Copy link to clipboard
Copied
Hi, everyone!.
I hope you are doing well.
I am creating gpu accelerated plugin right now.
I started to develop it from SDK_CrossDissolve and SDK_ProcAmp.
But there is no way to create new buffer on sample project.
So, I used following mode of metal, opencl and cuda for that.
But I am getting crash issue after 3 ~ 4 hours when I use them.
For metal:
id<MTLBuffer> tempEffectFrameData = [device newBufferWithLength : sizeof(float) * 4 * entireWidth * entireHeight options : MTLResourceStorageModeManaged];
for opencl and cuda:
int tempBufferSize = entireWidth * entireHeight * sizeof(float) * 4;
mGPUDeviceSuite->AllocateDeviceMemory(
index,
tempBufferSize,
&tempDeviceBuffer
);
Do you have another way for that?
Regards,
Igor.
Have something to add?