Lightwave benchmark for users with PPC, 060, PiStorm, Vampire accelerators
Q over at Hold and Modify on YouTube just posted a Lightwave Raytrace Scene benchmark and I absolutely love this type of stuff so if you're interested please follow the link below to follow the simple steps to run the benchmark and post your results & system specs https://www.amigans.net/uploads/smil3dbd4e5e7563a.gif https://www.amigans.net/uploads/smil3dbd4e4c2e742.gif
Hold and Modify Lightwave3D Benchmark Raytrace Scene! For those that don't want to watch the whole vid here's the quick setup: 1- File-Load Scene-Benchmarks folder-choose Raytrace scene 2- Camera Panel- set to low resolution- set to square pixels 3- click Render or hit F9 Q's system results: A3000, Phase5-MKII (060@50), OS3.2.1- 52m24s A1200, TF1260 (060@50)128MB, OS3.2.1-54m22s A4000, BFG060 (rev5) 060@50, 128MB, OS3.2.1- 43m46s A2500, PP&S (040@25), OS3.3.2.1- 2h2m8s |
my testing:
A4000, CS-MKIII, 060@50 (Rev1), OS3.2.1- 47m21s A1200, TF1260, 060@50 (Rev1), OS3.2.1- 51m48s A1200, M-Tec 030@50, 32MB Fast- 5h21m2s AmigaONE X1000, PPC, OS4.1FE- 17m10s as of April2023 firmware: A1200, PiStorm32 with Raspberry Pi3a+ (March 10/23 firmware), Amiga OS3.2.2- 5m59s A1200, PiStorm32 with Raspberry Pi3a+ (Sept/23 firmware), Amiga OS3.2.2- 5m59s |
There is a similar Lightwave test here on A1K with dozens of results for different accelerators, including the new BFG9060 at 100mhz
https://www.a1k.org/forum/index.php?threads/62176/ The results spreadsheet is here https://drive.google.com/file/d/1_Vx...ew?usp=sharing |
Wow! Apollo Icedrake vs RPi4 Emu68 Pistorm32 is 38 minutes vs 6 minutes! The PiStorm32 pushes the Classic hardware above G4 chips and almost matching WinUAE speeds on fast PCs! :cool
|
@ trixster
cool stuff! |
PiStorm seems to be in the wrong section of that spreadsheet. It's in the FPGA section when it should be in the 68K software emulation section. The CPLD or FPGA on the various PiStorm boards are only there to interface between the SMI bus of the Raspberry PI and the 68K bus/control signals. The ARM SoC on the PI is running the 68K CPU JIT emulation (Emu68) and may be using other hardware resources on the PI that could help a lot with bottlenecks.
|
Quote:
Two things that I find interesting:- 1. The big difference between the Pi 3B at 1.4GHz and the Pi 4B at 2.2GHz (2.4 x faster from only 1.6 x higher clock frequency). Looks like a Pi 4 is the way to go if you want ludicrous speed! 2. The Icedrake is the fastest of all the 68k hardware solutions, beating a 100MHz 060. That puts it close to what we might have expected from the 060 if Motorola hadn't thrown in the towel early. Quote:
I am concerned about what effect this might have on the retro community. We already have 'Amiga' software being produced that needs at least an 060 simply because it's there (for a vanishing small number of users), and programmers who can't be bothered coding for less powerful machines. This is the PC way, not the Amiga way. I am more impressed by games for the A500 like Dread and Metro Siege, which show what stock hardware can do in the hands of talented programmers. |
Quote:
There IS a place for MORE 040/060/RTG programs/games to be written now because you can upgrade your A500/A1200 to run them cheaply and without the hassle of times gone by. There will always be people that want to 'bang the metal' on stock A500s and A1200s but now people can develop more demanding software and games too! |
Quote:
|
hey guys chatting is cool but also turn on your miggies and run the scene and post your results here with your system specs:great
|
Quote:
Lightwave 5.00, MorphOS 3.17 (with some Wayfarer active tabs on the background) on a PMac Sawtooth with upgraded CPU 7451 (G4) Version 32768 Revision 515, CPUClock 1.6Ghz / BusClock 100Mhz Result: 5:41 Note: That PMac 1.76Ghz listed on the table result must have had some pretty heavy background tasks and imo should either be excluded from the list or rerun the test, as its results doesn't make any sense. |
@ Cool_amigaN
Thanks :great how much Fast Ram? @ All start up your Amiga's and get rendering :spin |
Quote:
|
A1200, M-Tec 030@50, 32MB Fast- 5h21m2s
|
from YouTube thread:
AmigaONE X5000/020-PPC-OS4.1FE- 3m27s |
Quote:
|
My impression of Vampire compared to 060 was that its strength was its memory bandwidth? I thought a vampire was a bit faster in terms of raw processing power over a fast 060 but its memory is orders of magnitude faster. So for stuff like this Lightwave test which is maybe not memory intensive then you don’t see a big difference between vamp and 100mhz 060, but for games like Quake and Diablo which rely on quick memory access you get a major difference. Is that a fair comparison?
|
I find the decrying of PiStorm as a non hardware solution to be disingenuous. As far as the Amiga hardware is concerned, it's a 68K. It's the interface that matters, not the implementation. To that end, although it's not in the emulator section it's in the best fit selection. A non 68K hardware category.
Claiming the speed and unpredictable nature of JIT execution of code makes it somehow "not" an Amiga. Well the same can be said of any faster CPU with bigger caches and other tricks to improving execution throughput. You should disqualify the 68060 if you are going down that route: no microcode? More than one execution unit? Branch prediction? 8K caches? Orders of magnitude faster than a 7MHz 68000? Missing instructions needing software traps? Outrageous! |
@Karlos
yup, exactly that |
Quote:
|
All times are GMT +2. The time now is 18:35. |
Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.