@KimmoK

Not really, if we look at the past, you can find that AmigaOS4 altivec support was added long before Morphos got it.

_________________
Cuida tus piedras gordas.

A1200/030 32Mb
A4000D
A1-X1000.

@Ami603
Quote:

Not really, if we look at the past, you can find that AmigaOS4 altivec support was added long before Morphos got it.

Actually, AltiVec support is also in there for a long time in MorphOS (long long before 2.0 was released), and it was made available for quite a few developers, to let them optimize their applications. AltiVec code was already there in some key MorphOS applications released during 1.4.x times, just sitting inactive, until 2.0. So people upgrading to the new system could immediately take advantage of it, without upgrading the applications or waiting for new versions to be released.

_________________
MorphOS, classic Amiga, demoscene, and stuff
"When a bridge is not enough, build a Viaduct!"
"Strip the Amiga community of speculation and we can fit every forum on a 720k floppy" (by resle)

@AP

Well, i didn't claim i was benchmarking the OS speed itself. Only real world applications.

To benchmark the os itself, you'd have to do unitary tests (memory allocation, disk/filesystem tests, 68k emu performance, ...). But this kind of tests has to be done with care, else you can easily misinterpret an apparent difference of performance.

That said, that Q3 test actually benchmarks the OS 3D driver performance here. The MorphOS Q3 port is indeed untouched on the OpenGL part, and i'd assume it's the same for the OS4 port. So what's really measured is the underlying OS component there, which is mainly the opengl layer. There's a reason why Q3 is often used as a benchmark.

Now, i have another "OS" test which comes to my mind, but it's not too fair:
Consider a golded:bak/ Golded backup directory containing 4000 files (SFS).

On OS4 Workbench, displaying the 4000 files in list mode took almost 30 seconds and needed 200MB RAM (and there's a refresh bug with so many files when selecting files, btw).
On MorphOS, listing the same directory took about 2 seconds and needed 3MB RAM.

But I perfectly know Workbench hasn't been really improved in that area since 3.x ages, so that's a cheap shot. :)

Last edited by Fab on 16-Feb-2009 at 06:48 PM.

@Fab

Thanks for tests!


Can you compare Quake II for both systems? I think that Q2 for OS4 is more optimised than current port of Q3.

How fast is DvPlayer?

@Fab
Many thanx for your tests in real world applications.
And apart from speed tests I think that other winning areas for MorphOS from an user pov would be comparing Ambient vs Workbench and MuiShell vs OS4 shell. These are another 2 easy shots...
Ciao.

_________________
Excuse me in advance for my very poor english.
----------------------------------------------------------------------
PowerBook G4 15" 1,67ghz running OSX (for now), MorphOS (when it's done!).

Clearly, MOS team should port Ambient to AOS4.x.

_________________
- KimmoK
// For freedom, for honor, for AMIGA
//
// Thing that I should find more time for: CC64 - 64bit Community Computer?

@KimmoK
But...
if you port Ambient to OS4 then we have OS4=MorphOS. AARGHH this IS evil! 666!

_________________
Excuse me in advance for my very poor english.
----------------------------------------------------------------------
PowerBook G4 15" 1,67ghz running OSX (for now), MorphOS (when it's done!).

@Fab

Quote:

Fab wrote: ... That said, that Q3 test actually benchmarks the OS 3D driver performance here. The MorphOS Q3 port is indeed untouched on the OpenGL part, and i'd assume it's the same for the OS4 port. So what's really measured is the underlying OS component there, which is mainly the opengl layer. There's a reason why Q3 is often used as a benchmark. Now, i have another "OS" test which comes to my mind, but it's not too fair: Consider a golded:bak/ Golded backup directory containing 4000 files (SFS). On OS4 Workbench, displaying the 4000 files in list mode took almost 30 seconds and needed 200MB RAM (and there's a refresh bug with so many files when selecting files, btw). On MorphOS, listing the same directory took about 2 seconds and needed 3MB RAM. But I perfectly know Workbench hasn't been really improved in that area since 3.x ages, so that's a cheap shot. :)

Those are two parts of the OS that are earmarked for replacement. We know that they're not as good as they should be. Do you have any tests for other OS components?

