 Min. Time / Frame : 1mn 05s  - 5582.77 ppd
 Avg. Time / Frame : 1mn 05s  - 5582.77 ppd
 Cur. Time / Frame : 1mn 18s  - 4652.31 ppd
 R3F. Time / Frame : 1mn 15s  - 4838.40 ppd
 Eff. Time / Frame : 1mn 15s  - 4838.40 ppd


 -- 9450-2 gpu 3 --

 Min. Time / Frame : 1mn 04s  - 5670.00 ppd
 Avg. Time / Frame : 1mn 05s  - 5582.77 ppd
 Cur. Time / Frame : 1mn 16s  - 4774.74 ppd
 R3F. Time / Frame : 1mn 15s  - 4838.40 ppd
 Eff. Time / Frame : 1mn 13s  - 4970.96 ppd


 -- G0-4-gpu 0 --

 Min. Time / Frame : 58s  - 6256.55 ppd
 Avg. Time / Frame : 58s  - 6256.55 ppd
 Cur. Time / Frame : 1mn 02s  - 5852.90 ppd
 R3F. Time / Frame : 1mn 02s  - 5852.90 ppd
 Eff. Time / Frame : 1mn 03s  - 5760.00 ppd


 -- G0-4 gpu 1 --

 Min. Time / Frame : 33s  - 10996.36 ppd
 Avg. Time / Frame : 2mn 23s  - 2537.62 ppd
 Cur. Time / Frame : 1mn 26s  - 4219.53 ppd
 R3F. Time / Frame : 1mn 29s  - 4077.30 ppd
 Eff. Time / Frame : 1mn 32s  - 3944.35 ppd


 -- G0-4 gpu-2 --

 Min. Time / Frame : 47s  - 7720.85 ppd
 Avg. Time / Frame : 57s  - 6366.32 ppd
 Cur. Time / Frame : 59s  - 6150.51 ppd
 R3F. Time / Frame : 59s  - 6150.51 ppd
 Eff. Time / Frame : 1mn 15s  - 4838.40 ppd


 -- G0-4 gpu 3 --

 Min. Time / Frame : 56s  - 6480.00 ppd
 Avg. Time / Frame : 57s  - 6366.32 ppd
 Cur. Time / Frame : 1mn 09s  - 5259.13 ppd
 R3F. Time / Frame : 1mn 09s  - 5259.13 ppd
 Eff. Time / Frame : 1mn 09s  - 5259.13 ppd


 -- G0-2-gpu-0 --

 Min. Time / Frame : 1mn 28s  - 4123.64 ppd
 Avg. Time / Frame : 1mn 31s  - 3987.69 ppd
 Cur. Time / Frame : 1mn 44s  - 3489.23 ppd
 R3F. Time / Frame : 1mn 39s  - 3665.45 ppd
 Eff. Time / Frame : 1mn 38s  - 3702.86 ppd


 -- G0-2 gpu 1 --

 Min. Time / Frame : 1mn 29s  - 4077.30 ppd
 Avg. Time / Frame : 1mn 31s  - 3987.69 ppd
 Cur. Time / Frame : 1mn 36s  - 3780.00 ppd
 R3F. Time / Frame : 1mn 39s  - 3665.45 ppd
 Eff. Time / Frame : 1mn 39s  - 3665.45 ppd


 -- I-7 gpu 0 --

 Min. Time / Frame : 26s  - 13956.92 ppd
 Avg. Time / Frame : 39s  - 9304.62 ppd
 Cur. Time / Frame : 49s  - 7405.71 ppd
 R3F. Time / Frame : 47s  - 7720.85 ppd
 Eff. Time / Frame : 46s  - 7888.70 ppd


 -- AMD 0 --

 Min. Time / Frame : 1mn 01s  - 5948.85 ppd
 Avg. Time / Frame : 1mn 02s  - 5852.90 ppd
 Cur. Time / Frame : 1mn 13s  - 4970.96 ppd
 R3F. Time / Frame : 1mn 11s  - 5110.99 ppd
 Eff. Time / Frame : 1mn 11s  - 5110.99 ppd


