Falcon and NemBench

Hardware, coding, music, graphic and various applications

Moderators: Mug UK, [ProToS], moondog/.tSCc., lp, Moderator Team

Post Reply
User avatar
Cyprian
10 GOTO 10
10 GOTO 10
Posts: 2189
Joined: Fri Oct 04, 2002 11:23 am
Location: Warsaw, Poland

Falcon and NemBench

Post by Cyprian »

Hi all,
Here you can find benchmarks for TT: http://www.atari-forum.com/viewtopic.php?f=27&t=17970
Could be cool to have also NemBench 2.1 benchmark for stock/overclocked Falcon. Could you please post yours Falcon's benchmark here??
Mega ST 1 / 7800 / Portfolio / Lynx II / Jaguar / TT030 / Mega STe / 800 XL / 1040 STe / Falcon030 / 65 XE / 520 STm / SM124 / SC1435
DDD HDD / AT Speed C16 / TF536 / SDrive / PAK68/3 / Lynx Multi Card / LDW Super 2000 / XCA12 / SkunkBoard / CosmosEx / SatanDisk / UltraSatan / USB Floppy Drive Emulator / Eiffel / SIO2PC / Crazy Dots / PAM Net
Hatari / Steem SSE / Aranym / Saint
http://260ste.atari.org
User avatar
Xerus
Moderator
Moderator
Posts: 1240
Joined: Fri Dec 13, 2002 9:31 pm
Location: France

Re: Falcon and NemBench

Post by Xerus »

With which resolution?
Moulinaie
Captain Atari
Captain Atari
Posts: 378
Joined: Wed Feb 01, 2012 9:34 pm

Re: Falcon and NemBench

Post by Moulinaie »

Xerus wrote:With which resolution?
That would be interesting to see how the resolution influences the results.

Guillaume.
User avatar
Cyprian
10 GOTO 10
10 GOTO 10
Posts: 2189
Joined: Fri Oct 04, 2002 11:23 am
Location: Warsaw, Poland

Re: Falcon and NemBench

Post by Cyprian »

I propose do a test in two screen resolutions:
- low memory bandwidth: Compatibility mode: ST-High;
- high memory bandwidth: TV: 80 column, interlaced, truecolor or in case of VGA: 80 column, doubleline off, 256 colors;

thanks
Mega ST 1 / 7800 / Portfolio / Lynx II / Jaguar / TT030 / Mega STe / 800 XL / 1040 STe / Falcon030 / 65 XE / 520 STm / SM124 / SC1435
DDD HDD / AT Speed C16 / TF536 / SDrive / PAK68/3 / Lynx Multi Card / LDW Super 2000 / XCA12 / SkunkBoard / CosmosEx / SatanDisk / UltraSatan / USB Floppy Drive Emulator / Eiffel / SIO2PC / Crazy Dots / PAM Net
Hatari / Steem SSE / Aranym / Saint
http://260ste.atari.org
User avatar
Xerus
Moderator
Moderator
Posts: 1240
Joined: Fri Dec 13, 2002 9:31 pm
Location: France

Re: Falcon and NemBench

Post by Xerus »

All tests were performed with the control key down (not activated CPX or ACC) and RGB monitor:

640x400x2C RGB 68030-16MHz 68882-32MHz

Code: Select all

NemBench v2.1 - precision CPU/FPU profiler.

Integer multiply (16bit)     -> 0.616 Mips (~100%)
Integer divide (16bit)       -> 0.363 Mips (~100%)
Linear (stalled) integer     -> 8.007 Mips (~100%)
Interleaved (piped) integer  -> 8.007 Mips (~100%)

Float multiply (64bit)       -> 0.540 MegaFlops (~203%)
Float divide (64bit)         -> 0.349 MegaFlops (~201%)
Linear (stalled) float       -> 0.921 MegaFlops (~172%)
Interleaved (piped) float    -> 0.916 MegaFlops (~172%)

16bit read (100% hit)        -> 7.898 MByte/sec (~100%)
16bit write (100% hit)       -> 6.211 MByte/sec (~103%)
32bit read (100% hit)        -> 15.785 MByte/sec (~100%)
32bit write (100% hit)       -> 6.887 MByte/sec (~103%)

Linear 32bit read (ST-Ram)   -> 5.475 MByte/sec (~103%)
Linear 32bit write (ST-Ram)  -> 6.660 MByte/sec (~103%)
Linear 32bit copy (ST-Ram)   -> 3.336 MByte/sec (~103%)
640x400xTC RGB 68030-16MHz 68882-32MHz

Code: Select all

NemBench v2.1 - precision CPU/FPU profiler.

