[opendtv] Re: RGB mania

  • From: jeroen.stessen@xxxxxxxxxxx
  • To: opendtv@xxxxxxxxxxxxx
  • Date: Fri, 14 Jan 2005 15:06:04 +0100

Hello, 

Alan Roberts wrote: 
> Not quite right Jeroen.  ITU.601 etc all include specifications for the
> coding of RGB as well as YUV signals. They form a closed set. 

Okay... I was under the false impression that 601 was invented 
at a time when all RGB->YUV and YUV->RGB conversions were still 
done by analog matrix circuits, and that storage and signal 
processing were to be done only on the (cheaper) YUV 4:2:2 
signals. Then it would not be necessary to specify anything for 
digital RGB signals, if they did not exist at the time... 

> At no point is there any flexibility in the coding of Y, since
> it normally doesn't exist in analogue these days.

Huh ? So the situation has been reversed, all matrices are now 
digital... Then I hope that the signals are 10 bits or better ! 
What helps is if the gain RGB->Y and Y->RGB (for monochrome 
signals) is always exactly 1.00, to avoid quantisation of grey. 


Prin wrote (after some decyphering): 
> In an 8 bit ITU-R BT.601 system the code for 'white' is 235 [decimal] -
> all seems clear and simple.
> (etc.) 

We were mostly discussing what happens with RGB below black. 

Of course a similar discussion can be held about above white. 
This is both simpler (because peak white usually occurs in smaller 
areas than peak black) and more complicated (because there are 
many more solutions possible, and we have to make choices). 

> if the broadcaster did not apply the soft clipper - the display 
> would probably hard clip - which would be even worse.

Fear not. The situation will occur anyway because we allow our 
viewers to turn up the (contrast) gain beyond 1.0x. This is for 
viewing dark images in bright rooms. So we must have a cure for 
white clipping anyway. In the old days this was called "peak 
drive limiting", now we prefer to call it "contrast reserve". 
See patent WO2002071745(A1). Soft clipping is optional, see 
patents EP0589513(B1) and WO2002085037(A1). I am now extended 
this principle in new directions for exotic display types. 

You never followed up on that idea for digital cinema that I 
launched in the campus restaurant, on the second day, of doing 
something special with the amplitude... 

Best regards, 
-- Jeroen 

+-------------------------------+----------------------------------------+
| From:     Jeroen H. Stessen   | E-mail:  Jeroen.Stessen@xxxxxxxxxxx    |
| Building: SFJ-5.22 Eindhoven  | Philips Digital Systems Laboratories   |
| Phone:    ++31.40.2732739     | Visiting & mail address: Glaslaan 2    |
| Mobile:   ++31.6.44680021     | NL 5616 LW Eindhoven, the Netherlands  |
| Pager:    ++31.6.65133818     | Website: http://www.pdsl.philips.com/  |
+-------------------------------+----------------------------------------+

 
 
----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:

- Using the UNSUBSCRIBE command in your user configuration settings at 
FreeLists.org 

- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word 
unsubscribe in the subject line.

Other related posts: