@pavlor

here you can a HW accel warp 3d for winuae that run with blitzquake gl Version.

http://quarktex.kontechs.de/

you can run very high resolution and full scene 4* antialiasing and 16* anisotropic filtering and get high framerates.

have i miss something where is the Cyberstorm PPC benchmark?

_________________
The Machine: Bride Of The Pin•Bot by Williams Electronics

@bernd_afa

Quote:

bernd_afa wrote: @Pavlor >GLQuake (BlitzQuake_68k) on WinUAE box of my brother: Core 2 Q6600 2.4 GHz, >GForce 8600GT; WinUAE 1.5.0: 68040 JIT, 128 MB Fast, 16 MB GFX, OS3.9; scale >instead of switching resolution used). Wazp3D beta 45. you should not use scale resolution, because it cost lots performance to scale the 320*200.please try without it. but anyway i try your test too.I get 16 frmaes too on my AMD64 with real 1,8 GHZ.but i have only 256 kb 2. Level Cache, so i thought the core duo with the large caches must be lots faster in software 3d. doing real 32 bit 3d with CPU seem really performance hungry.so the 8 bit software renderer quake have and give lower quality is important. Or maybe there is a Problem in wazp, i notice drawing problems and the status bar flicker.

I get ~40 FPS from Intel Core 2 Duo P8700 (25 Watts, 2.5Ghz, 3MB L2 cache) , demo001 timedemo, normal settings, 640x480.

This is rendered via Direct3D 9b JIT-X86 software renderer i.e. Swiftshader 2.01. I used WinMESA OpenGL-to-Direct3D bridge.

I have haven't tried QuarkTex+WinUAE with Swiftshader.
You can download Swiftshader demo from http://www.transgaming.com/business/swiftshader/

Last edited by Hammer on 11-Dec-2009 at 10:33 AM.
Last edited by Hammer on 11-Dec-2009 at 10:32 AM.
Last edited by Hammer on 11-Dec-2009 at 10:30 AM.

_________________
Amiga 1200 (rev 1D1, KS 3.2, PiStorm32/RPi CM4/Emu68)
Amiga 500 (rev 6A, ECS, KS 3.2, PiStorm/RPi 4B/Emu68)
Ryzen 9 7950X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB

here's one card missing from the list Blizzard PPC 321Mhz 80.333Mhz bus. OS4.0


Speed test for FFT + iFFT: (float)
time needed 17928ms for 413696 samples, => .261626005172729x speed @44100Hz/stereo
Speed test for FFT + iFFT: (integer)
time needed 15377ms for 413696 samples, => .30502900481224x speed @44100Hz/stereo
Speed test for FFT + iFFT: (integer handoptimized 68K ASM)
time needed 12774ms for 413696 samples, => .367185741662979x speed @44100Hz/stereo


im not happy with the results. this card will now be modify/reviewed again as it needs to show better performance.

Last edited by delshay on 12-Dec-2009 at 06:51 PM.
Last edited by delshay on 12-Dec-2009 at 10:52 AM.
Last edited by delshay on 12-Dec-2009 at 10:51 AM.
Last edited by delshay on 12-Dec-2009 at 10:32 AM.

_________________
The Machine: Bride Of The Pin•Bot by Williams Electronics

@delshay

I think here you cant do anything.Your mem transfer speed is lower as a 604 because 604 have 64 bit access and memtransger is 2* faster as on 68k.the 68k and 603e have only 32 bit and reach same mem transfer speed on cybppc.

but the 68k is only clocked with 50 MHZ and need not so much wait for data.

also the 604 have 32 kb data cache the 603e only 16 kb.this with the the lower memory bandwidth make the 603e speed not increase lot if you clock it higher.

The 604e have 3 integer units and the 603e is a slim CPU designs same as 440ep.It does not perform so good in MHZ /performance as a 604 or G4 core.

thanks for your values, i calc performance per MHZ now and add it to the other thread.

Last edited by bernd_afa on 13-Dec-2009 at 02:30 PM.
Last edited by bernd_afa on 12-Dec-2009 at 01:33 PM.
Last edited by bernd_afa on 12-Dec-2009 at 01:30 PM.

@all

Something wrong with the forum. I cant see my full post only some part...

OWB (AOS4.1) and Firefox (WinXP) shows only part of my post.

Last edited by mr2 on 19-Jan-2010 at 10:10 PM.
Last edited by mr2 on 19-Jan-2010 at 10:02 PM.

_________________
Sam440ep-flex 800MHz 1GB RAM R9250 128MB SB Live!

can someone with a Cyberstorm PPC 233 do this test please.

Last edited by delshay on 20-Jan-2010 at 04:50 AM.

