So that enhancement is disabled in all other consumer cards?

I'd rather say "this functionality is not present in other consumer cards".

MrS

---

Scanning for our furry friends since Jan 2002

We already have a rating system for video cards that is broken down into 4 categories: most recommended, highly recommended, recommended and not recommend. So, shouldn't we have the same for video memory size?

most recommended : 4 GB +
highly recommended : 2 to 4 GB
recommended: 1 to 2 GB
not recommended : less than 1 GB

That is a good idea. I'll suggest the following classification:

- sufficient to run all current tasks at maximum speed
- sufficient to run short runs at maximum speed
- sufficient to run short runs
- insufficient

These boundaries would change with apps and WUs and we'd have a hard time figuring them out, so the developers would have to take care of updating this table.

MrS

---

Scanning for our furry friends since Jan 2002

acemd uses two types of algorithms, one faster and more memory expensive and one slower that uses less memory. The decision is made at dynamically depending on the gpu memory you have available.

Most of the simulations are small enough to use the faster algorithm these days with 2GB of memory, but some might still end up higher.

Anyway, as soon as we get one, we will report the performance.

gdf

I can give you a remote access to my gtx titan if it helps.

We already have a rating system for video cards that is broken down into 4 categories: most recommended, highly recommended, recommended and not recommend. So, shouldn't we have the same for video memory size?

most recommended : 4 GB +
highly recommended : 2 to 4 GB
recommended: 1 to 2 GB
not recommended : less than 1 GB

That is a good idea. I'll suggest the following classification:

- sufficient to run all current tasks at maximum speed
- sufficient to run short runs at maximum speed
- sufficient to run short runs
- insufficient

Yikes, it looks like this project might be getting a little rich for my pocketbook, or are you not saying that we should need 4GB cards to do the long runs? Maybe you're not making a 1 to 1 relationship for the 2 classifications above. Right now the long run TONI WUs do OK on the GTX 460 768mb cards (about 12.5 hours). The older long run NATHANS were slow but the last one I received was smaller and ran fine. The 650 TI 1gb card seems to run everything fine currently.

Sorry if this wasn't clear: Bedrich gave an example of what he had in mind. I proposed different categories, also 4 by accident, totally unrelated to the number he wrote down for his categories. I didn't give any numbers because I don't know them, but I'm sure they'd be smaller than what Bedrich had in mind.

MrS

---

Scanning for our furry friends since Jan 2002

The analyst and cruncher:
In theory you could run one app across 4 Titans and use up to 24GB, and at the other extreme you could run more than on task on a 1GB (or less) card. It's really down to the research parameters. While I expect some in-house usages to be more exploratory, the live apps need to accommodate the main resources (our GPU's). For this reason the researchers work out what cards are available and how many are attached... With that in mind they develop the apps and perform research that utilizes the majority of what's available.

If there are enough 3GB cards available then perhaps some research will be designed to utilize from 2GB up to 3GB, and the tasks will prefer 3GB cards. Just how much they will prefer 3GB is a question in itself, and an unknown. The performance difference between a 2GB card and a 3GB card might be large or negligible, and the difference in price + running cost may or may not make the 3GB card the choicest. If there are enough 4GB or 6GB cards, ditto...

There is a ~£30 difference between the GTX660ti 2GB and 3GB; ~£180 vs ~£210. It's unknown if tasks would be 15% faster (the price difference) just because of the video memory capacity. More GDDR might also require more power, which would mean a slight loss of power to the rest of the card (capped at 150W) and thus the GPU speed might be a bit less. So the real difference to make up might be 20% from the GPU-only point of view. However, if you add the system overheads and price tag, and the perspective might change again.

The Scientist:
I'm all for using more GDDR5; it would expand the scientific boundaries (thee scientific goal) and we could be looking at larger molecular interactions, or the same interactions in more detail. This expansion of science obviously completely outweighs the projects performance in terms of steps or performance differences between 2GB and 3GB. It should thus be a key project goal.

---

FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

First test ends with error for short and long 4.2

<message>
- exit code -9 (0xfffffff7)
</message>

OD !!

