I also received a short GIANNI, and it errored out after 1 second on my Titan:

I16R72-GIANNI_VIL1-1-3-RND0548_0
-9 (0xfffffffffffffff7) Unknown error number

GIANNI_VIL1 are not beta tasks and should fail on titans.
GIANNI_TEST are beta tasks.

gdf

Operator, Uncheck the Project Preferences option to run other WU's, if no work is available for your chosen apps,

If no work for selected applications is available, accept work from other applications? NO

Don't know if I did this correctly but I set my machine with the two Titans to run test apps only (no long, short, beta boxes ticked), and it downloaded two WUs, a long and a short.

From client_state.xml;

I1R93-GIANNI_VIL1-2-3-RND5226_0 computation error and...
<![CDATA[ <message> (unknown error) - exit code -9 (0xfffffff7) </message> <stderr_txt> </stderr_txt> ]]>


I22R1-NATHAN_KIDc22_2-0-8-RND8049_0 computation error and...
<![CDATA[ <message> (unknown error) - exit code -9 (0xfffffff7) </message> <stderr_txt> </stderr_txt> ]]>

So, no joy today.

Operator

---

FAQ's

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

My Titan is ready but no beta tasks.

I received 1 on my 660TI. Completed and validated in 66.50 seconds?

http://www.gpugrid.net/result.php?resultid=6893514

Operator, Uncheck the Project Preferences option to run other WU's, if no work is available for your chosen apps,

So I did that leaving just the test apps box ticked and when I update the prefs it goes back and selects all apps.

So I unchecked everything again and selected only the beta apps tick box and made sure the accept other WUs was unchecked.

Now I'm getting that there are no WUs available.

This is all using the prefs for "Home" not the default ones. The box with the Titans in it is set to use the Home prefs.

Sorry if I seem obtuse.

Operator

---

If you want something else, thank you to answer me.

I don't have any answers for you, but I would like to say I'm amazed with you're computers. When they get the bugs worked out and get the Titans working, you'll become the number 1 cruncher, I can't believe you have 8 of them (Titans).

I suggest that you check his computer link again. It's an 8-core FX-8350 with one, and only one, Titan.

---

I suggest you reread the thread, he detached and reattached 8 times, he needed to merge all 8 computers in to 1 which he hadn't done at the time I looked at his profile.

And about the Titan/780 crunch, any news?

I'm standing by to test the beta app, whenever new workunits become available. In the meantime my Titan will continue to crunch Einstein (5 concurrent workunits in sp mode *grin*) or Milkyway (6 concurrent workunits in dp mode *grin*).

In that context: Would be cool if most colleagues here without a GTX Titan could deselect beta apps in their account to leave a few workunits for those of us who have a Titan. ;) ... provided that the project staff has no other suggestions.

More beta workunits available.

gdf

More beta workunits available.

gdf

Thank you sir.

I tried 8 of them and all errored out in about 3 secs. I am using the latest driver (320.18).

I'm assuming you get the stderr file back when this happens?

Operator

---

I tried 8 of them and all errored out in about 3 secs. I am using the latest driver (320.18).

'exit code -1073741819 (0xc0000005)'

At least one of them was successfully completed by a GTX 660 Ti, so it isn't a configuration issue this time.

Two valid results with a GTX Titan!

I got two beta workunits today and both results were good.

32-GIANNI_TESTX8-0-1-RND5844_1
31-GIANNI_TESTX8-0-1-RND6917_1

Both ran a bit under a minute, here's a stderr output of one:

<core_client_version>7.0.64</core_client_version>
<![CDATA[
<stderr_txt>
MDIO: cannot open file "restart.coor"
# Time per step (avg over 25000 steps): 2.097 ms
# Approximate elapsed time for entire WU: 52.432 s
14:13:52 (2640): called boinc_finish

</stderr_txt>
]]>


Anybody else got valid results with a GTX Titan so far?

I tried again with my Titan, the units are always in error.

@+
*_*

---

Hello Zarck,

it seems you only got non-beta workunits for your Titan today. Make sure you set your account as follows:

use Nvidia GPU -> yes
run test jobs -> yes

Run only the selected applications
ACEMD short runs for CUDA 4.2: -> no
ACEMD short runs for CUDA 3.1: -> no
ACEMD beta: -> yes
ACEMD long runs for CUDA 4.2: -> no

If no work for selected applications is available, accept work from other applications? -> no
Use Graphics Processing Unit (GPU) if available -> yes
Use Central Processing Unit (CPU) -> yes

Btw, you can use one of the venues (home, work, school) to make sure your hosts w/o a Titan are not affected by above settings.

In the meantime I got three more beta workunits which also validated ok. Seems these are all resends, send out first on the 27th of May.

Regards

Thank you Beorn

I checked my parameters, they seem correct. But I do not get units.

@ +
*_*

---

I received another beta on my 660TI. It errored out in 1 second. This was a different version than the first one that completed successfully.
http://www.gpugrid.net/result.php?resultid=6919729

(unknown error) - exit code -1073741511 (0xc0000139)

This workunit failed on another 660 Ti and finished ok on a 550 Ti (Fermi). A Kepler problem? And it was also a resend from the 27th.

Btw, any plans to distribute more betas? Currently my Titan is offline, I'm plan to move it to another hosts on the weekend.

My 780 is currently at folding@home. Waiting for more betas from here.