@recedent

nice thank you!

tomorrow will update :)

_________________
Simone"Tuxedo"Monsignori, Perugia, ITALY.

@Tuxedo

Some more speed tests for PowerBook G4 1.67 GHz, Radeon 9700, MorphOS 3.9:

Ram Disk:> AFinchM68.20
Amiga Fibonacci Benchmark V1.0 By Chris De Maeyer
(C)1999 Visionary Software - Released as Freeware

Calculating, please wait...
Run Time Seconds = 9.080

Ram Disk:> AFinchWOS
Amiga Fibonacci Benchmark V1.0 By Chris De Maeyer
(C)1999 Visionary Software - Released as Freeware

Calculating, please wait...
Run Time Seconds = 3.820

EDIT: I need to re-run the AFlops 68k, the results seem awkward.
EDIT2: There's a huge speed difference in AFlops68k under MorphOS. The 020 version runs slow (as there was no JIT translation done by Trance). Here are the Aflops68k results for both 020 and 040 versions:

7_6gb:> aflops/AFlopsM68.20

Amiga Flops Benchmark (Double Precision), V2.0

Module Error RunTime MFLOPS
(usec)
1 -1.5632e-13 3.0925 4.5271
2 -1.0347e-13 1.5700 4.4586
3 -3.1197e-14 3.7000 4.5946
4 7.7938e-14 3.4900 4.2980
5 -3.2641e-14 6.5600 4.4207
6 -9.9920e-16 6.5450 4.4309
7 -5.5650e-11 2.8575 4.1995
8 2.7700e-14 6.7550 4.4412

Iterations = 8000000
NullTime (usec) = 0.0050
MFLOPS(1) = 4.5022
MFLOPS(2) = 4.3656
MFLOPS(3) = 4.4242
MFLOPS(4) = 4.4412

7_6gb:> aflops/AFlopsM68.40

Amiga Flops Benchmark (Double Precision), V2.0

Module Error RunTime MFLOPS
(usec)
1 2.8422e-14 0.1605 87.2444
2 2.5047e-13 0.1311 53.3969
3 -7.6605e-15 0.1647 103.2258
4 2.2771e-13 0.1330 112.8085
5 3.8858e-14 0.2847 101.8661
6 7.5495e-15 0.2666 108.7925
7 -1.1369e-13 0.2522 47.5836
8 1.2612e-13 0.2634 113.8790

Iterations = 128000000
NullTime (usec) = 0.0080
MFLOPS(1) = 63.4026
MFLOPS(2) = 75.3174
MFLOPS(3) = 95.7377
MFLOPS(4) = 109.9490

Ram Disk:> AFlopsWOS

Amiga Flops Benchmark (Double Precision), V2.0

Module Error RunTime MFLOPS
(usec)
1 4.0146e-13 0.0362 387.0410
2 -1.4166e-13 0.0238 293.7705
3 4.7184e-14 0.0270 628.9017
4 -1.2546e-13 0.0398 376.4706
5 -1.3800e-13 0.0661 438.7707
6 3.2374e-13 0.0429 675.5232
7 -8.4583e-11 0.0951 126.2120
8 3.4855e-13 0.0460 651.9525

Iterations = 512000000
NullTime (usec) = 0.0012
MFLOPS(1) = 355.7456
MFLOPS(2) = 256.5908
MFLOPS(3) = 413.4056
MFLOPS(4) = 584.0060

Last edited by recedent on 18-Jan-2018 at 11:44 AM.
Last edited by recedent on 18-Jan-2018 at 11:10 AM.
Last edited by recedent on 18-Jan-2018 at 11:04 AM.

@recedent

I did some more tests on PowerBook 1.67.

lame_vmx (Altivec):

Ram Disk:lame-3.99.5-morphos> lame_vmx ram:hearth.wav RAM:hearth.mp3
LAME 3.99.5 32bits (http://lame.sf.net)
Using polyphase lowpass filter, transition band: 16538 Hz - 17071 Hz
Encoding ram:hearth.wav to RAM:hearth.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
12205/12205 (100%)| 0:26/ 0:26| 0:26/ 0:26| 12.146x| 0:00
-------------------------------------------------------------------------------
kbps LR MS % long switch short %
128.0 4.1 95.9 96.5 2.2 1.3
Writing LAME Tag...done
ReplayGain: +3.5dB

===

lame no Altivec:

