Attila Kinali wrote:
On Wed, 23 Jan 2008 09:37:56 -0700 James Richard Tyrer
<[EMAIL PROTECTED]> wrote:
Sad that a major brand like Sanyo would get the decode wrong.
It seems to be a general problem caused by the chips used for
decode. I noticed this in reading reviews for DVD players.
The problem is that it is hard to get full quality decoders in
realtime with limited power, area and money budget. Thus engineers
tend to comprimise at some points, trying to simplify the problem by
allowing the system to act inacurate. How bad these approximations
are is most time not clear until customers complain.
This would appear to be a motion compensation issue. IIUC, TI went
to hardware motion compensation for the TMS320DM6467.
Where do you see hardware motion compensation?
http://focus.ti.com/lit/ml/sprn249/sprn249.pdf
The TMS320DM6467 looks to me just like a VLIW DSP combined with an
ARM chip and a lot of peripherial stuff. No fancy decoder unit.
Read more! It also includes
Dual Programmable High-Definition Video Image Co-Processor
(HDVICP) Engines
Video Data Conversion Engine (VDCE)
Horizontal and Vertical Downscaling
Chroma Conversion (4:2:2↔4:2:0)
BTW: What is a hardware motion compensation for you? I ask because
the process of motion compensation is something that looks always
very similar, no matter how much you do in hardware.
Hardware motion compensation would be motion compensation done (or
partially done) by hardware dedicated to that use only.
--
JRT
_______________________________________________
Open-hardware mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-hardware
