11.2.3 Mighty&Nemesis MightySonic und die Nemesis-Karte im Falcon
Ein Artikel von
Michael Grove
For 40MHz CPU and 20MHz bus (VGA only)
1).
2).
3).
4).
One, you place a capacitor across ground and DMA clock-in with
Two, you use a resistor (no capacitor) across DMA clock-in and
These do not work with the Mighty. Only use the DMA buffer
5).
6).
The Mighty can run 24/24 CPU/BUS for better TC Video modes (VGA).
Run Nemesis at 48MHz, but do not run 32MHz.PRG on the Mighty. Mighty
gets 24MHz clock from expansion bus. My FPU will not work in this
mode.
I tried to run the Mighty at 48MHz. It would work 1 of 20 times.
Not reliable. I habe 50Mhz 68030, but I think the problem is the simm
speed/chips on the Mighty. The Mighty runs the whole card at CPU clock
speed. Maybe Nemesis mit 42,44,46 MHz clock, I don't know. Aixit
Falcon and TT fast ram cards habe same problem. Will not run much over
40MHz.
Special note on Nemesis. Both oscillators must be installed even
if you only use one. Nemesis will not work properly if only one is
installed.
7).
GEM Bench v4.01 Ofir Gal - 14 Jan 95 ============================================ Falcon 030 TOS 4.04 AES v3.40 GEMDOS v0.48 MiNT not present Blitter Disabled NVDI 4.11 present Video Mode: 640*480*2 LineF FPU installed Run and Malloc from FastRAM Ref: F030 + FPU, 640*480*2 Mighty Sonic 40/20 CPU/BUS Nemesis installed FPU mit 64Mhz clock ============================================ GEM Dialog Box: 0.645 542% VDI Text: 0.220 1545% VDI Text Effects: 0.460 1628% VDI Small Text: 0.270 1388% VDI Graphics: 1.135 769% GEM Window: 0.425 329% Integer Division: 1.230 252% Float Math: 0.185 183% RAM Access: 0.815 269% ROM Access: 0.815 257% Blitting: 0.910 123% VDI Scroll: 0.785 336% Justified Text: 0.930 350% VDI Enquire: 0.270 585% New Dialogs: 1.045 425% ============================================ Graphics: 729% CPU: 240% Average: 598% ******************************************************** NemBench v1.0 - precision CPU/FPU profiler. Integer multiply (16bit) -> 1.532 Mips (~249%) Integer divide (16bit) -> 0.907 Mips (~250%) Linear (stalled) integer -> 19.980 Mips (~250%) Interleaved (piped) integer -> 19.980 Mips (~250%) Float multiply (64bit) -> 0.662 MegaFlops (~250%) Float divide (64bit) -> 0.432 MegaFlops (~250%) Linear (stalled) float -> 1.288 MegaFlops (~242%) Interleaved (piped) float -> 1.288 MegaFlops (~242%) 16bit read (100% hit) -> 19.704 MByte/sec (~250%) 16bit write (100% hit) -> 9.940 MByte/sec (~166%) 32bit read (100% hit) -> 39.447 MByte/sec (~250%) 32bit write (100% hit) -> 19.920 MByte/sec (~300%) Linear 256k read (STRam) -> 6.472 MByte/sec (~123%) Linear 256k write (STRam) -> 7.142 MByte/sec (~112%) Linear 256k copy (STRam) -> 3.577 MByte/sec (~112%) Linear 256k read (FastRAM) -> 21.522 MByte/sec (~409%) Linear 256k write (FastRAM) -> 19.176 MByte/sec (~300%) Linear 256k copy (FastRAM) -> 6.136 MByte/sec (~192%) ********************************************************* GEM Bench v4.01 Ofir Gal - 14 Jan 95 ============================================ Falcon 030 TOS 4.04 AES v3.40 GEMDOS v0.48 MiNT not present Blitter Disabled NVDI 4.11 present Video Mode: 640*480*2 LineF FPU installed Run and Malloc from FastRAM Ref: F030 + FPU, 640*480*2 Mighty Sonic 24Mhz CPU & BUS Nemesis installed/FPU mit 64Mhz clock ============================================ GEM Dialog Box: 0.745 469% VDI Text: 0.280 1214% VDI Text Effects: 0.615 1217% VDI Small Text: 0.315 1190% VDI Graphics: 1.485 587% GEM Window: 0.485 288% Integer Division: 2.055 150% Float Math: 0.190 178% RAM Access: 0.995 221% ROM Access: 0.995 211% Blitting: 1.250 89% VDI Scroll: 0.840 314% Justified Text: 1.125 289% VDI Enquire: 0.325 486% New Dialogs: 1.280 347% ============================================ Graphics: 590% CPU: 190% Average: 483% ******************************************************** NemBench v1.0 - precision CPU/FPU profiler. Integer multiply (16bit) -> 0.921 Mips (~150%) Integer divide (16bit) -> 0.544 Mips (~150%) Linear (stalled) integer -> 11.959 Mips (~150%) Interleaved (piped) integer -> 11.959 Mips (~150%) Float multiply (64bit) -> 0.798 MegaFlops (~301%) Float divide (64bit) -> 0.518 MegaFlops (~299%) Linear (stalled) float -> 1.409 MegaFlops (~264%) Interleaved (piped) float -> 1.402 MegaFlops (~264%) 16bit read (100% hit) -> 11.820 MByte/sec (~150%) 16bit write (100% hit) -> 6.684 MByte/sec (~112%) 32bit read (100% hit) -> 23.640 MByte/sec (~150%) 32bit write (100% hit) -> 13.368 MByte/sec (~201%) Linear 256k read (STRam) -> 6.798 MByte/sec (~129%) Linear 256k write (STRam) -> 8.115 MByte/sec (~127%) Linear 256k copy (STRam) -> 3.998 MByte/sec (~125%) Linear 256k read (FastRAM) -> 10.164 MByte/sec (~193%) Linear 256k write (FastRAM) -> 12.850 MByte/sec (~201%) Linear 256k copy (FastRAM) -> 6.425 MByte/sec (~201%)
Copyright © Robert Schaffner (support@doitarchive.de) Letzte Aktualisierung am 23. Dezember 2003 |