Integer multiply (16bit)     -> 0.616 Mips (~100%)
Integer divide (16bit)       -> 0.363 Mips (~100%)
Linear (stalled) integer     -> 8.000 Mips (~100%)
Interleaved (piped) integer  -> 8.007 Mips (~100%)

Float multiply (64bit)       -> 0.540 MegaFlops (~203%)
Float divide (64bit)         -> 0.349 MegaFlops (~201%)
Linear (stalled) float       -> 0.920 MegaFlops (~172%)
Interleaved (piped) float    -> 0.912 MegaFlops (~171%)

16bit read (100% hit)        -> 7.898 MByte/sec (~100%)
16bit write (100% hit)       -> 4.672 MByte/sec (~77%)
32bit read (100% hit)        -> 15.760 MByte/sec (~100%)
32bit write (100% hit)       -> 5.154 MByte/sec (~77%)

Linear 32bit read (ST-Ram)   -> 4.113 MByte/sec (~77%)
Linear 32bit write (ST-Ram)  -> 4.983 MByte/sec (~77%)
Linear 32bit copy (ST-Ram)   -> 2.501 MByte/sec (~77%)
640x400x2C RGB 68060-100MHz-BUS 16MHz

Code: Select all

NemBench v2.1 - precision CPU/FPU profiler.

Integer multiply (16bit)     -> 51.200 Mips (~8352%)
Integer divide (16bit)       -> 4.571 Mips (~1262%)
Linear (stalled) integer     -> 101.135 Mips (~1270%)
Interleaved (piped) integer  -> 204.800 Mips (~2572%)

Float multiply (64bit)       -> 34.133 MegaFlops (~12880%)
Float divide (64bit)         -> 2.767 MegaFlops (~1599%)
Linear (stalled) float       -> 51.200 MegaFlops (~9606%)
Interleaved (piped) float    -> 48.188 MegaFlops (~9057%)

16bit read (100% hit)        -> 198.019 MByte/sec (~2521%)
16bit write (100% hit)       -> 200.000 MByte/sec (~3324%)
32bit read (100% hit)        -> 392.156 MByte/sec (~2498%)
32bit write (100% hit)       -> 384.615 MByte/sec (~5769%)

Linear 32bit read (ST-Ram)   -> 5.925 MByte/sec (~111%)
Linear 32bit write (ST-Ram)  -> 9.959 MByte/sec (~154%)
Linear 32bit copy (ST-Ram)   -> 3.749 MByte/sec (~116%)

Linear 32bit read (FastRAM)  -> 104.439 MByte/sec (~1965%)
Linear 32bit write (FastRAM) -> 103.206 MByte/sec (~1600%)
Linear 32bit copy (FastRAM)  -> 41.088 MByte/sec (~1272%)

Linear burst copy (ST-Ram)   -> 3.636 MByte/sec (~112%)
Linear burst copy (FastRAM)  -> 50.803 MByte/sec (~1573%)
Linear burst copy (ST->Fast) -> 5.610 MByte/sec (~173%)
Linear burst copy (Fast->ST) -> 8.856 MByte/sec (~274%)
640x400xTC RGB 68060-100MHz-BUS 16MHz

Code: Select all

NemBench v2.1 - precision CPU/FPU profiler.

Integer multiply (16bit)     -> 51.200 Mips (~8352%)
Integer divide (16bit)       -> 4.571 Mips (~1262%)
Linear (stalled) integer     -> 101.135 Mips (~1270%)
Interleaved (piped) integer  -> 204.800 Mips (~2572%)

Float multiply (64bit)       -> 34.133 MegaFlops (~12880%)
Float divide (64bit)         -> 2.767 MegaFlops (~1599%)
Linear (stalled) float       -> 48.188 MegaFlops (~9040%)
Interleaved (piped) float    -> 51.200 MegaFlops (~9624%)

16bit read (100% hit)        -> 198.019 MByte/sec (~2521%)
16bit write (100% hit)       -> 200.000 MByte/sec (~3324%)
32bit read (100% hit)        -> 392.156 MByte/sec (~2498%)
32bit write (100% hit)       -> 384.615 MByte/sec (~5769%)

Linear 32bit read (ST-Ram)   -> 4.398 MByte/sec (~82%)
Linear 32bit write (ST-Ram)  -> 7.421 MByte/sec (~115%)
Linear 32bit copy (ST-Ram)   -> 2.769 MByte/sec (~85%)

Linear 32bit read (FastRAM)  -> 104.857 MByte/sec (~1972%)
Linear 32bit write (FastRAM) -> 103.206 MByte/sec (~1600%)
Linear 32bit copy (FastRAM)  -> 41.088 MByte/sec (~1272%)

Linear burst copy (ST-Ram)   -> 2.714 MByte/sec (~84%)
Linear burst copy (FastRAM)  -> 50.803 MByte/sec (~1573%)
Linear burst copy (ST->Fast) -> 4.217 MByte/sec (~130%)
Linear burst copy (Fast->ST) -> 6.616 MByte/sec (~204%)
640x400x2C RGB 68060-100MHz-BUS 25MHz

Code: Select all

NemBench v2.1 - precision CPU/FPU profiler.