Ram Disk:lame-3.99.5-morphos> lame ram:hearth.wav RAM:hearth.mp3
LAME 3.99.5 32bits (http://lame.sf.net)
Using polyphase lowpass filter, transition band: 16538 Hz - 17071 Hz
Encoding ram:hearth.wav to RAM:hearth.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
12205/12205 (100%)| 0:58/ 0:58| 0:59/ 0:59| 5.4185x| 0:00
-------------------------------------------------------------------------------
kbps LR MS % long switch short %
128.0 4.1 95.9 96.5 2.2 1.3
Writing LAME Tag...done
ReplayGain: +3.5dB

===
===
and here's a result for AFlops version 020 from my Dual 2,7 GHz G5 PowerMac (dramaticaly slower than the 040 version used before):

Amiga Flops Benchmark (Double Precision), V2.0

Module Error RunTime MFLOPS
(usec)
1 -1.5632e-13 1.9575 7.1520
2 -1.0347e-13 0.9525 7.3491
3 -3.1197e-14 2.2850 7.4398
4 7.7938e-14 2.2200 6.7568
5 -3.2641e-14 4.1375 7.0091
6 -9.9920e-16 4.0675 7.1297
7 -5.5650e-11 1.8475 6.4953
8 2.7700e-14 4.2375 7.0796

Iterations = 8000000
NullTime (usec) = 0.0050
MFLOPS(1) = 7.3785
MFLOPS(2) = 6.8911
MFLOPS(3) = 7.0353
MFLOPS(4) = 7.1038

From EAB on the new "Rapture" board under AmigaOS 3.9/Warpos.
Dnetc
RC5-72
4,112,400 keys/sec ---> Altivec ---> (Rapture G4@450MHz)
OGR-NG
8,892,416 nodes/sec ---> Altivec ---> (Rapture G4@450MHz)
5,051,323 nodes/sec ---> NON AltiVec ---> (Rapture G4@450MHz)

7.Ram Disk:> AflopsWOS

Amiga Flops Benchmark (Double Precision), V2.0

Module Error RunTime MFLOPS
(usec)
1 -7.6739e-13 0.1022 137.0031
2 -5.7021e-13 0.0823 85.0902
3 -2.4314e-14 0.0578 294.0541
4 6.8612e-14 0.0823 182.3362
5 -1.6209e-14 0.1600 181.2500
6 1.3961e-13 0.0889 326.1863
7 -3.6152e-11 0.2823 42.5131
8 8.9373e-15 0.0956 313.7255

Iterations = 256000000
NullTime (usec) = 0.0045
MFLOPS(1) = 110.8410
MFLOPS(2) = 93.8136
MFLOPS(3) = 167.9971
MFLOPS(4) = 280.3369

@Tuxedo

MPlayer benchmark from PowerBook G4 1,67 Ghz (MorphOS 3.9):

BENCHMARKs: VC: 28.152s VO: 0.017s A: 0.000s Sys: 0.898s = 29.067s
BENCHMARK%: VC: 96.8516% VO: 0.0573% A: 0.0000% Sys: 3.0911% = 100.0000%

Quote:

Goodtar wrote: From EAB on the new "Rapture" board under AmigaOS 3.9/Warpos.

cough cough "Rapture board under AmigaOS 3.9/Warpos"

_________________
____________________________
c64-2sids, A1000, A1200T-060@50(finally working!),A4000-CSMKIII
! My Master Miggies- Amiga 1000 & AmigaOne X1000 !
mancave-ramblings
X1000 I BELIEVE

@klx300r

yep and thats just the slow version(7410@450) will be testing G4 7457/744x soon and other PPC's
the final board will be a 7457 or 7447a that is the fastest RAPTURE G4

Last edited by kev7112001 on 22-Jan-2018 at 07:43 PM.
Last edited by kev7112001 on 22-Jan-2018 at 07:41 PM.
Last edited by kev7112001 on 22-Jan-2018 at 07:40 PM.

@recedent

Thank you for your and X5000 benchmarks

However seems that X5000 perform quite similar to AmigaOS4 and that will increase my thoughts about that X1000 was simlar if not faster(especially with AltiVec on) than X5000 on heavy tasks(like watching a video)...that was not good imho unfortunately...

@Goodtar and @kev7112001 

Really nice board, but...AmigaOS4 or MOS will run on top of it? If not, sorry but really dont see the reason of such a board...that remember me the first EyeTech/Amiga idea about equipping Amiga classics with a ppc...

Last edited by Tuxedo on 22-Jan-2018 at 09:02 PM.

