Fix function encoder capability mismatched with driver's table.

[xuguangxin]

https://github.com/intel/media-delivery/blob/master/doc/benchmarks/intel-data-center-gpu-flex-series/intel-data-center-gpu-flex-series.rst#fixed-function-encode-vdenc

see https://github.com/intel/media-driver/blob/master/docs/media_features.md#hardware-encoding-low-power-encodingvdenchuc
for hevc 8 bits.

Instead of maintaining a local copy of capability. could we point to driver capability?

thank you

[dvrogozh]

Can you, please, be specific which mismatch you are talking about? resolution?

Instead of maintaining a local copy of capability. could we point to driver capability?

No. This table in media-delivery is part of collateral material and just highlights certain platform features in one place. It does not pretend to be full list of capabilities. For the full list other documentation should be looked at (as you pointed in the driver). We can however try to add links for this kind of documentation.

[xuguangxin]

@dvrogozh thank you for the reply.
Yes. the driver can support the max resolution for hevc 8 bits is 16K, but the media delivery said the max is 8K,
It will mislead customers.

[xuguangxin]

No. This table in media-delivery is part of collateral material and just highlights certain platform features in one place. It does not pretend to be full list of capabilities. For the full list other documentation should be looked at (as you pointed in the driver). We can however try to add links for this kind of documentation.

Not a full list is understandable. But, we'd better have no conflict information like 8k vs 16k.
thank you