home *** CD-ROM | disk | FTP | other *** search
- On Fri, 2 Feb 1996, Magnus Kollberg wrote:
-
- > Wasn't it someone else who had tested the other speeds that Petr couldn't test?
- > Can't that person fill in those figures here?
-
- Yep. I'm back again with full test thanks to Georg Acher (father of
- FX-card) and his new software update (only pre-alpha, but helps me a lot!)
-
- Standard DOOM.WAD (4234124 bytes long), Falcon with FX-card installed
- (=> DSP @ 50 MHz), output to VGA monitor, standard resolution.
-
- Without FastRAM CPU @ | 16 MHz | 20 MHz | 32 MHz | 36 MHz | 40 MHz | 68040 |
- ----------------------+--------+--------+--------+--------+--------+-------+
- CPU version of BM132a | 6.25 | 8.33 | 7.73 | 9.09 | 9.70 | ?!?!? |
- DSP version of BM132a | 9.30 | 12.12 | 11.68 | 13.45 | 14.29 | ?!?!? |
- texture mapped BM170a | 4.38 | 5.52 | 5.84 | 6.67 | 7.31 | ?!?!? |
- ----------------------+--------+--------+--------+--------+--------+-------+
- With FX FastRAM (3-cycle-per-write modification) |
- ----------------------+--------+--------+--------+--------+--------+-------+
- CPU version of BM132a | 7.62 | 9.70 | 9.88 | 11.27 | 12.60 |!!!?!!!|
- DSP version of BM132a | 10.74 | 13.45 | 13.68 | 15.53 | 17.39 |!!!?!!!|
- texture-mapped BM170a | 5.10 | 6.18 | 6.84 | 7.66 | 8.47 |!!!?!!!|
- ----------------------+--------+--------+--------+--------+--------+-------+
-
- I didn't include results of AfterBurner test because I don't believe AB
- is so slow. The only chance is that that guy run BM from ST memory.
- Moreover I'd like to see BM test on AB with accelerated DSP! :-)
-
- Petr
-
- --
- +--------------------------------------------------------------------------+
- | stehlik@leo1.zlin.vutbr.cz -- 2:421/36@fidonet.org -- 90:1200/2@nest.ftn |
- | Falcon030/FPU/FX/40MHz/4ST+8FX RAM/IDE 1080 MB running TOS || Bad Mood |
- | MegaSTE/16MHz/4MB RAM/240MB/Riebl card/USR DS 16k8 running Joy BBS 24hrs |
- +--------------------------------------------------------------------------+
-
-