Integer multiply (16bit)     -> 51.200 Mips (~8352%)
Integer divide (16bit)       -> 4.571 Mips (~1262%)
Linear (stalled) integer     -> 101.135 Mips (~1270%)
Interleaved (piped) integer  -> 204.800 Mips (~2572%)

Float multiply (64bit)       -> 34.133 MegaFlops (~12880%)
Float divide (64bit)         -> 2.767 MegaFlops (~1599%)
Linear (stalled) float       -> 51.200 MegaFlops (~9606%)
Interleaved (piped) float    -> 48.188 MegaFlops (~9057%)

16bit read (100% hit)        -> 198.019 MByte/sec (~2521%)
16bit write (100% hit)       -> 200.000 MByte/sec (~3324%)
32bit read (100% hit)        -> 392.156 MByte/sec (~2498%)
32bit write (100% hit)       -> 384.615 MByte/sec (~5769%)

Linear 32bit read (ST-Ram)   -> 8.600 MByte/sec (~161%)
Linear 32bit write (ST-Ram)  -> 13.884 MByte/sec (~215%)
Linear 32bit copy (ST-Ram)   -> 5.363 MByte/sec (~166%)

Linear 32bit read (FastRAM)  -> 104.025 MByte/sec (~1957%)
Linear 32bit write (FastRAM) -> 103.206 MByte/sec (~1600%)
Linear 32bit copy (FastRAM)  -> 41.088 MByte/sec (~1272%)

Linear burst copy (ST-Ram)   -> 5.209 MByte/sec (~161%)
Linear burst copy (FastRAM)  -> 50.803 MByte/sec (~1573%)
Linear burst copy (ST->Fast) -> 8.131 MByte/sec (~251%)
Linear burst copy (Fast->ST) -> 12.058 MByte/sec (~373%)
640x400xTC RGB 68060-100MHz-BUS 25MHz

Code: Select all

NemBench v2.1 - precision CPU/FPU profiler.

Integer multiply (16bit)     -> 51.200 Mips (~8352%)
Integer divide (16bit)       -> 4.571 Mips (~1262%)
Linear (stalled) integer     -> 101.135 Mips (~1270%)
Interleaved (piped) integer  -> 204.800 Mips (~2572%)

Float multiply (64bit)       -> 34.133 MegaFlops (~12880%)
Float divide (64bit)         -> 2.767 MegaFlops (~1599%)
Linear (stalled) float       -> 51.200 MegaFlops (~9606%)
Interleaved (piped) float    -> 51.200 MegaFlops (~9624%)

16bit read (100% hit)        -> 198.019 MByte/sec (~2521%)
16bit write (100% hit)       -> 200.000 MByte/sec (~3324%)
32bit read (100% hit)        -> 392.156 MByte/sec (~2498%)
32bit write (100% hit)       -> 384.615 MByte/sec (~5769%)

Linear 32bit read (ST-Ram)   -> 7.162 MByte/sec (~134%)
Linear 32bit write (ST-Ram)  -> 11.681 MByte/sec (~181%)
Linear 32bit copy (ST-Ram)   -> 4.440 MByte/sec (~137%)

Linear 32bit read (FastRAM)  -> 104.439 MByte/sec (~1965%)
Linear 32bit write (FastRAM) -> 103.206 MByte/sec (~1600%)
Linear 32bit copy (FastRAM)  -> 41.088 MByte/sec (~1272%)

Linear burst copy (ST-Ram)   -> 4.386 MByte/sec (~135%)
Linear burst copy (FastRAM)  -> 50.803 MByte/sec (~1573%)
Linear burst copy (ST->Fast) -> 6.884 MByte/sec (~213%)
Linear burst copy (Fast->ST) -> 10.224 MByte/sec (~316%)
User avatar
Frank B
Atari God
Atari God
Posts: 1040
Joined: Wed Jan 04, 2006 1:28 am
Location: Glasgow

Re: Falcon and NemBench

Post by Frank B »

Cyprian wrote:I propose do a test in two screen resolutions:
- low memory bandwidth: Compatibility mode: ST-High;
- high memory bandwidth: TV: 80 column, interlaced, truecolor or in case of VGA: 80 column, doubleline off, 256 colors;

thanks
Instruction/data cache on and off to? That might be interesting.
User avatar
Xerus
Moderator
Moderator
Posts: 1240
Joined: Fri Dec 13, 2002 9:31 pm
Location: France

Re: Falcon and NemBench

Post by Xerus »

Yes, but you need the app for that and I'm too lazy to install Devpac and code it :)
User avatar
wongck
Ultimate Atarian
Ultimate Atarian
Posts: 13038
Joined: Sat May 03, 2008 2:09 pm
Location: Far East
Contact:

Re: Falcon and NemBench

Post by wongck »

I used to have this on my Falcon... FalconSX.
Suppose to be able to turn on/off CPU cache.
Is this something you're looking for?
My Stuff: FB/Falcon CT63 CTPCI ATI RTL8139 USB 512MB 30GB HDD CF HxC_SD/ TT030 68882 4+32MB 520MB Nova/ 520STFM 4MB Tos206 SCSI
Shared SCSI Bus:ScsiLink ethernet, 9GB HDD,SD-reader @ http://phsw.atari.org
My Atari stuff for sale - click here for list
User avatar
Frank B
Atari God
Atari God
Posts: 1040
Joined: Wed Jan 04, 2006 1:28 am
Location: Glasgow

Re: Falcon and NemBench

Post by Frank B »

Xerus wrote:Yes, but you need the app for that and I'm too lazy to install Devpac and code it :)
A falcon without devoack installed? Sacrilege ;)

Maybe try with overscan modes too.
User avatar
Xerus
Moderator
Moderator
Posts: 1240
Joined: Fri Dec 13, 2002 9:31 pm
Location: France

Re: Falcon and NemBench

Post by Xerus »

Frank B wrote:
Xerus wrote:Yes, but you need the app for that and I'm too lazy to install Devpac and code it :)
A falcon without devoack installed? Sacrilege ;)
Yes shame on me :)
wongck wrote:I used to have this on my Falcon... FalconSX.
Suppose to be able to turn on/off CPU cache.
Is this something you're looking for?
Almost, but this tool disables all caches at the same time, you can not choose only the instruction cache or data cache.
I tried Backward III but it's a resident program and there are some active codes in memory, the end result is not 100% accurate.

