Tony (Atanas) Zlatinski

Aha, okay. Thank you, [charlie-ht](https://github.com/charlie-ht) and [airlied](https://github.com/airlied). Indeed, we'll need a fix for that in the sample app. I'll look into fixing this soon.

Thank you, Mark! We agree with your findings and will be working on fixing this in the CTS and the drivers before fixing the sample app.

We already support P and B frames with all encoders. Closing this issue.

Yes, we need a separate transfer queue for both AMD and Intel implementations. Can you please try the ToT. It should be fixed now.

Thank you, Stephane! Can you please tell us if this is happening during the pre-processing stage or the encode stage? An applicaiton call- stack would be very helpful.

Thank you, for the backtrace, Stephane! It looks like the open-source driver does not support creating individual views against YCbCr image. We'll have to deal with this by using properties,...

Thank you, [lolzballs](https://github.com/lolzballs)! > I was able to work around this by setting VkImageCreateInfo::flags = VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT Yes, indeed, this is the correct use when the individual planes of YCbCr MP...

Hi Dave, Please also check the m_samplerYcbcrModelConversion and m_samplerYcbcrRange in VulkanFrame::DrawFrame(). Currently, the app has it fixed to BT709 NARROW range, but this should reflect the settings from the compressed...

Thank you for your feedback and suggestions for app improvements, Kurt! Would you mind submitting a pull request with all of the changes you'd like to see in the sample?

Hi Kurt, what is the MESA driver that you are using - what version/distro and HW (Intel, AMD)? About 4. > The next decode step will then assert that the...