Pages: 1 [2]   Go Down

Author Topic: Time for a new JPEG alterntive?  (Read 4056 times)

hjulenissen

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2051
Re: Time for a new JPEG alternative? How about JPEG 9.1!
« Reply #20 on: February 24, 2015, 04:12:31 am »

In the case of jpeg and jpeg 2000, for example, quite a lot of code would be common anyway.  
I have some practical knowledge of JPEG and some superficial knowledge of JPEG2k. This leads me to believe that the common code between the two would be very little. Perhaps some setup/teardown, scaling/colorspace conversion could be shared, but the core codecs seems to be totally different?

-h
Logged

hjulenissen

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2051
Re: Time for a new JPEG alterntive?
« Reply #21 on: February 24, 2015, 04:17:30 am »

Not seeing any issues visual issues with JPEG an NEC PA272. Even one saved in ProPhoto RGB!
Some photographers (myself included) use their monitor as a softproofing device. If you limit your monitor to 100:1 or 200:1 of CR, then >8bits may be irrelevant.

But what if your display (and content and viewing conditions) allowed for 1000:1 or 10000:1 CR (possibly doing some tricks like dynamic backlighting). At some point, gradations would be a)visible in hand-picked corner cases, b)visible in a large number of cases. My only reference is Poynton and 9 bits for 50:1 television quality, and the developer of a well-known open-source color calibration package saying something to the effect of "10 bits aint worth it".

-h
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Re: Time for a new JPEG alternative? How about JPEG 9.1!
« Reply #22 on: February 24, 2015, 12:11:49 pm »

JPEG 9.1 is backwards compatible (9.1 decoders can read earlier spec jpegs) but not forward compatible (earlier decoders can't read 12-bit jpegs).
Agreed that JPEG 9.1 might not have the best possible compatibility with older JPEG; indeed I think it quite likely that the current JPEG will persist for a very long time, due to being an incumbent that is "good enough" for almost all users.  One further point: oversampling and then down-sampling on display (like 24MP JPEG displayed on even 4K displays) also improves perceived DR over the "per pixel DR", so that is one way that even the current 8-bit JPEG can keep up with viewing needs, short of 100% pixel-peeping.

My point was only about the relative prospects of adoption for the various proposed replacement for current JPEG; the combination of some compatibility (including the potential for a phase-in of hardware support in newer devises) and the support of the the main standards body give JPEG 9.1 a lot more chance than the others.
Logged
Pages: 1 [2]   Go Up