Otherwise, who wants to send its results on an unmodified Falcon unlike mine?
User avatar
qq1975b
Atari God
Atari God
Posts: 1140
Joined: Tue May 15, 2012 9:15 am
Location: Barcelona

Re: Falcon and NemBench

Post by qq1975b »

Plain Falcon 4MB with 68882+Magnum 8MB FastRAM (ST High video mode and NemBench 2.1). ALT key used (loaded from FDD and only Magnum drivers loaded):
Falcon 030 Magnum.jpg
You do not have the required permissions to view the files attached to this post.
Learning...
Paolo
Captain Atari
Captain Atari
Posts: 463
Joined: Tue Sep 28, 2004 3:43 pm
Location: Italy

Re: Falcon and NemBench

Post by Paolo »

Falcon 060+SuperVidel(+SvethLana) @95MHz, 128MB + 14MB RAM, 1024x768@16bit running MINT (all caches on, checked in mintsett.acc)

NemBench v2.1 - precision CPU/FPU profiler.

Integer multiply (16bit) -> 48.761 Mips (~7954%)
Integer divide (16bit) -> 4.302 Mips (~1188%)
Linear (stalled) integer -> 94.160 Mips (~1182%)
Interleaved (piped) integer -> 186.181 Mips (~2338%)

Float multiply (64bit) -> 31.507 MegaFlops (~11889%)
Float divide (64bit) -> 2.625 MegaFlops (~1517%)
Linear (stalled) float -> 48.188 MegaFlops (~9040%)
Interleaved (piped) float -> 45.511 MegaFlops (~8554%)

16bit read (100% hit) -> 186.915 MByte/sec (~2380%)
16bit write (100% hit) -> 185.185 MByte/sec (~3078%)
32bit read (100% hit) -> 370.370 MByte/sec (~2359%)
32bit write (100% hit) -> 357.142 MByte/sec (~5357%)

Linear 32bit read (ST-Ram) -> 8.489 MByte/sec (~159%)
Linear 32bit write (ST-Ram) -> 13.624 MByte/sec (~211%)
Linear 32bit copy (ST-Ram) -> 5.328 MByte/sec (~165%)

Linear 32bit read (FastRAM) -> 98.181 MByte/sec (~1847%)
Linear 32bit write (FastRAM) -> 97.090 MByte/sec (~1505%)
Linear 32bit copy (FastRAM) -> 38.664 MByte/sec (~1197%)

Linear burst copy (ST-Ram) -> 5.160 MByte/sec (~159%)
Linear burst copy (FastRAM) -> 48.011 MByte/sec (~1487%)
Linear burst copy (ST->Fast) -> 7.992 MByte/sec (~247%)
Linear burst copy (Fast->ST) -> 11.926 MByte/sec (~369%)

(And I get a lot of "buffer_RX error" in the process. This doesn't happen out of MiNT... I must open a thread about it...)
User avatar
Cyprian
10 GOTO 10
10 GOTO 10
Posts: 2189
Joined: Fri Oct 04, 2002 11:23 am
Location: Warsaw, Poland

Re: Falcon and NemBench

Post by Cyprian »

below figures for Petr's Hades 060/68 MHz
https://atari.joska.no/Falcon040-ml/0413.html

Code: Select all

NemBench v2.1 - precision CPU/FPU profiler.

Integer multiply (16bit) -> 35.310 Mips (~5760%)
Integer divide (16bit) -> 3.084 Mips (~851%)
Linear (stalled) integer -> 67.702 Mips (~850%)
Interleaved (piped) integer -> 136.533 Mips (~1715%)
Float multiply (64bit) -> 22.755 MegaFlops (~8586%)
Float divide (64bit) -> 1.861 MegaFlops (~1076%)
Linear (stalled) float -> 34.133 MegaFlops (~6403%)
Interleaved (piped) float -> 32.768 MegaFlops (~6159%)

16bit read (100% hit) -> 133.333 MByte/sec (~1698%)
16bit write (100% hit) -> 22.624 MByte/sec (~376%)
32bit read (100% hit) -> 266.666 MByte/sec (~1698%)
32bit write (100% hit) -> 45.454 MByte/sec (~681%)

Linear 32bit read (ST-Ram) -> 66.197 MByte/sec (~1245%)
Linear 32bit write (ST-Ram) -> 43.983 MByte/sec (~681%)
Linear 32bit copy (ST-Ram) -> 26.859 MByte/sec (~832%)
Linear burst copy (ST-Ram) -> 33.781 MByte/sec (~1046%) 
Mega ST 1 / 7800 / Portfolio / Lynx II / Jaguar / TT030 / Mega STe / 800 XL / 1040 STe / Falcon030 / 65 XE / 520 STm / SM124 / SC1435
DDD HDD / AT Speed C16 / TF536 / SDrive / PAK68/3 / Lynx Multi Card / LDW Super 2000 / XCA12 / SkunkBoard / CosmosEx / SatanDisk / UltraSatan / USB Floppy Drive Emulator / Eiffel / SIO2PC / Crazy Dots / PAM Net
Hatari / Steem SSE / Aranym / Saint
http://260ste.atari.org
User avatar
Cyprian
10 GOTO 10
10 GOTO 10
Posts: 2189
Joined: Fri Oct 04, 2002 11:23 am
Location: Warsaw, Poland

Re: Falcon and NemBench

Post by Cyprian »

Thanks to our colleague @jury, below you can find some figures for Falcon with CT63 060 94MHz and VGA video modes.


