NetNews Offline 2/NetNews Offline Volume 2.iso/news/comp/sys/amiga/programmer/5308

Internet Message Format | 1996-08-05 | 3.8 KB

Path: informatik.tu-muenchen.de!fischerj
From: [e-mail] (Juergen "Rally" Fischer)
Newsgroups: comp.sys.amiga.programmer
Subject: Re: doubling pixels horizontally
Date: 12 Mar 1996 18:38:38 GMT
Organization: Technische Universitaet Muenchen, Germany
Distribution: world
Message-ID: <4i4gbe$[e-mail]>
References: <4f4ibc$[e-mail]> <[e-mail]><[e-mail]><4faoe1$[e-mail]><[e-mail]><[e-mail]><[e-mail]>   <[e-mail]> <4gbjg3$[e-mail]>  <[e-mail]> <4h4hv5$[e-mail]> <[e-mail]>
NNTP-Posting-Host: hphalle10.informatik.tu-muenchen.de
Originator: [e-mail]


In article <[e-mail]>, [e-mail] (Ludvig Pedersen) writes:

|> >so anyone of us got a bug or 030 is very different here ?
|>
|> Its not a bug, bustest also confirmed that. Reading from fastram, move.l is
|> faster than movem. But on writing movem is faster! :-)

well, it rather looks like movem read has become slower vs. 020 :\

|>
|> >Do those "free c2p" routines really run 5.6 mb/sec ?
|>
|> Hehe, no, but you usually don't play Doom without any bitmap-DMA, do you! ;)

grrrrrr ;) But it's a useful information for a coder what speed a routine
can go with dma off.

|> If you turn on 256 colors you get a slow-down.

uh, I didn't know that. :D

|> >I said "very fast cpu" and "aproxiamte 7mb/sec" :)
|>
|> You should have said something like a fast cpu with 1+ MB internal cache! ;)
|> But that would be cheating! :-)

or maybe a cool PPC board...

|> What is LAME? >:( That you have problems with you math again? :-)

probs _again_ ? what ?

anyway I parsed your sentense wrong, I thought you get 203300 dbra's per
second. more a language prob than a math prob ;)

|> >what ?
|>
|> What you forget is that a move uses 2 clock cyles.

but if I do dbra I won't look at the move cycle-table ;))

|>
|> ONE clock cycles is 20 ns. So TWO clock cycles must be 40 ns.
|>
|> How *hard* can it really be??? This is pretty elementary math! ;-)

hey wait a minute, I'm master of theoretic cycle calculation if
given proper data, I already released 020-14 fastmem timings
before having fastmem in my A1200. baeh! :)

we talk about mem moves. i.e

>> move.l (an)+,(an)+ << and nothing more.

where is the move reg,reg you (and writer of bustest) talk about ? ;)

|> >yep. no need for ghost-look doing doom, descent etc.
|> >just intuiscreen, supporting all monitors, or genlock or whatever :)
|>
|> That would be great! :)

yeah! :) imagine your routine supports rendering into a intuition window
and you got a genlock.

Then you could watch TV action and simultaneously play in a
small 160x100 window having your interactive action :)

hmm genlock supporting doom engine should maybe not use color 0 for rendering.
and genlock supporting demo should always use color 0 for background :)
mixing amigademo with mtv :D
well, genlocks are not ghostlook-compatible I guess, demo must be
switched to 4pass...

|> >i.e 1084 -> most users no problemo. ghost-look looks even better
|> >over composite video vs. rgb-port, because of smear (hardware-dither ;)
|>
|> Dithering looks ok on a multisync too. (Atleast in PAL) But you don't get any
|> smear! ;) "Lace"ing the dither is the way to go in both cases.

yep, no smear as their bandwidth is able to display white-black-white...

|>
|>
|>
|> <sb>Ludde - Amiga Demo Coder
|> <sb>Virtual Reality & Official Be developer
|> <sb>[e-mail]
|>
------------------------------------------------------------------------
   [e-mail] (Juergen "Rally" Fischer)   =:)