[theora] NHW Project - quality improvement - proper subband quantization

[Raphael Canut]

Hello,

I have improved quality for -l1->-l4 settings.I have properly set up the
wavelet coeff threshold of the dead-zone according to the different wavelet
subbands, I have then saved another 1Ko per .nhw files without noticeable
difference and so I have raised again the quantization.The reconstructed
images have now a better precision and still a good neatness.

Update at: http://nhwcodec.blogspot.com/

Else I realize that JPEG XR (2009) and WebP (2011) are quite good codecs,
they are backed furthermore by big companies Microsoft and Google, they
claim and guarantee that their codecs are royalty-free, and so if at the
moment of their release in 2009 and 2011, they didn't succeed in replacing
JPEG, I can not see why today the NHW codec that is globally on par with
them will succeed in doing so.Could be even seen as unfair...

So it is nearly impossible for the NHW Project to penetrate the market,
except maybe for devices that have a microcontroller, low power processor
<400Mhz, and where speed is a critical and imperative need, because the NHW
codec is very faster than these codecs.I will maybe look on this side in
the next days...

If you have time, let me know about the quality improvement!

Cheers,
Raphael
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.xiph.org/pipermail/theora/attachments/20180317/2e347f00/attachment.html>