640x480x2C VGA 68060-94MHz-BUS 16MHz CT63

Code: Select all

NemBench v2.1 - precision CPU/FPU profiler.

Integer multiply (16bit)     -> 46.545 Mips (~7593%)
Integer divide (16bit)       -> 4.231 Mips (~1168%)
Linear (stalled) integer     -> 93.090 Mips (~1169%)
Interleaved (piped) integer  -> 186.181 Mips (~2338%)

Float multiply (64bit)       -> 31.507 MegaFlops (~11889%)
Float divide (64bit)         -> 2.560 MegaFlops (~1479%)
Linear (stalled) float       -> 45.511 MegaFlops (~8538%)
Interleaved (piped) float    -> 48.188 MegaFlops (~9057%)

16bit read (100% hit)        -> 183.486 MByte/sec (~2336%)
16bit write (100% hit)       -> 178.571 MByte/sec (~2968%)
32bit read (100% hit)        -> 370.370 MByte/sec (~2359%)
32bit write (100% hit)       -> 357.142 MByte/sec (~5357%)

Linear 32bit read (ST-Ram)   -> 5.606 MByte/sec (~105%)
Linear 32bit write (ST-Ram)  -> 9.609 MByte/sec (~148%)
Linear 32bit copy (ST-Ram)   -> 3.589 MByte/sec (~111%)

Linear 32bit read (FastRAM)  -> 90.707 MByte/sec (~1706%)
Linear 32bit write (FastRAM) -> 89.775 MByte/sec (~1391%)
Linear 32bit copy (FastRAM)  -> 36.008 MByte/sec (~1115%)

Linear burst copy (ST-Ram)   -> 3.504 MByte/sec (~108%)
Linear burst copy (FastRAM)  -> 44.431 MByte/sec (~1376%)
Linear burst copy (ST->Fast) -> 5.363 MByte/sec (~166%)
Linear burst copy (Fast->ST) -> 8.516 MByte/sec (~263%)

640x480x16C VGA 68060-94MHz-BUS 16MHz CT63

Code: Select all

NemBench v2.1 - precision CPU/FPU profiler.

Integer multiply (16bit)     -> 46.545 Mips (~7593%)
Integer divide (16bit)       -> 4.231 Mips (~1168%)
Linear (stalled) integer     -> 93.090 Mips (~1169%)
Interleaved (piped) integer  -> 186.181 Mips (~2338%)

Float multiply (64bit)       -> 32.768 MegaFlops (~12365%)
Float divide (64bit)         -> 2.560 MegaFlops (~1479%)
Linear (stalled) float       -> 45.511 MegaFlops (~8538%)
Interleaved (piped) float    -> 51.200 MegaFlops (~9624%)

16bit read (100% hit)        -> 183.486 MByte/sec (~2336%)
16bit write (100% hit)       -> 178.571 MByte/sec (~2968%)
32bit read (100% hit)        -> 370.370 MByte/sec (~2359%)
32bit write (100% hit)       -> 357.142 MByte/sec (~5357%)

Linear 32bit read (ST-Ram)   -> 4.787 MByte/sec (~90%)
Linear 32bit write (ST-Ram)  -> 8.181 MByte/sec (~126%)
Linear 32bit copy (ST-Ram)   -> 3.031 MByte/sec (~93%)

Linear 32bit read (FastRAM)  -> 91.022 MByte/sec (~1712%)
Linear 32bit write (FastRAM) -> 89.775 MByte/sec (~1391%)
Linear 32bit copy (FastRAM)  -> 36.008 MByte/sec (~1115%)

Linear burst copy (ST-Ram)   -> 2.989 MByte/sec (~92%)
Linear burst copy (FastRAM)  -> 44.431 MByte/sec (~1376%)
Linear burst copy (ST->Fast) -> 4.576 MByte/sec (~141%)
Linear burst copy (Fast->ST) -> 7.261 MByte/sec (~224%)

640x480x256C VGA 68060-94MHz-BUS 16MHz CT63

Code: Select all

NemBench v2.1 - precision CPU/FPU profiler.

