mfakto: an OpenCL program for Mersenne prefactoring

[Bdot]

Originally Posted by kracker

A bit different here.
Running two instances on integrated (6550D) to max it.

Very well, so 42 it is (I always knew that answer!). Higher SievePrimes may even bring it to 50 GHz-days/day.

This shows that James needs more data to update his table. I also have the feeling that the "older generation" AMD's should be ~1/3 higher than what is in the list.

[kracker]

Originally Posted by Bdot

Very well, so 42 it is (I always knew that answer!). Higher SievePrimes may even bring it to 50 GHz-days/day.

This shows that James needs more data to update his table. I also have the feeling that the "older generation" AMD's should be ~1/3 higher than what is in the list.

Higher: about how much higher?

[Bdot]

Originally Posted by sdbardwick

Progress...I think.
Hangs on selftest

C:\hd>mfakto.hd4000 -d 11
mfakto 0.12-Win-HD4000 (64bit build)

...

Select device - Get device info - Compiling kernels ..........
WARNING: Unknown GPU name, assuming VLIW5 type. Please post the device name "Intel(R) HD Graphics 4000 (Intel(R) Corporation)" to http://www.mersenneforum.org/showthread.php?t=15646 to have it added to mfakto. Set GPUType in mfakto.ini to select a GPU type yourself and avoid this warning.

OpenCL device info
  name                      Intel(R) HD Graphics 4000 (Intel(R) Corporation)
  device (driver) version   OpenCL 1.1  (9.17.10.2932)
  maximum threads per block 512
  maximum threads per grid  134217728
  number of multiprocessors 16 (1280 compute elements)
  clock rate                350MHz

Automatic parameters
  threads per grid          1048576
  optimizing kernels for    VLIW5

########## testcase 1/1559 ##########
Starting trial factoring M50804297 from 2^67 to 2^68 (0.59GHz-days)
  k_min = 1599999998520 - k_max = 1900000000000
Using GPU kernel "barrett15_75"
  done |    ETA |     GHz |time/class|    #FCs | avg. rate | SieveP. |CPU idle
1048576 FCs copied in 0.37 ms (11287.15 MB/s), proc'd in 167.74 ms (6.25 M/s)
Error -5: Copying h_ktab(clEnqueueWriteBuffer)
ERROR from tf_class.
Error exit as selftest failed

Oh, boy. Results, after all.

• 6.25 M/s translates to ~5-6 GHz-days/day.
• this kernel certainly is not optimal, there may be better ones
• Error -5 is CL_OUT_OF_RESOURCES, we may have less memory available, or fewer streams. This already happens when copying the second memory block to the device.
• no factor found still needs to be analyzed

Next round: In order to make it a little easier for the iGPU, please create (or copy) an mfakto.ini file, with these settings changed from the default:
NumStreams=1
GridSize=1
VectorSize=1

and retry the 'mfakto.hd4000-pi -d 11 -st'.
They all have a serious effect on resources (and performance).

[Bdot]

Originally Posted by kracker

Higher: about how much higher?

Give it a try My guess would be SievePrimes=60000 should hit the 50 mark.

[sdbardwick]

Bdot, I'm away from that computer for the next 14 hours, so get some sleep!

[Bdot]

Originally Posted by sdbardwick

Bdot, I'm away from that computer for the next 14 hours, so get some sleep!

Agreed.

[kladner]

Originally Posted by kracker

A bit different here.
Running two instances on integrated (6550D) to max it.

Congratz! It's been a long time coming.....

[Bdot]

Originally Posted by kladner

Congratz! It's been a long time coming.....

Ummm ... that 6550D is the AMD integrated GPU, which has been working for a long time. Not sure if you thought that was a screenshot of the Intel HD4000 running - we're not yet there.

[kracker]

Success.

[Bdot]

Originally Posted by kracker

Success.

an improvement of 20%! Not that easy to get these days!

[kracker]

My 7770 is running at 200M/s, 150 GHZ/days, at SievePrime 15000, do you think I can get more if I increase it?

(it's OC'ed)