_________________
Simone"Tuxedo"Monsignori, Perugia, ITALY.

Quote:

kev7112001 wrote: @klx300r yep and thats just the slow version(7410@450) will be testing G4 7457/744x soon and other PPC's the final board will be a 7457 or 7447a that is the fastest RAPTURE G4

1st off Welcome aboard
2nd this is crazy cool news, thanks for posting

btw, this deserves its own thread when you get a chance

Last edited by klx300r on 22-Jan-2018 at 09:06 PM.
Last edited by klx300r on 22-Jan-2018 at 09:05 PM.

_________________
____________________________
c64-2sids, A1000, A1200T-060@50(finally working!),A4000-CSMKIII
! My Master Miggies- Amiga 1000 & AmigaOne X1000 !
mancave-ramblings
X1000 I BELIEVE

@klx300r

yeah project is at EAB little birdy told me someone posted some of our results here so i had to see:)

@Tuxedo

we are tryin to get as much support as possible this is just being used as a co-pro in os3.x WOS

Last edited by kev7112001 on 22-Jan-2018 at 10:10 PM.

@kev7112001

Yes like bppc, but if OS4 will run on top of it was really a nice thing...how to put a gfx board on it? only mediator?

_________________
Simone"Tuxedo"Monsignori, Perugia, ITALY.

@Tuxedo
yes atm you will need a mediator as we need pci for this to work im am looking at other possibilities like designing a new pci expansion with what we need to boot it (OSs4/MOS etc)which can be done most our issues are mediator drivers etc i like to do like a cyberstormppc and have pci on the faster side of the bus but baby steps for now as many things need work as with a 7457/7447a a voodoo is realy holding us back in gfx

Last edited by kev7112001 on 22-Jan-2018 at 10:21 PM.
Last edited by kev7112001 on 22-Jan-2018 at 10:19 PM.

@kev7112001
It looks like a great project :) hopefully these are are easier to possess then a Crescendo !

Quote:

kev7112001 wrote: @Tuxedo yes atm you will need a mediator as we need pci for this to work im am looking at other possibilities like designing a new pci expansion with what we need to boot it (OSs4/MOS etc)which can be done most our issues are mediator drivers etc i like to do like a cyberstormppc and have pci on the faster side of the bus but baby steps for now as many things need work as with a 7457/7447a a voodoo is realy holding us back in gfx

I have to bug a buddy of mine to see if he still has his old PowerMac G4 with PPC7455@1.4GHz as my 1200T with mediator is waiting ...but right now she's humming with a Radeon 9250 with 256 MB ram with Picasso...

_________________
____________________________
c64-2sids, A1000, A1200T-060@50(finally working!),A4000-CSMKIII
! My Master Miggies- Amiga 1000 & AmigaOne X1000 !
mancave-ramblings
X1000 I BELIEVE

@klx300r

Quote:

I have to bug a buddy of mine to see if he still has his old PowerMac G4 with PPC7455@1.4GHz as my 1200T with mediator is waiting

And how you're going to stick the PowerMac G4's CPU module to your Mediator? With a roll of duct tape? PowerMac G4 has got a custom, 300-pin connector for that, not a PCI slot. Better bug him to boot MorphOS 3.9 on it and do some benchmarks to get this thread back on topic.

@Tuxedo

fs-uae on threadripper 1950x

dnetc v2.9112-521-CTR-16020313 for AmigaOS (OS 4.1, PowerPC).
Please provide the *entire* version descriptor when submi ...
The distributed.net bug report pages are at http://bugs.d ...

[Jan 24 20:01:44 UTC] Automatic processor type detection ...
a PowerPC 604e processor.
[Jan 24 20:01:44 UTC] OGR-NG: using core #0 (KOGE 3.1 Sca ...
[Jan 24 20:02:04 UTC] OGR-NG: Benchmark for core #0 (KOGE ...
0.00:00:17.10 [6,159,255 nodes/sec]
[Jan 24 20:02:04 UTC] OGR-NG benchmark summary :
Default core : #0 (KOGE 3.1 Scalar) ...
Fastest core : #0 (KOGE 3.1 Scalar) ...
[Jan 24 20:02:04 UTC] RC5-72: using core #0 (MH 2-pipe).
[Jan 24 20:02:24 UTC] RC5-72: Benchmark for core #0 (MH 2 ...
0.00:00:16.54 [4,902,850 keys/sec]
[Jan 24 20:02:24 UTC] RC5-72: using core #1 (KKS 2-pipe).
[Jan 24 20:02:43 UTC] RC5-72: Benchmark for core #1 (KKS ...
0.00:00:16.37 [4,774,123 keys/sec]
[Jan 24 20:02:43 UTC] RC5-72: using core #2 (KKS 604e).
[Jan 24 20:03:03 UTC] RC5-72: Benchmark for core #2 (KKS ...
0.00:00:16.84 [3,854,975 keys/sec]
[Jan 24 20:03:03 UTC] RC5-72: using core #5 (MH 1-pipe).
[Jan 24 20:03:22 UTC] RC5-72: Benchmark for core #5 (MH 1 ...
0.00:00:17.07 [5,399,288 keys/sec]
[Jan 24 20:03:22 UTC] RC5-72: using core #6 (MH 1-pipe 604e).
[Jan 24 20:03:42 UTC] RC5-72: Benchmark for core #6 (MH 1 ...
0.00:00:17.06 [5,631,761 keys/sec]
[Jan 24 20:03:42 UTC] RC5-72 benchmark summary :
Default core : #6 (MH 1-pipe 604e) ...
Fastest core : #6 (MH 1-pipe 604e) ...




