home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!mcsun!Germany.EU.net!mpifr-bonn.mpg.de!uniol!caty!cbmger!peterk
- From: peterk@cbmger.de.so.commodore.com (Dr Peter Kittel Germany)
- Newsgroups: comp.sys.amiga.programmer
- Subject: Re: Chunky Pixels vs. Bitplanes (was: Chunky Chip Set...)
- Message-ID: <10408@cbmger.de.so.commodore.com>
- Date: 11 Jan 93 14:53:06 GMT
- References: <Karsten_Weiss.0n2o@ibase.stgt.sub.org> <1hbngoINNglt@uwm.edu> <1993Jan1.141207.20262@mpifr-bonn.mpg.de> <doiron.0kil@starpt.UUCP>
- Reply-To: peterk@cbmger.de.so.commodore.com (Dr Peter Kittel Germany)
- Organization: Commodore Germany
- Lines: 75
-
- I know I'm coming a little late in this debate, but only some few remarks:
-
- In article <doiron.0kil@starpt.UUCP> doiron@starpt.UUCP (Glenn Doiron) writes:
- >In article <1993Jan1.141207.20262@mpifr-bonn.mpg.de> mlelstv@speckled.mpifr-bonn.mpg.de (Michael van Elst) writes:
- >> In <doiron.0kd5@starpt.UUCP> doiron@starpt.UUCP (Glenn Doiron) writes:
- >> >>
- >> >> So if you have a plane oriented memory (which is no real problem with VRAMs
- >> >> although you may get some alignment restrictions)
-
- Yes, severe restrictions, but then it should be possible.
-
- >> >... which use DRAM's, not VRAMS. No amount of 'alignment restrictions' is
- >> >going to cut it for VRAM planar, unless, as I said, you do funky
- >> >address/data schemes and put one plane in each VRAM,
-
- Yes, sure. Or put it more generally: Every bitplane has to start at the same
- offset in every VRAM chip. That's an alignment of the size of one VRAM chip
- as compared to 16-bit alignment in <= ECS or greater portions in AA. But not
- at all impossible IMHO.
-
- >> >then mess further with your RAMDAC feeding setup.
-
- Why that??? Every VRAM delivers one bit for the current pixel, *exactly*
- the same situation as in a chunky environment. What additional effort
- would be necessary?
-
- >> >Using VRAMS like DRAMS will result in a loss of
- >> >all the performance gains VRAMS were supposed to give you in the first
- >> >place.
-
- Don't get that.
-
- >As I said, you lose all the performance gains VRAMS were supposed to give
- >you in the first place, since you can't use the serial registers for video
- >fetch in planar setups.
-
- Of course you can, if you obey the mentioned alignment rules. Not impossible
- at all in my eyes.
-
- >No. It is not. You are doing N time the amount of work in planar, where N
- >is the number of planes.
-
- This is overgeneralization. Whenever you have data that also have some
- width > 1 pixel, then you can cut down the number of fetches and stores
- with appropriate (I'm not saying "tricky" or "genius-like") coding.
-
- >I'm not going to argue with you about depth scaling. Given the same depth,
- >chunky will be faster for all of the above operations, with exception of a
- >few break-even cases. Sometimes it will even be faster on chunky *even
- >with a reduced scale for planar*, since chunky doesn't need 'bit
- >thrashing'.
-
- All this becomes immediately invalid and turns into a pro argument for
- planes when your bits per pixel are not multitudes of 8!
-
- >:) Look, if you want a 128-color display, I won't forbid you from using a
- >planar architecture. However, drawing a line will be far faster on a
- >256-color chunky than a 128-color planar display.
-
- Yes, but 8 bits per pixel is not the be-all-end-all of all worlds! There
- are many situations where you can very nicely do with fewer bits, say
- 5 or 6, and then you are going to waste memory *and* processor cycles
- with a chunky architecture. I think it's *not* realistic to caculate
- always with an 8-bit case only. (I say "realistic", not "unfair", because
- we want to deal with what comes across us in our dayly needs.)
-
- Yes, for n*8 bits/pixel, chunky is in most cases the best option. (I still
- believe there also situations where planar can be better also here.)
- If you have some odd number of bits, then I guess the planar structure
- has more advantages.
-
- --
- Best regards, Dr. Peter Kittel // E-Mail to \\ Only my personal opinions...
- Commodore Frankfurt, Germany \X/ {uunet|pyramid|rutgers}!cbmvax!cbmger!peterk
- Wer's nicht kann, soll's bleiben klopfen oder Steine lassen!
-