 -- AMD 1 --

 Min. Time / Frame : 58s  - 6256.55 ppd
 Avg. Time / Frame : 1mn 02s  - 5852.90 ppd
 Cur. Time / Frame : 1mn 19s  - 4593.42 ppd
 R3F. Time / Frame : 1mn 16s  - 4774.74 ppd
 Eff. Time / Frame : 1mn 17s  - 4712.73 ppd


 -- AMD 2 --

 Min. Time / Frame : 58s  - 6256.55 ppd
 Avg. Time / Frame : 1mn 01s  - 5948.85 ppd
 Cur. Time / Frame : 1mn 12s  - 5040.00 ppd
 R3F. Time / Frame : 1mn 16s  - 4774.74 ppd
 Eff. Time / Frame : 1mn 10s  - 5184.00 ppd


 -- AMD 3 --

 Min. Time / Frame : 1mn 01s  - 5948.85 ppd
 Avg. Time / Frame : 1mn 02s  - 5852.90 ppd
 Cur. Time / Frame : 1mn 19s  - 4593.42 ppd
 R3F. Time / Frame : 1mn 12s  - 5040.00 ppd
 Eff. Time / Frame : 1mn 11s  - 5110.99 ppd


 -- I-7 gpu 1 --

 Min. Time / Frame : 27s  - 13440.00 ppd
 Avg. Time / Frame : 39s  - 9304.62 ppd
 Cur. Time / Frame : 44s  - 8247.27 ppd
 R3F. Time / Frame : 45s  - 8064.00 ppd
 Eff. Time / Frame : 42s  - 8640.00 ppd


 -- G0-3 gpu 1 --

 Min. Time / Frame : 39s  - 9304.62 ppd
 Avg. Time / Frame : 1mn 16s  - 4774.74 ppd
 Cur. Time / Frame : 1mn 56s  - 3128.28 ppd
 R3F. Time / Frame : 1mn 51s  - 3269.19 ppd
 Eff. Time / Frame : 1mn 58s  - 3075.25 ppd


 -- G0-3-gpu-2 --

 Min. Time / Frame : 1mn 07s  - 5416.12 ppd
 Avg. Time / Frame : 1mn 16s  - 4774.74 ppd
 Cur. Time / Frame : 1mn 40s  - 3628.80 ppd
 R3F. Time / Frame : 1mn 50s  - 3298.91 ppd
 Eff. Time / Frame : 1mn 39s  - 3665.45 ppd


 -- G0-3 gpu 3 --

 Min. Time / Frame : 1mn 08s  - 5336.47 ppd
 Avg. Time / Frame : 1mn 09s  - 5259.13 ppd
 Cur. Time / Frame : 1mn 14s  - 4903.78 ppd
 R3F. Time / Frame : 1mn 13s  - 4970.96 ppd
 Eff. Time / Frame : 1mn 12s  - 5040.00 ppd


 -- G0-3 gpu 0 --

 Min. Time / Frame : 1mn 04s  - 5670.00 ppd
 Avg. Time / Frame : 1mn 07s  - 5416.12 ppd
 Cur. Time / Frame : 1mn 13s  - 4970.96 ppd
 R3F. Time / Frame : 1mn 12s  - 5040.00 ppd
 Eff. Time / Frame : 1mn 14s  - 4903.78 ppd

[Sparky]

I'm about ready to put advmethods on all mine. Been going smoothly on the one 8800GS so far. Sure it scores lower than some of the 300-something point units, but those are few, usually I get the 511s which drop down to under 3000 ppd. So a consistent 3550 ppd per card isn't bad at all. CPU usage is still 0% typically (XP) and temps are much lower too.