Integer multiply (16bit)     -> 46.545 Mips (~7593%)
Integer divide (16bit)       -> 4.231 Mips (~1168%)
Linear (stalled) integer     -> 93.090 Mips (~1169%)
Interleaved (piped) integer  -> 186.181 Mips (~2338%)

Float multiply (64bit)       -> 32.768 MegaFlops (~12365%)
Float divide (64bit)         -> 2.560 MegaFlops (~1479%)
Linear (stalled) float       -> 48.188 MegaFlops (~9040%)
Interleaved (piped) float    -> 48.188 MegaFlops (~9057%)

16bit read (100% hit)        -> 183.486 MByte/sec (~2336%)
16bit write (100% hit)       -> 178.571 MByte/sec (~2968%)
32bit read (100% hit)        -> 370.370 MByte/sec (~2359%)
32bit write (100% hit)       -> 357.142 MByte/sec (~5357%)

Linear 32bit read (ST-Ram)   -> 3.708 MByte/sec (~69%)
Linear 32bit write (ST-Ram)  -> 6.206 MByte/sec (~96%)
Linear 32bit copy (ST-Ram)   -> 2.343 MByte/sec (~72%)

Linear 32bit read (FastRAM)  -> 91.022 MByte/sec (~1712%)
Linear 32bit write (FastRAM) -> 89.468 MByte/sec (~1387%)
Linear 32bit copy (FastRAM)  -> 36.008 MByte/sec (~1115%)

Linear burst copy (ST-Ram)   -> 2.272 MByte/sec (~70%)
Linear burst copy (FastRAM)  -> 44.431 MByte/sec (~1376%)
Linear burst copy (ST->Fast) -> 3.504 MByte/sec (~108%)
Linear burst copy (Fast->ST) -> 5.551 MByte/sec (~171%)


320x240x16C VGA 68060-94MHz-BUS 16MHz CT63

Code: Select all

NemBench v2.1 - precision CPU/FPU profiler.

Integer multiply (16bit)     -> 46.545 Mips (~7593%)
Integer divide (16bit)       -> 4.231 Mips (~1168%)
Linear (stalled) integer     -> 93.090 Mips (~1169%)
Interleaved (piped) integer  -> 186.181 Mips (~2338%)

Float multiply (64bit)       -> 32.768 MegaFlops (~12365%)
Float divide (64bit)         -> 2.560 MegaFlops (~1479%)
Linear (stalled) float       -> 45.511 MegaFlops (~8538%)
Interleaved (piped) float    -> 45.511 MegaFlops (~8554%)

16bit read (100% hit)        -> 183.486 MByte/sec (~2336%)
16bit write (100% hit)       -> 178.571 MByte/sec (~2968%)
32bit read (100% hit)        -> 370.370 MByte/sec (~2359%)
32bit write (100% hit)       -> 357.142 MByte/sec (~5357%)

Linear 32bit read (ST-Ram)   -> 5.332 MByte/sec (~100%)
Linear 32bit write (ST-Ram)  -> 9.127 MByte/sec (~141%)
Linear 32bit copy (ST-Ram)   -> 3.413 MByte/sec (~105%)

Linear 32bit read (FastRAM)  -> 91.339 MByte/sec (~1718%)
Linear 32bit write (FastRAM) -> 89.775 MByte/sec (~1391%)
Linear 32bit copy (FastRAM)  -> 36.008 MByte/sec (~1115%)

Linear burst copy (ST-Ram)   -> 3.330 MByte/sec (~103%)
Linear burst copy (FastRAM)  -> 44.431 MByte/sec (~1376%)
Linear burst copy (ST->Fast) -> 5.096 MByte/sec (~157%)
Linear burst copy (Fast->ST) -> 8.090 MByte/sec (~250%)

320x240x256C VGA 68060-94MHz-BUS 16MHz CT63

Code: Select all

NemBench v2.1 - precision CPU/FPU profiler.

Integer multiply (16bit)     -> 46.545 Mips (~7593%)
Integer divide (16bit)       -> 4.231 Mips (~1168%)
Linear (stalled) integer     -> 93.090 Mips (~1169%)
Interleaved (piped) integer  -> 186.181 Mips (~2338%)

Float multiply (64bit)       -> 32.768 MegaFlops (~12365%)
Float divide (64bit)         -> 2.560 MegaFlops (~1479%)
Linear (stalled) float       -> 45.511 MegaFlops (~8538%)
Interleaved (piped) float    -> 48.188 MegaFlops (~9057%)

16bit read (100% hit)        -> 183.486 MByte/sec (~2336%)
16bit write (100% hit)       -> 178.571 MByte/sec (~2968%)
32bit read (100% hit)        -> 370.370 MByte/sec (~2359%)
32bit write (100% hit)       -> 357.142 MByte/sec (~5357%)

Linear 32bit read (ST-Ram)   -> 4.858 MByte/sec (~91%)
Linear 32bit write (ST-Ram)  -> 8.181 MByte/sec (~126%)
Linear 32bit copy (ST-Ram)   -> 3.088 MByte/sec (~95%)

Linear 32bit read (FastRAM)  -> 91.339 MByte/sec (~1718%)
Linear 32bit write (FastRAM) -> 89.775 MByte/sec (~1391%)
Linear 32bit copy (FastRAM)  -> 36.008 MByte/sec (~1115%)

Linear burst copy (ST-Ram)   -> 2.989 MByte/sec (~92%)
Linear burst copy (FastRAM)  -> 44.431 MByte/sec (~1376%)
Linear burst copy (ST->Fast) -> 4.576 MByte/sec (~141%)
Linear burst copy (Fast->ST) -> 7.261 MByte/sec (~224%)

320x240xTC VGA 68060-94MHz-BUS 16MHz CT63

Code: Select all

NemBench v2.1 - precision CPU/FPU profiler.