You're saying WUs are not running on your Titan?

Mine's in the mail.

This doesn't bode well....

Operator

---

Yes. Maybe application has to be updated for titan.

It's possible that it does not run until we get one here.

gdf

Ah, -9, my favourite error message. It's an ongoing mystery where this comes from, since there's nothing in the application source that causes a termination with that exit code.

HA-SOFT: Since your machine with the Titan seems to be new (wasn't crunching before with an older GPU), my guess is that there is something about its setup that is causing a problem for the app (or possibly the BOINC client) making it immediately abort. I have no idea what that might be, but if you are able to get it working by playing about with the configuration, please do let me know.

MJH

I have tested 7.0.28, 7.0.52, project reset, no luck.

boinc log:

GPUGRID | [cpu_sched_debug] Ann145_r2-TONI_AGGd8-13-100-RND9865_0 sched state 0 next 2 task state 0
28.2.2013 13:10:10 | GPUGRID | Starting task Ann145_r2-TONI_AGGd8-13-100-RND9865_0 using acemdlong version 618 (cuda42) in slot 10
28.2.2013 13:10:10 | GPUGRID | [css] running Ann145_r2-TONI_AGGd8-13-100-RND9865_0 (0.81 CPUs + 1 NVIDIA GPU)
28.2.2013 13:10:10 | | [cpu_sched_debug] enforce_schedule: end
28.2.2013 13:10:10 | | [network_status] status: online
28.2.2013 13:10:10 | | [slot] removed file slots/10/init_data.xml
28.2.2013 13:10:10 | GPUGRID | [slot] linked ../../projects/www.gpugrid.net/Ann145_r2-TONI_AGGd8-13-LICENSE to slots/10/LICENSE
28.2.2013 13:10:10 | GPUGRID | [slot] linked ../../projects/www.gpugrid.net/Ann145_r2-TONI_AGGd8-13-COPYRIGHT to slots/10/COPYRIGHT
28.2.2013 13:10:10 | GPUGRID | [slot] linked ../../projects/www.gpugrid.net/Ann145_r2-TONI_AGGd8-13-pdb_file to slots/10/structure.pdb
28.2.2013 13:10:10 | GPUGRID | [slot] linked ../../projects/www.gpugrid.net/Ann145_r2-TONI_AGGd8-13-psf_file to slots/10/structure.psf
28.2.2013 13:10:10 | GPUGRID | [slot] linked ../../projects/www.gpugrid.net/Ann145_r2-TONI_AGGd8-13-par_file to slots/10/parameters
28.2.2013 13:10:10 | GPUGRID | [slot] linked ../../projects/www.gpugrid.net/Ann145_r2-TONI_AGGd8-13-conf_file_enc to slots/10/input
28.2.2013 13:10:10 | GPUGRID | [slot] linked ../../projects/www.gpugrid.net/Ann145_r2-TONI_AGGd8-13-100-RND9865_0_0 to slots/10/progress.log
28.2.2013 13:10:10 | GPUGRID | [slot] linked ../../projects/www.gpugrid.net/Ann145_r2-TONI_AGGd8-13-100-RND9865_0_1 to slots/10/output.coor
28.2.2013 13:10:10 | GPUGRID | [slot] linked ../../projects/www.gpugrid.net/Ann145_r2-TONI_AGGd8-13-100-RND9865_0_2 to slots/10/output.vel
28.2.2013 13:10:10 | GPUGRID | [slot] linked ../../projects/www.gpugrid.net/Ann145_r2-TONI_AGGd8-13-100-RND9865_0_3 to slots/10/output.idx
28.2.2013 13:10:10 | GPUGRID | [slot] linked ../../projects/www.gpugrid.net/Ann145_r2-TONI_AGGd8-13-100-RND9865_0_4 to slots/10/output.dcd
28.2.2013 13:10:10 | GPUGRID | [slot] linked ../../projects/www.gpugrid.net/Ann145_r2-TONI_AGGd8-13-100-RND9865_0_8 to slots/10/output.vel.dcd
28.2.2013 13:10:11 | GPUGRID | [sched_op] Deferring communication for 2 min 56 sec
28.2.2013 13:10:11 | GPUGRID | [sched_op] Reason: Unrecoverable error for task Ann145_r2-TONI_AGGd8-13-100-RND9865_0
28.2.2013 13:10:11 | | [slot] cleaning out slots/10: handle_exited_app()
28.2.2013 13:10:11 | | [slot] removed file slots/10/acemd.2865P.exe
28.2.2013 13:10:11 | | [slot] removed file slots/10/boinc_lockfile
28.2.2013 13:10:11 | | [slot] removed file slots/10/COPYRIGHT
28.2.2013 13:10:11 | | [slot] removed file slots/10/cudart32_42_9.dll
28.2.2013 13:10:11 | | [slot] removed file slots/10/cufft32_42_9.dll
28.2.2013 13:10:11 | | [slot] removed file slots/10/init_data.xml
28.2.2013 13:10:11 | | [slot] removed file slots/10/input
28.2.2013 13:10:11 | | [slot] removed file slots/10/input.coor
28.2.2013 13:10:11 | | [slot] removed file slots/10/input.idx
28.2.2013 13:10:11 | | [slot] removed file slots/10/input.vel
28.2.2013 13:10:11 | | [slot] removed file slots/10/LICENSE
28.2.2013 13:10:11 | | [slot] removed file slots/10/META_INP
28.2.2013 13:10:11 | | [slot] removed file slots/10/output.coor
28.2.2013 13:10:11 | | [slot] removed file slots/10/output.dcd
28.2.2013 13:10:11 | | [slot] removed file slots/10/output.idx
28.2.2013 13:10:11 | | [slot] removed file slots/10/output.vel
28.2.2013 13:10:11 | | [slot] removed file slots/10/output.vel.dcd
28.2.2013 13:10:11 | | [slot] removed file slots/10/parameters
28.2.2013 13:10:11 | | [slot] removed file slots/10/progress.log
28.2.2013 13:10:11 | | [slot] removed file slots/10/stderr.txt
28.2.2013 13:10:11 | | [slot] removed file slots/10/structure.pdb
28.2.2013 13:10:11 | | [slot] removed file slots/10/structure.psf
28.2.2013 13:10:11 | | [slot] removed file slots/10/tcl85.dll
28.2.2013 13:10:11 | | [cpu_sched_debug] Request CPU reschedule: application exited
28.2.2013 13:10:11 | GPUGRID | Computation for task Ann145_r2-TONI_AGGd8-13-100-RND9865_0 finished