EXTRA

THREADRIPPER SAY HELLO TO A-EON :)

hellooooooooooo

Last edited by tlosm on 24-Jan-2018 at 10:29 PM.

_________________
I love Amiga and new hope by AmigaNG
A 500 + ; CDTV; CD32;
PowerMac G5 Quad 8GB,SSD,SSHD,7800gtx,Radeon R5 230 2GB;
MacBook Pro Retina I7 2.3ghz;
#nomorea-eoninmyhome

@recedent

Guess it's good he didn't have it anymore then now back on topic

_________________
____________________________
c64-2sids, A1000, A1200T-060@50(finally working!),A4000-CSMKIII
! My Master Miggies- Amiga 1000 & AmigaOne X1000 !
mancave-ramblings
X1000 I BELIEVE

Nobody gives new results, so I have to work it out on my own. Some benchmarks from a 1,33 GHz 12" iBook G4 that I've just bought:

===

aFlopsWOS:

Amiga Flops Benchmark (Double Precision), V2.0

Module Error RunTime MFLOPS
(usec)
1 4.0146e-13 0.0453 308.9655
2 -1.4166e-13 0.0300 233.6375
3 4.7184e-14 0.0338 503.1214
4 -1.2546e-13 0.0500 300.2346
5 -1.3800e-13 0.0807 359.5157
6 3.2374e-13 0.0538 538.7518
7 -8.4583e-11 0.1191 100.7874
8 3.4855e-13 0.0577 520.3252

Iterations = 512000000
NullTime (usec) = 0.0015
MFLOPS(1) = 283.2340
MFLOPS(2) = 205.4596
MFLOPS(3) = 331.6121
MFLOPS(4) = 466.1064

===

MPlayer

MPlayer SVN-r37401 (C) 2000-2015 MPlayer Team
211 audio & 442 video codecs

Playing RAM:Fashion_DivX720p_ASP.divx.
libavformat version 56.33.101 (internal)
AVI file format detected.
[aviheader] Video stream found, -vid 0
[aviheader] Audio stream found, -aid 1
VIDEO: [DX50] 1280x720 24bpp 23.976 fps 8604.2 kbps (1050.3 kbyte/s)
Load subtitles in RAM:
==========================================================================
Opening video decoder: [ffmpeg] FFmpeg's libavcodec codec family
libavcodec version 56.39.100 (internal)
[mpeg4 @ 0x23df8270]Warning: not compiled with thread support, using thread emulation
Selected video codec: [ffodivx] vfm: ffmpeg (FFmpeg MPEG-4)
==========================================================================
Audio: no sound
Starting playback...
Movie-Aspect is undefined - no prescaling applied.
VO: [null] 1280x720 => 1280x720 Planar YV12
Movie-Aspect is 1.78:1 - prescaling to correct movie aspect.
VO: [null] 1280x720 => 1280x720 Planar YV12
[VD_FFMPEG] DRI failure.


BENCHMARKs: VC: 28.152s VO: 0.017s A: 0.000s Sys: 0.898s = 29.067s
BENCHMARK%: VC: 96.8516% VO: 0.0573% A: 0.0000% Sys: 3.0911% = 100.0000%

Exiting... (End of file)

===

more to come.

Hi ALL!

Why new A1222 user dont participate here to see how the new system performs?
Also just a test with the well supported app form the TAbor was ok, just to see the better performace we can get from that board...
but also the worst an be usefull if you want to particiapte!


Thank you!

_________________
Simone"Tuxedo"Monsignori, Perugia, ITALY.