Today we released a new CUDA102 application v102.14 for both Windows & Linux.
A bug was fixed where application causes 100% utilisation of whole CPU core (thread in hyperthreaded CPUs)

Radim Vančo (FoxKyong)

I've tried it out and can confirm that the CPU usage is significantly lower.

Here are some results from my machine. GTX 1070 and GTX 1050Ti on Ryzen 3700X.

I'm not sure how to tell if my installation is using (or indeed, can use) a "CUDA" version.
It's just running on a laptop at the moment, no idea if it uses any fancy graphics card of any sort, CUDA or not (if that is a graphics card type?).

Is there a download link, or does the BOINC manager download the updates automatically?

---

CUDA is for Nvidia GPUs.
I would think it'd automatically d/l the app if you have the right GPU.

You could use GPU-Z to see which GPU you have.

---

Team AnandTech - SETI@H, Muon1 DPAD, Folding@H, MilkyWay@H, Asteroids@H, LHC@H, POGS, Rosetta@H, Einstein@H,DHPE & CPDN

Main rig - Ryzen 3600, 32GB DDR4 3200, RX 580 8GB, Win10
2nd rig - i7 4930k @4.1 GHz, 16GB DDR3 1866, HD 7870 XT 3GB(DS), Win7

PRO RADIMA: Jedu to na 14 vláknech. Verzi GPU se nevyplácí provozovat, nebo mám nějakou chybu v konfiguraci? ale, pojedu krom turnajů, pokud nebudou kovy kompletní :D

---

Možná budete muset jít do svých počítačových preferencí v BOINC a povolit mu používat GPU.

Možná budete muset jít do svých počítačových preferencí v BOINC a povolit mu používat GPU.

Hello,

It takes me twice as long to process the work units as with Cuda55!

Regards Klaus

http://asteroidsathome.net/boinc/results.php?userid=57813

Already received CUDA102 application v102.15 tasks.
Some show weird behavior, task on time 0:00:02 starts from begin again and again.

Hi folks

Relatively new to Asteroids@home since the "passing" of SETI (wipes away a tear) but cannot see why my work done is still at zero. The (three) other projects I'm working on are showing as active but nothing on Asteroids.....

It's not showing as suspended or anything else obvious, but if anyone would be so kind as to remind me what other basics I should be checking, it would be greatly appreciated.

Many thank yous.

Nick

Work units are sent to two computers, which must both send back the result. Then the results are compared and validated.

You have sent back 1 result that could be validated. Therefore you have 480 credits.

Then you have sent back another 5 results, which wait for the second cruncher to send in the result. When they turn up, your work will be validated. They have up to 10.5 days to do so.

You have done nothing wrong and there is nothing that you must or can do. You just need to wait. Everyone has at any time pending units. I have currently 261 units waiting to be validated. No problem.

Thank you Robert. :)

Much appreciated.

You're welcome! Oh, I see that a second work unit has been validated. No instant gratification here, but what are a few hours or days in the life of an asteroid?

Hahaha - absolutely.

I just wondered if I had missed something as the other tasks I had joined at the same time were ticking along nicely. :)

Many thanks again.

I have over 32 WU waiting to report for two days which started a servers out of space. Any ETA on restoring?

---

I have seen multiple comments regarding a doubling of run time on the new work units compared to the CUDA55 units. I am seeing the same thing, though the variation seems to be from double to triple the run time.
This isn't encouraging and suggests the new app is poorly optimised.
If this matter could be passed up to the devs for revision that would be much appreciated by the volunteers.

why does the server not upload?

why does the server not upload?

Because it's out of disc space, someone has to delete clutter.

---

Grüße vom Sänger