First test ends with error for short and long 4.2
<message>
- exit code -9 (0xfffffff7)
</message>

Ouch, bleeding edge strikes again. Sorry to hear it and hope there's a resolution soon.

Installed my titan last night - no luck with gpugrid.

I loaded up gpuz and the cuda box isn't ticked. Sounds like no cuda drivers yet.


-- Craig

314.09 is CUDA 5 cappable.

Are you using remote desktop?

The Titan card is installed in a machine with a GTX580. I wonder if that's the issue. Can I embed a screenshot in this forum?

Not using remote desktop.

-- Craig

This may be the case. Driver 314.09 support Titan only. Older drivers do not support titan.

Can I embed a screenshot in this forum?

If you don't have your own HTML server or a wqebsite such as most ISPs include with your ISP service, you can upload the image to a free image hosting service like photobucket. They will give you a URL to the image. Then put the URL in a post here. You can embed it by using BBCode tags which are similar to HTML tags. While composing the message find the link that says "Use BBCode tags to format your text" to the left of the edit box and up a little bit to get a summary of what the various tags do. If you can't make the BBCode tags work then just post the URL, we can copy 'n paste it into a browser address bar.

---

BOINC <<--- credit whores, pedants, alien hunters

nucleon;

I will be installing my Titan in the same machine my Tesla is in.

I have already downloaded the Titan specific driver (314.09) from Nvidia.

Did you manage to get yours working along with your GTX 580?

Operator

---

We need to refresh the app for Titan. Probably get this done on Monday.

MJH