Integer multiply (16bit)     -> 46.545 Mips (~7593%)
Integer divide (16bit)       -> 4.196 Mips (~1159%)
Linear (stalled) integer     -> 93.090 Mips (~1169%)
Interleaved (piped) integer  -> 186.181 Mips (~2338%)

Float multiply (64bit)       -> 31.507 MegaFlops (~11889%)
Float divide (64bit)         -> 2.560 MegaFlops (~1479%)
Linear (stalled) float       -> 45.511 MegaFlops (~8538%)
Interleaved (piped) float    -> 48.188 MegaFlops (~9057%)

16bit read (100% hit)        -> 183.486 MByte/sec (~2336%)
16bit write (100% hit)       -> 178.571 MByte/sec (~2968%)
32bit read (100% hit)        -> 370.370 MByte/sec (~2359%)
32bit write (100% hit)       -> 357.142 MByte/sec (~5357%)

Linear 32bit read (ST-Ram)   -> 3.706 MByte/sec (~69%)
Linear 32bit write (ST-Ram)  -> 6.206 MByte/sec (~96%)
Linear 32bit copy (ST-Ram)   -> 2.338 MByte/sec (~72%)

Linear 32bit read (FastRAM)  -> 91.022 MByte/sec (~1712%)
Linear 32bit write (FastRAM) -> 89.468 MByte/sec (~1387%)
Linear 32bit copy (FastRAM)  -> 36.008 MByte/sec (~1115%)

Linear burst copy (ST-Ram)   -> 2.272 MByte/sec (~70%)
Linear burst copy (FastRAM)  -> 44.431 MByte/sec (~1376%)
Linear burst copy (ST->Fast) -> 3.504 MByte/sec (~108%)
Linear burst copy (Fast->ST) -> 5.558 MByte/sec (~172%)


320x480x16C VGA 68060-94MHz-BUS 16MHz CT63

Code: Select all

NemBench v2.1 - precision CPU/FPU profiler.

Integer multiply (16bit)     -> 46.545 Mips (~7593%)
Integer divide (16bit)       -> 4.231 Mips (~1168%)
Linear (stalled) integer     -> 93.090 Mips (~1169%)
Interleaved (piped) integer  -> 186.181 Mips (~2338%)

Float multiply (64bit)       -> 32.768 MegaFlops (~12365%)
Float divide (64bit)         -> 2.560 MegaFlops (~1479%)
Linear (stalled) float       -> 45.511 MegaFlops (~8538%)
Interleaved (piped) float    -> 45.511 MegaFlops (~8554%)

16bit read (100% hit)        -> 183.486 MByte/sec (~2336%)
16bit write (100% hit)       -> 178.571 MByte/sec (~2968%)
32bit read (100% hit)        -> 370.370 MByte/sec (~2359%)
32bit write (100% hit)       -> 357.142 MByte/sec (~5357%)

Linear 32bit read (ST-Ram)   -> 5.332 MByte/sec (~100%)
Linear 32bit write (ST-Ram)  -> 9.127 MByte/sec (~141%)
Linear 32bit copy (ST-Ram)   -> 3.445 MByte/sec (~106%)

Linear 32bit read (FastRAM)  -> 91.022 MByte/sec (~1712%)
Linear 32bit write (FastRAM) -> 89.468 MByte/sec (~1387%)
Linear 32bit copy (FastRAM)  -> 36.008 MByte/sec (~1115%)

Linear burst copy (ST-Ram)   -> 3.330 MByte/sec (~103%)
Linear burst copy (FastRAM)  -> 44.431 MByte/sec (~1376%)
Linear burst copy (ST->Fast) -> 5.096 MByte/sec (~157%)
Linear burst copy (Fast->ST) -> 8.085 MByte/sec (~250%)

320x480x256C VGA 68060-94MHz-BUS 16MHz CT63

Code: Select all

NemBench v2.1 - precision CPU/FPU profiler.

Integer multiply (16bit)     -> 46.545 Mips (~7593%)
Integer divide (16bit)       -> 4.231 Mips (~1168%)
Linear (stalled) integer     -> 93.090 Mips (~1169%)
Interleaved (piped) integer  -> 186.181 Mips (~2338%)

Float multiply (64bit)       -> 32.768 MegaFlops (~12365%)
Float divide (64bit)         -> 2.560 MegaFlops (~1479%)
Linear (stalled) float       -> 45.511 MegaFlops (~8538%)
Interleaved (piped) float    -> 48.188 MegaFlops (~9057%)

16bit read (100% hit)        -> 183.486 MByte/sec (~2336%)
16bit write (100% hit)       -> 178.571 MByte/sec (~2968%)
32bit read (100% hit)        -> 370.370 MByte/sec (~2359%)
32bit write (100% hit)       -> 357.142 MByte/sec (~5357%)

Linear 32bit read (ST-Ram)   -> 4.825 MByte/sec (~90%)
Linear 32bit write (ST-Ram)  -> 8.181 MByte/sec (~126%)
Linear 32bit copy (ST-Ram)   -> 3.085 MByte/sec (~95%)

Linear 32bit read (FastRAM)  -> 91.022 MByte/sec (~1712%)
Linear 32bit write (FastRAM) -> 89.775 MByte/sec (~1391%)
Linear 32bit copy (FastRAM)  -> 36.008 MByte/sec (~1115%)

Linear burst copy (ST-Ram)   -> 2.989 MByte/sec (~92%)
Linear burst copy (FastRAM)  -> 44.431 MByte/sec (~1376%)
Linear burst copy (ST->Fast) -> 4.576 MByte/sec (~141%)
Linear burst copy (Fast->ST) -> 7.265 MByte/sec (~225%)

320x480xTC VGA 68060-94MHz-BUS 16MHz CT63

Code: Select all

NemBench v2.1 - precision CPU/FPU profiler.