[coo-coo-clocker]

Hmmm, not so good:

Project:5900, Core:Unknown, Frames:100, Credit:420

-- 8800 GT 1728 Drv 180.48 --
Avg. Time / Frame : 1mn 42s - 3557.65 ppd

-- 9800 GX2 1716 Drv 182.6 --
Avg. Time / Frame : 1mn 18s - 4652.31 ppd

-- 9800 GX2 1716 Drv 182.6 --
Avg. Time / Frame : 1mn 19s - 4593.42 ppd

I think I'm doing better with more 5768s..

[Kondik]

That 5900 project is nuts tbh

From 7000 PPD to 10.000 on one card ... flies up and down like a poop in a tornado ...
One WU from this project is done in an hour and heats the card to 68 degrees , one takes double the time and heats the card to 85 degrees celsia
and both are 420 points .... what the eck ?

[MikeB12]

poop in a tornado

[WFO]

That 5900 project is nuts tbh

From 7000 PPD to 10.000 on one card ... flies up and down like a poop in a tornado ...
One WU from this project is done in an hour and heats the card to 68 degrees , one takes double the time and heats the card to 85 degrees celsia
and both are 420 points .... what the eck ?

Mike explained the "poop in a tornado effect" in the last race with the benchmark dart board.

[Marvin_The_Martian]

Kondik are you sure it was 5900? It should not take double the time, not even close, variations could be 25% or so Also the temperature does not fluctuate like that with this same wu. It sound like you got another WU like the 5900 you're not supposed to get at this time

Which also explains why your scores are acting like poop in a tornado

[MikeB12]

lol, WFO, I guess the correct term now is "Poop darts in a tornado"..

[Kondik]

Remember kids, in tornado even a poop can be a deadly projectile that can blast a hole in a tank ! Deadly Poop ...

[DR3]

Noticed that this weekend my Nvidia GPU clients went from Core 11 to Core 14.
Have not had problems.

Is everyone else with Nvidia Windows clients seeing the version 14 cores?

[OldChap]

There are 2 cores in use core 11 and core 14. However, core 11 is now in 2 flavours the original, which runs your cards hot and a variant http://www.stanford.edu/~friedrim/FahCore_11.exe which you can read about at http://foldingforum.org/viewtopic.php?f=52&t=10404 It is supposed to alow you to reduce gpu run time in order to control temps (have not really tried yet)

[DR3]

Found the answer to the core 11 or 14.

Core 14 doesn't produce more science, it produces different science... otherwise all the projects would be using that core already. Suggesting replacing Core 11 with Core 14 is like replacing CPU Core 78 (the boggo Gromacs core) with CPU Core 80 (the Gromacs SREM core)... they serve different purposes even if they're based on the same package.

http://foldingforum.org/viewtopic.php?f=52&t=10128

So core 14 is not an improvement on 11, it is just designed to run different projects.

[570091D]

GPU3 CORE

i haven't found anything posted in here about the GPU3 core for fermi gpus... are there no other folders with fermi cards?!?

Stanford GPU3 thread

We have a new GPU core (core 15) going into an open beta test for NVIDIA clients. This core requires a new client (see below) as well as the latest drivers (197.45). This core is the first run of the GPU3 technology, derived from the OpenMM project at Stanford (http://simtk.org/home/openmm). You can find more information in our GPU3 FAQ (see url below).

While this release is for NVIDIA only to start, we are actively pushing ATI support (with the help of AMD/ATI), although we have no ETA at the moment. We do not recommend that you use this client with an ATI GPU at the moment since there is no advantage and you are potentially exposing yourself to new bugs.

This is the first open beta test of this new client and core, so there are likely bugs to be found as more donors try this out on more diverse sets of hardware. Also, the documentation (GPU3 FAQ) is new too and there are possibly some errors there too. However, the client has been QA'd both internally at Stanford and with our closed group of beta testers and is looking pretty good so far.

Some testers in the closed beta test have found problems with 8800 and 9800 class GPUs (we are working on this).