_________________
The Machine: Bride Of The Pin•Bot by Williams Electronics

@all
SAM 440flex 800MHz, RAM-1GB, HDD-320GB 7200rpm 2,5''(WB 1920x1200 32Bit):

mkfile hd:file 200000000 test

OS4.1(Beta):

SFS0->RAM: 14s cpu 98-100%
SFS0->SFS0: 13s cpu 65-71%
JXF->RAM: 15s cpu 95-100%

OS4.1 Update 1:

SFS0->RAM: 5s cpu 75-83%
SFS0->SFS0: 10s cpu 24-29%
JXF->RAM: 5s cpu 79-85%



cube test

OS4.1(Beta):

7776 frames drawn in 16.0 seconds
764 fps (1.308 milliseconds per frame)
489 fps (2.041 milliseconds per frame with swap)

OS4.1 Update 1:

7701 frames drawn in 15.9 seconds
777 fps (1.286 milliseconds per frame)
491 fps (2.036 milliseconds per frame with swap)

_________________
Sam440ep-flex 800MHz 1GB RAM R9250 128MB SB Live!

@Tuxedo

Here are my new results (Core 2 Quad Q6600 2400 MHz, WinUAE 1.5: 68040, 256 MB RAM, 16 MB GFX RAM, OS 3.9) - JIT is set to direct (my former results were with indirect and thus lower).


lame 3.98-2 test:

8.Applications:Utils/lame-3.98.2d-m68k/bin> lame040fpu DH3:archiv/WhereDoAllTheJunkiesComeFrom.mp3.wav
LAME 3.98.2 32bits (http://www.mp3dev.org/)
Using polyphase lowpass filter, transition band: 16537 Hz - 17071 Hz
Encoding DH3:archiv/WhereDoAllTheJunkiesComeFrom.mp3.wav
to DH3:archiv/WhereDoAllTheJunkiesComeFrom.mp3.wav.mp3
Encoding as 44.1 kHz j-stereo MPEG-1 Layer III (11x) 128 kbps qval=3
Frame | CPU time/estim | REAL time/estim | play/CPU | ETA
9467/9467 (100%)| 1:00/ 1:00| 1:02/ 1:02| 4.0742x| 0:00
-------------------------------------------------------------------------------
kbps LR MS % long switch short %
128.0 6.7 93.3 88.6 6.9 4.5
Writing LAME Tag...done
ReplayGain: -5.4dB


BYTEmark (tm) Native Mode Benchmark ver. 2 (3/95)
NUMERIC SORT: Iterations/sec.: 306.024754 Index: 7.908638
STRING SORT: Iterations/sec.: 6.786321 Index: 2.982998
BITFIELD: Iterations/sec.: 62357108.844797 Index: 10.696218
FP EMULATION: Iterations/sec.: 19.555419 Index: 9.401644
FOURIER: Iterations/sec.: 1623.814347 Index: 1.838641
ASSIGNMENT: Iterations/sec.: 4.057519 Index: 15.458981
IDEA: Iterations/sec.: 573.580724 Index: 8.775715
HUFFMAN: Iterations/sec.: 353.227579 Index: 9.815967
NEURAL NET: Iterations/sec.: 0.816089 Index: 1.380861
LU DECOMPOSITION: Iterations/sec.: 25.920606 Index: 1.530323
...done...
===========OVERALL============
INTEGER INDEX: 8.482253
FLOATING-POINT INDEX: 1.572080 (interestingly low result)
(90 MHz Dell Pentium = 1.00)
==============================

with JIT set to indirect:
===========OVERALL============
INTEGER INDEX: 4.782673
FLOATING-POINT INDEX: 31.663610
(90 MHz Dell Pentium = 1.00)
==============================

Aflops, AFlopsM68.40 (JIT direct)

Amiga Flops Benchmark (Double Precision), V2.0

Module Error RunTime MFLOPS
(usec)
1 2.6660e-07 0.0850 164.7059
2 -4.5477e-08 0.0773 90.5051
3 -3.4509e-09 0.0980 173.3865
4 -9.8962e-09 0.0759 197.7343
5 -4.3696e-09 0.1677 172.8924
6 -1.9106e-09 0.1560 185.8788
7 2.2564e-05 0.1535 78.1679
8 -2.5409e-09 0.1567 191.4257

Iterations = 256000000
NullTime (usec) = 0.0041
MFLOPS(1) = 107.2683
MFLOPS(2) = 127.0140
MFLOPS(3) = 163.5139
MFLOPS(4) = 186.9963

JIT - indirect:
Iterations = 128000000
NullTime (usec) = 0.0041
MFLOPS(1) = 44.4148
MFLOPS(2) = 57.4975
MFLOPS(3) = 81.2875
MFLOPS(4) = 96.8407

AmiGOD 2 (JIT direct)
Math (CPU) - 2405


Speedometer 4.0 (ShapeShifter)
(JIT Direct)
CPU - 28.016
Math - 648.531
(JIT indirect)
CPU - 15.648
Math - 236.687

Conclusion: Now, it is as fast as 750FX 800 MHz in some tests, in some much slower (I can´t still play DVDs at full speed).

