Version 390.77 - WHQL
Release Date Mon Jan 29, 2018
Operating System Windows 10 64-bit
Language English (US)
File Size 431.94 MB

and the first guinea pig is going to be....? ;-)

---

Oink :))

So far so good.
I hesitate to say this but it seems to me my GPU tasks are taking about 30 seconds less to complete with version. Too small a sample to be sure.

---

Just for giggles, since I am in the middle of my latest Linux build right now, I went to Nvidia's site and took a gander for the Linux version, and noticed that the latest and greatest is 390.25. I presume because its Linux, it is always going to be a number of versions behind the current Windows build?

---

Seems to be ok atm, I have noticed also that wu's are completing slightly faster..

So far so good.
I hesitate to say this but it seems to me my GPU tasks are taking about 30 seconds less to complete with version. Too small a sample to be sure.

Looks like there were changes to the OpenCL portion of the driver, so it's possible.

From the Release Notes

 Device side enqueue
•The current implementation is limited to 64-bit platforms only.
•OpenCL 2.0 allows kernels to be enqueued with global_work_size larger than the
compute capability of the NVIDIA GPU. The current implementation supports only
combinations of global_work_size and local_work_size that are within the compute
capability of the NVIDIA GPU.
The maximum supported CUDA grid and block size of NVIDIA GPUs is available
at http://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#compute-
capabilities. For a given grid dimension, the global_work_size can be determined by
CUDA grid size x CUDA block size.
•For executing kernels (whether from the host or the device), OpenCL 2.0 supports
non-uniform ND-ranges where global_work_size does not need to be divisible by
the local_work_size. This capability is not yet supported in the NVIDIA driver, and
therefore not supported for device side kernel enqueues.

---

Grant
Darwin NT

I thought we weren't used OpenCl 2.0, still with the OpenCl 1.2....Raistmer would know for sure

---

Yes, still using OpenCL 1.2 for the SoG app. Raistmer will have to rewrite that app to use the new OpenCL 2.0 capabilities of the driver.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

During the current NSDO (that's Non-Scheduled Data Outage) I d/l and installed 390.77.
Hopefully, it will like my two dualies...

---

It certainly isn't helping this computer;
State: All (1090) · In progress (199) · Validation pending (511) · Validation inconclusive (225) · Valid (70) · Invalid (84) · Error (1)
Unfortunately the Server has decided to send me a good number of his failures...there ought to be a law...

So, this new driver isn't going to be any better for the current SOG app?
Or has anybody tested it and found it beneficial?

Meow?

---

"Time is simply the mechanism that keeps everything from happening all at once."

So, this new driver isn't going to be any better for the current SOG app?
Or has anybody tested it and found it beneficial?

Meow?

As I posted. On my Intel GTX 960 Ti, Windows 7. About 20 seconds faster with the SOG app.

So far, nothing negative.

Thanks. I'll give it a spin then.
Maybe I can speed the rig up just a tad and drive it's RAC a little lower......LOL.
Meow!

---

"Time is simply the mechanism that keeps everything from happening all at once."

So, this new driver isn't going to be any better for the current SOG app?
Or has anybody tested it and found it beneficial?

Meow?

I haven't seen any positive or negative changes. Seems safe.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)