@Cyborg

I sincerly doubt there is still any "ancient almost 25 years old ####" in OS4-core parts (an old Desktop or "ed" wouldn't have any effect on these test) since it's been allmost 8 years since OS4 started. If thats not enough time to get rid of all the "ancient ###" someone has done a very poor job (which would bring us back to Krashans comment).

_________________
- We don't need good ideas, we haven't run out on bad ones yet
- blame Canada

Call me stupid, but it seems the most obvious answer is that the Peg2 port of OS4 was rushed (it was a big surprise to everyone after all), and presumably some of the HAL (drivers/etc) are not really optimised for the Peg2 yet, especially disk access going by the benchmarks. MOS has been running on Peg2 for *years* so it is hardly surprising that it runs well on that h/w!

Of course, USB is know to suck on OS4.1 , so no surprises that it was well behind there. Hopefully the (apparently) much improved USB that is in the works will shore things up a bit. And the 3D performance is also known to be non-optimal, so again no surprise that MOS beats OS4 here.

In addition to that, it is also possible that MOS has a small overall efficiency advantage over OS4 anyway (say somewhere between 1% & 10%).

Not sure about the JIT results, would be nice to see more tests on other hardware.

As to the 173Mb vs 10MB "memory usage", it seems that they are probably being fooled my OS4.1's new memory system. Measuring free memory is pretty much impossible.

Last edited by ChrisH on 11-Aug-2009 at 07:12 PM.
Last edited by ChrisH on 11-Aug-2009 at 07:07 PM.
Last edited by ChrisH on 11-Aug-2009 at 07:05 PM.

_________________
Author of the PortablE programming language.
It is pitch black. You are likely to be eaten by a grue...

You aren't probably really interested in facts

Of course I'm interested, as an AmigaOS 4 owner, user and application programmer.

OS4 has a heck of all that ancient almost 25 years old #### onboard

What for? And don't tell me about backward compatibility, it is a known fact that MorphOS is more backward compatible with classic Amiga software than AmigaOS 4 is. If AmigaOS 4 developers just haven't managed to remove this old #####, well, what can I say more?

_________________
Reggae · MorphOS Files · DigiBooster 3

@_PAB_ : no it's not. sam's oem version of os4.1 is (still...) in beta stage.

more info

Last edited by amije on 11-Aug-2009 at 07:21 PM.

_________________
A3040 AmigaOS3.9
A1200 060/BPPC AmigaOS3.9/4.0
Sam440ep AmigaOS4.1.2
Pegasos2/G3 AmigaOS4.1.2/Morphos2.7/Debian6.0/OpenSuse11.1
MacMini/G4 1.5 MorphOS 2.7/OSX 10.5.8

Long Live Amiga

@COBRA

Quote:

On the Pegasos2 anything which uses memory accesses will run faster under MorphOS because it uses a different memory setup making memory accesses on the Pegasos2 memory controller faster overall.

It still would not explain better 3D performance on MorphOS. Neither it would explain USB slowness on OS4.

On the other hand LAME is still equally fast on both operating systems.

_________________
Amiga Developer
Amiga 500, Efika, Mac Mini and PowerBook

Nice comparison.
Why 4 benchmarks for 68k JIT?

Speed difference between OS4 and MorphOS is minor (cca 16-20 %), except 3D and USB of course.

@itix

Quote:

It still would not explain better 3D performance on MorphOS. Neither it would explain USB slowness on OS4.

In case you haven't noticed, I was answering to Denil's question about why there would be a difference with the same sources (he mentioned mplayer and lha), I was not talking about 3D or USB, which are areas we all know OS4 needs to catch up on (and it's been discussed before already). The disk I/O results surprise me, because on my Peg2 things like copying files from one partition to another is like twice as fast under OS4 than MOS, in my tests this was the one thing where MOS seemed to be very left behind.

Sadly there not more hardware to run tests, i.e. MorphOS on sam440ep.

_________________
X1000|II/G4|440ep|2000/060|2000/040|1000

Quote:

Sadly there not more hardware to run tests, i.e. MorphOS on sam440ep.

You can run tests on OS4/MOS on classic PPC Amiga hardware.

I really don't understand this fascination with having to compare and benchmark all the time. Is this just a case of "mine is bigger than yours" ?

All I care about is how well the system can carry out what I want it to do, who cares if "x" is faster then "y". I mean, are we in that much of a rush these days?

