home *** CD-ROM | disk | FTP | other *** search
- Path: grafix.xs4all.nl!john.hendrikx
- Date: Sat, 02 Mar 96 22:53:16 GMT+1
- Newsgroups: comp.sys.amiga.programmer
- Distribution: world
- Subject: Re: Amiga doesn`t need Pl
- MIME-Version: 1.0
- Content-Type: text/plain; charset=iso-8859-1
- Content-Transfer-Encoding: 8bit
- From: john.hendrikx@grafix.xs4all.nl (John Hendrikx)
- Message-ID: <john.hendrikx.4ixx@grafix.xs4all.nl>
- Organization: Private
-
- In a message of 01 Mar 96 Stephan Schaem wrote to All:
-
- >> I don't think so, maybe the 50 MHz version, but they still won't get 25
- >> FPS orso at 320x256 1x1 (talking DOOM here, not some WolfenStein clone
- >> with floors which I see all too often).
-
- SS> rendering a static scene should acheive 25fps in 320x200 (60hz) with
- SS> a 25mhz 030 (50% walls 50% floor/ceiling). I'm just talking about
- SS> rendering
- SS> it to fastmem, no c2p into chipmem... I just wanted to point out that
- SS> the CPU is not the problem in the amiga to play something like doom.
- SS> (Well 14mhz 020 are)
-
- So you tested this? Remember TextDemo? It gets max 6 FPS on my 68030/22, of
- which about 30% is C2P (info: 320x240 1x1, 'DOOM' shading using a look-up
- table). That gets you let's say about 9 FPS (being VERY generous) if C2P was
- disabled. Not even near 25 FPS, and I'm pretty damn sure it cannot be sped up
- more than 25% orso.
-
- >> Do you really think so? On 030 the fastest instructions available take 2
- >> cycles, while most instructions on 486 take 1 cycle. 486 also has much
- >> faster Mul and Div instructions. You would be better of comparing the 040
- >> with the 486.
-
- SS> well, the 68030 as 16 'general' purpose register (Let not argue on
- SS> the details:) as a separate data/inst cache . Does the 486 halt during
- SS> buss access? Do you know how many cycles are needed to tmap a gouraud
- SS> lighted pixel on a 486? on a 030 is 28cycle or so.
-
- 28 cycles, with a huge look-up table, and all kinds of stupid limitations which
- are completely impractical for a REAL game.
-
- SS> Also this take 6 cycle on a 030: addi.l #32bitvalue,(localvariable,a7)
- SS> how many on a 486 and 386? (Sorry I dont have any x86 stuff:( )
-
- What's the point? Isn't it clear that the 486 is much faster than the 030? The
- difference in cache-size alone gives the 486 a huge speed-up.
-
- >> SS> Again the problem is not c2p but slow video memory...Does PC alway
- >> SS> cache video memory on the L1 cache? I hear many people rendering in
- >> SS> local mem then doing a copy.
-
- >> Of course the video memory is not cached in the L1 cache, for the same
- >> reason as ChipRAM isn't cached on Amiga. To copy the stuff to video ram
- >> why not simply ask the DMA controller to copy that shit for you while you
- >> render the next frame? Also why wouldn't the same trick to get 'free'
- >> cycles on Amiga while doing ChipRAM writes work with the clones much
- >> faster Video RAM? While writing the pixel to video ram the processor
- >> continues to calculate the next TMapped pixel.
-
- SS> I'm very unfamiliar with PC... but I heard that the PC dma is dead
- SS> slow like 4meg second to dma system memory... does all PC video card
- SS> have DMA?
-
- Most clones have got a DMA controller on the motherboard which you can simply
- ask to move memory to a specific location (ie, its very general purpose). The
- speed depends heavily on what kind of clone you have, but 4 MB/second is
- probably the lower-limit at the moment.
-
- >> I doubt this 10 cycle routine of yours is very usefull for realistic
- >> purposes judging from all the 'unrealistic' TMap routines I've seen here
- >> lately (ones with rely on 64K boundaries or too big or too small
- >> Textures).
-
- SS> It texture anything using 64K tmap with 24bit/16bit/8bit fixpoint
- SS> and gouraud lighting. I do not see why using segment is 'unrealistic'
- SS> ?!
-
- So how big are the look-up tables? What is the minimum size of a texture?
-
- >> It DIDN'T work in local fastmem (did I say that?). This included C2P
- >> time. It was run in 320x240 1x1, 8-bit, full floors, ceilings and walls
- >> in DOOM style.
-
- SS> I bet that if you had a 300mhz 060 the c2p wont be any faster then the
- SS> 25mhz 040... The 060 is not to blame.
-
- And that's exactly my point, you can't get a fast 'exact' DOOM clone for Amiga
- as the ChipRAM will always bog it down far too much. Requiring a fast 040 or
- 060 to get a 'playable but still not as fast as my friends 486/33' 'exact' DOOM
- clone is quite unrealistic.
-
- SS> This is what I beleive the amiga as enought power even with a 25mhz
- SS> 030 to do doom(but flat shaded tmaping) at ~15-20fps if it had a
- SS> local bus video card.
-
- Backing down from your 25 FPS claim above? If you back down a little bit
- further to say 10 FPS then I might actually agree with you.
-
- Grtz John
-
- -----------------------------------------------------------------------
- John.Hendrikx@grafix.xs4all.nl TextDemo/FastView/Etc... development
- -----------------------------------------------------------------------
- -- Via Xenolink 1.985B5, XenolinkUUCP 1.1
-