Tool for checking if plots are optimized
-
Is there a tool to check if plots are optimized,
I did plotting with GPU plotter, but read speeds are slow. Am mining with CPU,
Thanks in advance
-
You can check it by looking at the file name. If the amount of nonces is equal to the stagger size it is optimized. If the amount of nonces aren't equal it isn't optimized.
14750531867371072502_0_1024_64 = Not optimized
14750531867371072502_0_1024_1024 = Optimized.
-
My file name reads as below
XXXXXXXXXXXXXXX_300000001_335360_8384
As per your comments its not optimized
I got it now, i specified the wrong stagger size when using GPU plotter. I should have formed the filename as
XXXXXXXXXXXXXXX_300000001_335360_335360
Did I got it right
-
Kinda... It is better to plot the plot non-optimized and later you can optimize it. The stagger size is the memory you use to plot the file, and a too big stagger size might not work with your graphic card.
Edit: Use this if you want to optimize it
-
@tpriyan - as frilledshark points out, you can't set an arbitrary stagger size, it needs to be within your systems capacity. However if you do "gpuplotgenerator generate direct" rather "gpuplotgenerator generate buffer", the plotter will create an optimized plot - the process however is slower than using buffer.
-
@haitch How could I optimize a plot if I do not have enough space on my mining hard drive or computer hard drive to do it?
-
@mathew If you don't have the space available, you can't optimize it. Best option is to use the gpu plotter in direct mode so that the plot is created optimized.
-
This post is deleted!
-
under the devices.txt what do I put if I have a nividia geforce 750TI card?
-
@mathew Take a look here for the plotter : https://forums.burst-team.us/topic/58/gpu-plot-generator-v4-0-3-win-linux/2
To get the correct device number run "gpuplotgenerator setup", then select option 1 to list devices in your system You'll get something like:
[0] AMD Accelerated Parallel Processing (OpenCL 2.0 AMD-APP (1912.5))
[0] Tahiti (OpenCL 1.2 AMD-APP (1912.5))
[1] Tahiti (OpenCL 1.2 AMD-APP (1912.5))
[2] Tahiti (OpenCL 1.2 AMD-APP (1912.5))
[3] AMD FX(tm)-4130 Quad-Core Processor (OpenCL 1.2 AMD-APP (1912.5))The "[0] AMD Accelerated Parallel Processing (OpenCL 2.0 AMD-APP (1912.5))" is your platformId - in this case 0. The devices below are the devices on that platform. Pick the number that corresponds to your video card and that's the DeviceID.
-
@haitch it is stating that I am running out of resources while in step 2 kernel launch. Not sure how to fix it.
-
@mathew In devices.txt change "0 0 4096 256 8192" to "0 0 2048 256 8192" and try again.
-
@haitch said:
now unable to create opencl gpu buffer
-
@mathew okay, put the 2048 back to 4096, and try reducing the last number of your plot name from 10240 down to 4096.
-
@haitch Getting the same error.
-
@mathew What the output of gpuplotgenerator setup, option 1 and option 2 ?
-
@haitch not sure what you mean. If this is what I think you mean.
-
@Mathew : From the above screenshots you are using your intel HD graphics for generating the plot, which does not have sufficient memory
I guess you need to change the line in devices.txt to
1 0 4096 256 8192
If the above does not work, you can try the below
1 0 2048 256 8192
Run plot generator
If this doesn't work too reduce the 2048 further to 1024 and then to 512 and try again
If nothing works paste the output of below command here
gpuPlotGenerator listPlatforms
-
@mathew - yep, exactly what tpriyan said - the devices.txt needs to be updated to reflect you actual card.
-
@tpriyan @haitch Tried all of that and my driver then says it has stopped responding buy has recovered.