_________________
Simon

Comments made by me on any public fora are not representative of, or on behalf of, any company I may have, or assumed by the reader to have, any association with.

Any comments are a personal opinion, and should be accepted as such.

COBRA

Hm well, mos has implemented tlsf.

_________________
C= Amiga addict
,,,
(Oo)
⎛☮ໄ
ﮑὠՀ
Couldn't care less what other people think, seeing that there's concrete evidence they don't.

@COBRA

I have to say I didn't notice. But still it does not explain why 68k LhA runs 2 times faster on MorphOS than PPC LhA on OS4 (nor why PPC/LhA is 3 times faster on MorphOS than PPC LhA on OS4, OS4Emu comparison could be nice).

Last edited by itix on 11-Aug-2009 at 08:17 PM.
Last edited by itix on 11-Aug-2009 at 08:06 PM.

_________________
Amiga Developer
Amiga 500, Efika, Mac Mini and PowerBook

@rigo

Well, personally, I care :) It would make a huge difference for instance if OS4 had USB2 support because right now downloading stuff from the 2GB SD card of my camera takes a considerable amount of time, time which I could be spending on sorting and touching up the photos, and I usually have little time for these things. I'm not much into 3D games and stuff (and there aren't many to play on these platforms anyway) so USB is the first thing I would like to see an improvement on under OS4.

@Krashan
Quote:

Because MorphOS has better programmers.

When you resort to cheap personal attacks like this it only reflects badly on whatever system you happen to run on any computer. Yes, attack the system itself but keep any personal problems you have to yourself next time.

_________________
ExecSG Team Lead

Dwyloc : yes, the DMA is active. See hard drive speeds, AmigaOS 4.1 is fast.

ChrisH : low disk access was fixed in the first AmigaOS 4.1 fix for Pegasos II. Copy of files/directories are even faster than in MorphOS now. And for 170 MB/10 MB memory access, you can re-do the test (with ARexx) mesure it with the command "avail" before and after the test.

The main problem is perhaps like Cobra says : the memory controler.

Pavlor : why 4 tests why 68k JIT ? Simply we need more than one exemple to be fair. For exemple, one exemple for office, one other with SDL, others with graphics, etc.

Last edited by Daff on 11-Aug-2009 at 08:35 PM.
Last edited by Daff on 11-Aug-2009 at 08:26 PM.
Last edited by Daff on 11-Aug-2009 at 08:23 PM.

@itix

Quote:

I have to say I didn't notice. But still it does not explain why 68k LhA runs 2 times faster on MorphOS than PPC LhA on OS4.

Most compression algorithms make heavy use of memory and I believe lha is no exception, so MorphOS is at a clear advantage there on Pegasos2 hardware due to the memory setup I mentioned earlier. I'm not saying though that there cannot be other factors influencing the results, but this single issue will give MorphOS a considerable advantage in almost any test.

Quote:

There's a very simple explanation, actually: memory setup.

What if the explanation was even more simple: MorphOS is better written than OS4. Period.

Can't you guys accept that ?

OS4 is slower than MorphOS: it's the hardware's fault, it was released in a hurry, it's been in development for less time, MorphOS is more optimised, blabla...

_________________
http://www.warpdesign.fr/

When you resort to cheap personal attacks

Please hold the horses. If I say that person A is better programmer than person B - it is just an opinion, but not a personal attack. I will give you an example. Do you consider "Shakira is a better singer than Madonna" statement (whether it is true or not) a cheap personal attack at Madonna? Following your "logic" a statement "Western Digital makes better harddrives than Seagate" is a cheap attack at Seagate's engineering team. BTW there are no names in my opinion in question - how can you call it "personal" then?

_________________
Reggae · MorphOS Files · DigiBooster 3

zomg, they failed to benchmark the speed of Screen Dragging! OS4 would have ruled in that (as MorphOS can't do that, unless you install a 3rd party patch)

Quote:

When you resort to cheap personal attacks like this it only reflects badly on whatever system you happen to run on any computer. Yes, attack the system itself but keep any personal problems you have to yourself next time.

you shouldn't call it personal. Krashan is a very talented and skilled programer (have a look about his past and present work) and he also codes on OS4.1. So he knows both system's internals and is skilled enough to have a judgement based on facts (the code, the way the system works) better than personal.

OK, Krashan, what about swapping your statement for "MorphOS is better writen" ?

_________________