Audio time-compression (Was: On 80 columns should (not) be enough for everyone)

Walter Bright newshound2 at digitalmars.com
Mon Jan 31 16:59:38 PST 2011


Ulrik Mikaelsson wrote:
> I might be wrong, I hardly ever get to touch things that low-level,
> unfortunately.
> 
> However, I think the DSP:s included in most STB:s are severely limited
> in how much you can twist them. AFAIU, that's one quiet important
> aspect of the whole HTML5 video-codec debacle. I.E. in the boxes I
> work with, the entire video-processing-chain is done in one pipe-line,
> separated from the CPU almost directly at the network-interface, and
> color-keyed back with other graphics in the very last rendering-step.
> (Much as the old Voodoo2-cards worked.)
> 
>>From the economics stand-point, there's a HUGE cost-focus on them, so
> every chance of removing costs from hardware is taken.  Every dollar
> saved on hardware is roughly a dollar on the bottom-line, and in the
> volumes STB:s sell, that's millions of dollars. Sure, software costs
> might go up, but not likely by millons of dollars yearly. I.E. even if
> you CAN use the DSP-components to post-process audio, I would not be
> surprised to learn it hasn't got the power to do it simultaneously
> with video-playback.
> 
> Again, I don't work low-level enough to tell for sure, and TiVo might
> spend a lot more on the hardware than I'm used to, but hardware
> limitations in this industry is a very real challenge, especially for
> things like post-processing. I too would love 2X playback with working
> audio though. :)


The real money for Tivo is not the box cost, but the recurring subscription 
revenue. Even better if this idea is nontrivial to implement, as that would give 
the first mover an exclusive.


More information about the Digitalmars-d mailing list