@all

remake all cpu test with lame 3.98.4 (maybe os4 version is 3.98.2) and bladeENC 0.94.2
with this sample wave:

xoomer.virgilio.it/tuxcam/wave/osaka_theory.rar (it was an mp3 music from demoscene archives)

only on cpu from 500 to 1500 mhz (amigaOS4 and MOS machine) and x86 system (from 500 to 2000 mhz - P4 require more mhz to be like a 2 ghz AMD).

_________________

@HAmmer

Quote:

I used WinMESA OpenGL-to-Direct3D bridge.

I'd like to give this a try myself, but I couldn't find an OpenGL-to-Direct3D bridge called WinMESA. Could you point me to this software? Thanks!

@c0d1f1ed

Welcome!

@ALL

Here a nice SDL bench downloadable from:

http://code.google.com/p/os4sdl/downloads/list

My results(Pegasos2 G4@1131 - Radeon9250):


Mode = 320x240, software
Pitch = 320
Hardware surfaces avail = 1
Window manager avail = 1
Blitter hardware = 1
Colorkey blit hardware = 0
Alpha blit hardware = 0
Software->Hardware accel = 0
Video memory = 0

Slow points test
Fast points test
Rect fill test
32x32 Blitter test
Mode = 320x240, hardware
Slow points test
Fast points test
Rect fill test
32x32 Blitter test
Mode = 640x480, software
Slow points test
Fast points test
Rect fill test
32x32 Blitter test
Mode = 640x480, hardware
Slow points test
Fast points test
Rect fill test
32x32 Blitter test
320x240 320x240 640x480 640x480
software hardware software hardware
Slow points (frames/sec): 12.0846 148.148 1.59968 51.6129
Fast points (frames/sec): 755.162 118.849 194.973 29.8786
Rect fill (rects/sec): 21005.1 113778 6511.92 55351.4
32x32 blits (blits/sec): 55351.4 141241 53194.8 141241


Happy testing :)

Last edited by Tuxedo on 28-Jan-2011 at 08:08 PM.

_________________
Simone"Tuxedo"Monsignori, Perugia, ITALY.

@ALL

updated first page with my new results on my Peg2@1131...

If some Sam460 user will do some tests was really nice :)

_________________
Simone"Tuxedo"Monsignori, Perugia, ITALY.

@ALL

Updated with the c-ray test from OS4Depot(http://os4depot.net/share/utility/benchmark/c-ray.lha):

[G4@1131]

8.RAM Disk:c-ray-1.1> c-ray-f -i scene -o foo.ppm
Rendering took: 3 seconds (3960 milliseconds)
8.RAM Disk:c-ray-1.1> c-ray-f -i sphfract -o foo.ppm
Rendering took: 81 seconds (81147 milliseconds)

Last edited by Tuxedo on 29-May-2011 at 02:33 PM.

_________________
Simone"Tuxedo"Monsignori, Perugia, ITALY.

@Tuxedo


13.Programs:Benchmarks/c-ray-1.1/c-ray-f -i scene -o foo.ppm
Rendering took: 5 seconds (5435 milliseconds)
13.Programs:Benchmarks/c-ray-1.1> c-ray-f -i sphfract -o foo.ppm
Rendering took: 128 seconds (128769 milliseconds)

Quote:

Tuxedo wrote: @ALL Updated with the c-ray test from OS4Depot(http://os4depot.net/share/utility/benchmark/c-ray.lha): 8.RAM Disk:c-ray-1.1> c-ray-f -i scene -o foo.ppm Rendering took: 3 seconds (3960 milliseconds) 8.RAM Disk:c-ray-1.1> c-ray-f -i sphfract -o foo.ppm Rendering took: 81 seconds (81147 milliseconds)

_________________
Sam460ex : Radeon Rx550 Single slot Video Card : SIL3112 SATA card

5435 is compareable to (only 10% slower than) Intel Pentium3 800 MHz.
3960 is half of the speed of a Intel Pentium4 2.6 GHz.

or something like that

Last edited by KimmoK on 29-May-2011 at 07:11 PM.

_________________
- KimmoK
// For freedom, for honor, for AMIGA
//
// Thing that I should find more time for: CC64 - 64bit Community Computer?

@Seiya

Quote:

Seiya wrote: @all remake all cpu test with lame 3.98.4 (maybe os4 version is 3.98.2) and bladeENC 0.94.2 with this sample wave: xoomer.virgilio.it/tuxcam/wave/osaka_theory.rar (it was an mp3 music from demoscene archives) only on cpu from 500 to 1500 mhz (amigaOS4 and MOS machine) and x86 system (from 500 to 2000 mhz - P4 require more mhz to be like a 2 ghz AMD).

We now have Intel Atom....

_________________
Amiga 1200 (rev 1D1, KS 3.2, PiStorm32/RPi CM4/Emu68)
Amiga 500 (rev 6A, ECS, KS 3.2, PiStorm/RPi 4B/Emu68)
Ryzen 9 7950X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB

glquake
WinUAE with QuarkText
75 fps in 640x480 on Intel core2 E8500@3200

_________________

@KimmoK

Not so bad... :)

_________________
Simone"Tuxedo"Monsignori, Perugia, ITALY.