OK - I'm getting more than a little frustrated... My queue keeps going down, and down, and down... Event Log continues to state, "Job Cache Full." This is NOT the case... I have MAYBE two more days of cache of 1.17 Units for the XP system.
The MAC, on the other hand, DOES get new work. I turn in two Units, (1.17's), and I get AT LEAST two Units Downloaded. Anyone else seeing issues with Windows machines NOT getting cache replenished??? [EDIT:] OK - Things just changed on the MAC... Now, as of this morning, I'm getting "Job Cache Full" on the MAC, too...
I'm looking into deploying Beta apps that are the same as the current 1.17 versions but have a lower minimum VRAM requirement so GPUs with more than 766 MB VRAM can be tested. Currently the limit is 1GB VRAM because a task typically uses 750 MB and we want some buffer for normal operations.
By the way, what are the minimum specifications, memory and compute capability for each sub-project ?
Thanks by advance
Since the implementation of 1.15 onward, I've noticed a HUGE drop in RAC. I've gone from a high of 103K RAC from the 1.12 and 1.14 Units to 83K RAC with 1.15 onward... Anyone else notice this significant drop???
I'm still crunching 2 Units at a time per GPU card on a total of 3 cards. Two EVGA GTX-750TI SC cards in the MAC/Hackintosh, and one EVGA GTX-760 in the Win XP Pro x64 system. [EDIT:] Also, still crunching 15 Hours a day on each system.
I will continue monitoring the situation. Also, still monitoring for Invalids on NVIDIA on MAC in El Capitan 10.11.4. (Darwin 15.4.0) Last Invalid was a 1.14 Unit; all prior 1.13 and 1.12 Invalids have dropped off the Web Results Page.
It's my Win system with the 760 that's taking the beating! Dropped nearly 30K RAC since 1.15 Units went on line. The MAC is currently holding in the 60K's... [EDIT:] Just noticed a drop on the MAC with 1.17 Units, too... Went from a solid 63K RAC down to 58K RAC right now...
Just received a second 1.14 Invalid... Still have six 1.14 Units Pending. So, for the moment, the 1.14 Units onward have a LOWER Invalid ratio than the 1.12 and 1.13 Units on Darwin 15.4.0 and above; but, the OpenCL Bug seems to be still present.
I will keep monitoring and report back. So far, no 1.17 Invalids to report.
[Update on Win XP Pro x64
)
[Update on Win XP Pro x64 with GTX-760]
OK - I'm getting more than a little frustrated... My queue keeps going down, and down, and down... Event Log continues to state, "Job Cache Full." This is NOT the case... I have MAYBE two more days of cache of 1.17 Units for the XP system.
The MAC, on the other hand, DOES get new work. I turn in two Units, (1.17's), and I get AT LEAST two Units Downloaded. Anyone else seeing issues with Windows machines NOT getting cache replenished??? [EDIT:] OK - Things just changed on the MAC... Now, as of this morning, I'm getting "Job Cache Full" on the MAC, too...
TL
TimeLord04
Have TARDIS, will travel...
Come along K-9!
Join SETI Refugees
Hi all, I have a problem
)
Hi all,
I have a problem with FGRPopencl-nvidia on a host with a GTX 260 (896 MB) and a Quadro K2200 (4 GB) using 342.01 driver. Some WU are working (https://einsteinathome.org/fr/task/598587761) and others are failing with 69 error code (https://einsteinathome.org/fr/task/598618221). Can it be due to the amount of memory on GTX 260 ?
By the way, what are the minimum specifications, memory and compute capability for each sub-project ?
Thanks by advance
_AF_EDLS_GuL wrote:Hi
)
Yes, it is probably the GTX 260. I have one as well and the scheduler doesn't even give me any tasks because:
"OpenCL GPU RAM required min: 1071644672.000000, supplied: 938803200"
Looks like minimum of 1G required.
Thanks. It seems to work when
)
Thanks. It seems to work when excluding the gtx 260 in cc_config.xml
Filipe_9 wrote:Stranger7777
)
Looks like we are off of the crunching ship now with all the 32-bit platforms around. Just saying this
Since the implementation of
)
Since the implementation of 1.15 onward, I've noticed a HUGE drop in RAC. I've gone from a high of 103K RAC from the 1.12 and 1.14 Units to 83K RAC with 1.15 onward... Anyone else notice this significant drop???
I'm still crunching 2 Units at a time per GPU card on a total of 3 cards. Two EVGA GTX-750TI SC cards in the MAC/Hackintosh, and one EVGA GTX-760 in the Win XP Pro x64 system. [EDIT:] Also, still crunching 15 Hours a day on each system.
I will continue monitoring the situation. Also, still monitoring for Invalids on NVIDIA on MAC in El Capitan 10.11.4. (Darwin 15.4.0) Last Invalid was a 1.14 Unit; all prior 1.13 and 1.12 Invalids have dropped off the Web Results Page.
TL
TimeLord04
Have TARDIS, will travel...
Come along K-9!
Join SETI Refugees
Welcome to the club.
)
Welcome to the club.
Betreger wrote:Welcome to the
)
It's my Win system with the 760 that's taking the beating! Dropped nearly 30K RAC since 1.15 Units went on line. The MAC is currently holding in the 60K's... [EDIT:] Just noticed a drop on the MAC with 1.17 Units, too... Went from a solid 63K RAC down to 58K RAC right now...
TL
TimeLord04
Have TARDIS, will travel...
Come along K-9!
Join SETI Refugees
It seems that Windoz really
)
It seems that Windoz really takes a beating on this app, Linux and Mac systems seem to fare well with it.
[Update: MAC OS with dual
)
[Update: MAC OS with dual GTX-750 cards - ]
Just received a second 1.14 Invalid... Still have six 1.14 Units Pending. So, for the moment, the 1.14 Units onward have a LOWER Invalid ratio than the 1.12 and 1.13 Units on Darwin 15.4.0 and above; but, the OpenCL Bug seems to be still present.
I will keep monitoring and report back. So far, no 1.17 Invalids to report.
TL
TimeLord04
Have TARDIS, will travel...
Come along K-9!
Join SETI Refugees