Path: chuka.playstation.co.uk!scea!greg_labrec@interactive.sony.com
From: dblee <dblee@netmagic.net>
Newsgroups: scee.yaroze.programming.2d_graphics
Subject: Re: 16 bit question
Date: Mon, 09 Feb 1998 03:16:32 -0800
Organization: SCEA News Server
Lines: 30
Message-ID: <34DEE590.3ECB@netmagic.net>
References: <34C74080.6993D993@infores.com>
Reply-To: dblee@netmagic.net
NNTP-Posting-Host: ppp1-16.sj.netmagic.net
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 3.01 (Win95; I)

Luca Cappellini wrote:
> I understand that when i use 16 direct mode for sprite colors i do not
> have a CLUT table but in wich way i define the colors, with PaintShop
> pro witch kind of color table i have to use if i wanna define my colors,
> and if i can define my colors is right think that in 16 bit mode i can
> only have one BIG set of colors for all my sprites ?

Hello,

As Lewis pointed out, only 4-bit and 8-bit sprites use CLUTs.  If you
are creating 16-color or 256-color TIMs from bitmaps, each has its
own CLUT, kinda like a GIF.  Each sprite could reference another CLUT
in another sprite, or a CLUT you define yourself.  Each pixel is mapped
to a CLUT entry.

On the other hand, a 16-bit sprite is made up of individual RGB
values (5-bits per channel plus one extra bit as a flag) for a
grand total of 32K colours.  That's why it doesn't need a CLUT.
(The same applies to the 24-bit RGB TIM mode, too.)

I use PhotoShop and usually work with 24-bit RGB color.  I just save
the image as a 24-bit BMP and let timutil.exe convert that into 32K
colours.  So far, the results have been pretty good (8-bits per channel
reduced down to the 5-bits).  If you want to squeeze things down
further, jump from RGB to Indexed Color Mode.  Dithering helps a
little.

- e!
  tenchi@netmagic.net
  http://www.netmagic.net/~tenchi/yaroze/