AmigaOS4 Petunia benchmarks

[HyperionMP]

Just wait until the entire graphics subsystem runs PPC native and then compare again.

[Lando]

>Does MorphOS 1.4 run on anything else than the
>Pegasos??

Yes it also runs on Piru's BPPC, hence how he got the benchmark results.  Duh!

[Lando]

Maybe someone could re-code the apps he's using to benchmark Petunia so that they don't render anything, just do the calculations (so we can factor out the graphics subsystem).  

As it is these Petunia benchmarks are disappointingly slow... I was expecting much more, given all the hype that has been spread about it over the last year or so. :-(

[Piru]

Just wait until the entire graphics subsystem runs PPC native and then compare again.

Excuse my ignorance but isn't the graphics subsystem JITted aswell?

And even if it wasn't:

mandel does no OS calls in innerloop. It has 4 exec calls, 2 intuition calls and 200 graphics calls.

c2ptest does no OS calls in innerloop. It has 4 exec calls, and two intuition calls. It does open graphics.library but does nothing with it.

demoeffect does no OS calls in innerloop. For CGFX it has 14 exec calls, 3 dos calls, 1 cybergraphics call, 2 intuition calls and 2 timer.device calls.

Could you please explain how graphics subsystem being 68k can have such an impact on the results?

[EDIT: Added exact number of OS calls]

[Piru]

@hnl_dk

@ Piru

MorphOS 1.4 JIT on 603/240 system is several times faster than the Petunia on 604/180.

Does MorphOS 1.4 run on anything else than the Pegasos??

Yes. It runs on CyberStorm PPC and Blizzard PPC boards. I performed the tests on my A1200+BPPC system.

If not, then why do you troll?

I don't.

[hnl_dk]

@ Piru

Thanks for the clarification :-D

About the graphics subsystem ... IIRC one of the Friedens told some time ago that the 68k system components use the interpreted emulation ... Petunia is for aplications only ... but I might be wrong ;-)

[PulsatingQuasar]

About the graphics subsystem ... IIRC one of the Friedens told some time ago that the 68k system components use the interpreted emulation ... Petunia is for aplications only ... but I might be wrong

I read that somewhere too. Didn't that have something to do with reliability?

[HyperionMP]

Edit by Kees - Please solve this matter in private.

[Piru]

Edit by Kees - Please solve this matter in private.

[HyperionMP]

Edit by Kees - Please solve this matter in private.

[Jupp3]

HyperionMP:

Scroll up, and what you see?

"AmigaOS4 Petunia benchmarks"

It's not equivalent to "Piru's BPPC"

So, do you have anything to say about the subject?

(Of course, no-one prevents you from starting other thread with that "other" subject...)

[shIva]

great work ... but i think morphos jit is faster (?)

besides all this : have you seen the stats button at the bottom of the page ? press it and wonder about the 18th :-D

[JoannaK]

BenH. I really don't know what's been going on between you two in past
(and I really don't want to hear all gross details).. I just have to
say this is quite strange place and time to do any collecting of old
debts. You both must have each others E-mail addresses (or at least
able to get one ) not to mention post regularly to various boards.
Besides this must be real old stuff cause Piru has been on
mos-development quite a while?

So why here and now? Is this just normal
evasive-truth-avoidance-routine ... :-?

[HyperionMP]

>So why here and now? Is this just normal
>evasive-truth-avoidance-routine

Afraid not, it was the knee jerk reaction of somebody who was stupid enough to trust somebody and was ripped off by the person for more than 1000 euros.

[ikir]

more than 060 50Mhz in a "poor" 604 180Mhz? And you say slow? Shut up piru.

[rachy]

Let's make some things clear:

1. In every test emulation setup time calculated into the results. It takes 1.5 secs for a simple "RTS" program. (Later this time will be gone, eg. buliding jump tables take a lot of time, but it has to do only once.) So, I could decrease the running times with this value, but I want to be as correct as possible, and that vaules wouldn't be the ones what I actually measured.

2. Emulation is highly clock-speed dependant. On a 604/233 system results were a lot better, than on 604/180 actually is. (I could have explanation for this, but it is not really interesting, rather technical.) I had just no opportunity of getting such system right now. So, measuring the speed on a higher clocked system WILL imply better results.

3. The tests are sort. This is true, but on some system these tests take AGES to run. At the beginning I had a slower machine, and the emulation was slower too, that is why I chose these tests. Now everything run better, but I don't want to change the tests, because of these are the base of comparsion of the recent results.

4. Emulation is beta, not finished yet. AOS4 is beta not finished yet. I have ideas for improvements, but right now I am about stabilization and integration.

(BTW, I don't know what is wrong with julia test, it is running just fine on AmigaOS3.x, AmigaOS4 and UAE. Except MorphOS. Where is the fault then?  :-D Just a joke, don't take too serious...)