Integer multiply (16bit)     -> 46.545 Mips (~7593%)
Integer divide (16bit)       -> 4.196 Mips (~1159%)
Linear (stalled) integer     -> 93.090 Mips (~1169%)
Interleaved (piped) integer  -> 186.181 Mips (~2338%)

Float multiply (64bit)       -> 31.507 MegaFlops (~11889%)
Float divide (64bit)         -> 2.560 MegaFlops (~1479%)
Linear (stalled) float       -> 45.511 MegaFlops (~8538%)
Interleaved (piped) float    -> 48.188 MegaFlops (~9057%)

16bit read (100% hit)        -> 183.486 MByte/sec (~2336%)
16bit write (100% hit)       -> 178.571 MByte/sec (~2968%)
32bit read (100% hit)        -> 370.370 MByte/sec (~2359%)
32bit write (100% hit)       -> 357.142 MByte/sec (~5357%)

Linear 32bit read (ST-Ram)   -> 3.677 MByte/sec (~69%)
Linear 32bit write (ST-Ram)  -> 6.206 MByte/sec (~96%)
Linear 32bit copy (ST-Ram)   -> 2.343 MByte/sec (~72%)

Linear 32bit read (FastRAM)  -> 90.394 MByte/sec (~1700%)
Linear 32bit write (FastRAM) -> 89.164 MByte/sec (~1382%)
Linear 32bit copy (FastRAM)  -> 36.008 MByte/sec (~1115%)

Linear burst copy (ST-Ram)   -> 2.272 MByte/sec (~70%)
Linear burst copy (FastRAM)  -> 44.431 MByte/sec (~1376%)
Linear burst copy (ST->Fast) -> 3.504 MByte/sec (~108%)
Linear burst copy (Fast->ST) -> 5.558 MByte/sec (~172%)
Mega ST 1 / 7800 / Portfolio / Lynx II / Jaguar / TT030 / Mega STe / 800 XL / 1040 STe / Falcon030 / 65 XE / 520 STm / SM124 / SC1435
DDD HDD / AT Speed C16 / TF536 / SDrive / PAK68/3 / Lynx Multi Card / LDW Super 2000 / XCA12 / SkunkBoard / CosmosEx / SatanDisk / UltraSatan / USB Floppy Drive Emulator / Eiffel / SIO2PC / Crazy Dots / PAM Net
Hatari / Steem SSE / Aranym / Saint
http://260ste.atari.org
Rustynutt
Atari God
Atari God
Posts: 1140
Joined: Wed Mar 21, 2012 7:38 am
Location: Oregon

Re: Falcon and NemBench

Post by Rustynutt »

Since the feature is extensively used in this thread, how do you all post your code and text inside a scrolling box?
I have a lot of test saved from my Mighty Sonic and Afterburner with Nemisis and NOVA ATI installed.
User avatar
Cyprian
10 GOTO 10
10 GOTO 10
Posts: 2189
Joined: Fri Oct 04, 2002 11:23 am
Location: Warsaw, Poland

Re: Falcon and NemBench

Post by Cyprian »

Rustynutt wrote: Fri Jan 15, 2021 7:59 pm Since the feature is extensively used in this thread, how do you all post your code and text inside a scrolling box?
I have a lot of test saved from my Mighty Sonic and Afterburner with Nemisis and NOVA ATI installed.
use "code" tag or that pointed on my screenshot icon

Code: Select all

example
tmp.png
You do not have the required permissions to view the files attached to this post.
Mega ST 1 / 7800 / Portfolio / Lynx II / Jaguar / TT030 / Mega STe / 800 XL / 1040 STe / Falcon030 / 65 XE / 520 STm / SM124 / SC1435
DDD HDD / AT Speed C16 / TF536 / SDrive / PAK68/3 / Lynx Multi Card / LDW Super 2000 / XCA12 / SkunkBoard / CosmosEx / SatanDisk / UltraSatan / USB Floppy Drive Emulator / Eiffel / SIO2PC / Crazy Dots / PAM Net
Hatari / Steem SSE / Aranym / Saint
http://260ste.atari.org
Rustynutt
Atari God
Atari God
Posts: 1140
Joined: Wed Mar 21, 2012 7:38 am
Location: Oregon

Re: Falcon and NemBench

Post by Rustynutt »

Looking back through these bench, and only at Integer multiply, the results with nembench are the same regardless of video mode. In the "Falcon 16-32 bus myths" thread where I'd u/l some old test, cyprian asked what video mode was used, and stated I thought that didn't matter with nembench. Didn't check all test though.
Should of placed those benches here, just lost where I was at on the board.
Post Reply

Return to “Professionals”