Hans

_________________
http://hdrlab.org.nz/ - Amiga OS 4 projects, programming articles and more. Home of the RadeonHD driver for Amiga OS 4.x project.
https://keasigmadelta.com/ - More of my work.

@Hans

filesystem test, both on Peg2? MOS has an old SFS IIRC, vs 'newest' os4 sfs + the other one?

_________________
Are we not done with the same silly arguments and flames yet??!

@Fab

Quote:

Now, i have another "OS" test which comes to my mind, but it's not too fair: Consider a golded:bak/ Golded backup directory containing 4000 files (SFS). On OS4 Workbench, displaying the 4000 files in list mode took almost 30 seconds and needed 200MB RAM (and there's a refresh bug with so many files when selecting files, btw). On MorphOS, listing the same directory took about 2 seconds and needed 3MB RAM.

Well Workbech identify files by opening etch file and look inside to check what type of files they are, also I get the impression that Workbench keeps a copy of etch icon even if they are the same for two files or more files, when the icon is 32bit then that takes a lot of memory.

_________________
http://lifeofliveforit.blogspot.no/
Facebook::LiveForIt Software for AmigaOS

@NutsAboutAmiga

"Well Workbech identify files by opening etch file and look inside to check what type of files they are"

Ambient does exacly the same.

@kiero

From the videos I have seen MorphOS uses different technique for cleaning up icons, I believe it uses a fixed matrix x,y number of pixels between etc icon, whiles AmigaOS4 tries to fit icons depending on icon width and height, and filename lengths in pixels.

The different types of display techniques might explain some of slowdowns.

When I have lots of files in directory I normal display content as list, to speed things up, or use Directory Opus 4, to copy move files.

I like the slow way AmigaOS does it when I have where few icons (20 icons) and like organize things, but on directories whit lots of files, it can become over complicated.

I have also notices that Scalos for OS4.x (the workbench replacement) is a lot faster then then Workbench at displaying lots of files and preview images,

_________________
http://lifeofliveforit.blogspot.no/
Facebook::LiveForIt Software for AmigaOS

@NutsAboutAmiga

"From the videos I have seen MorphOS uses different technique for cleaning up icons, I believe it uses a fixed matrix x,y number of pixels between etc icon, whiles AmigaOS4 tries to fit icons depending on icon width and height, and filename lengths in pixels."

Fab's test was about list mode, not icon mode. Icon view is obviously bit slower.

"The different types of display techniques might explain some of slowdowns."

There is no complicated layout in list mode. maybe he could do same test in icon mode and post timing/memory results.

What about some more benchmarks using MorphOS/OS4Emu and OS4 ?

This way there wouldn't be any excuse "the application isn't the same", etc...

_________________
http://www.warpdesign.fr/

@pavlor

Quote:

How fast is DvPlayer?

I discovered a few interesting things while testing DvPlayer on OS4-Pegasos2. Playback actually turned out to be slower than on my microA1, and after some investigations it turned out to be caused by two reasons:

1. The version of peg2ide.device which is included on the ISO contains a bug that it apparently falls back to PIO mode with transfers larger than 32k. This affects a lot of applications because pretty much everything on OS4 uses 64k buffers, even C:Copy or Workbench copy. The bug is already fixed and the fix will be available in an update.

2. The copy routine I used in DvPlayer to transfer the frames to videomemory was reaching the limitations of the AGP bus on the AmigaOne, however on the Pegasos2 a different technique is needed to reach the maximum speed. Using a different routine gave me a significant speed increase.

I think there are probably quite a few OS components (such as P96) which could be improved to give the best performance also on the Pegasos2, not only on the AmigaOne, but that takes time. The same goes for the Sam440.

Anyway, currently on my system when I play DVDs or VOB files with DvPlayer, CPU usage is typically between 40-45%. I haven't had time yet to install MorphOS so I don't know how fast DVD playback is there.

I managed to get a few more tests done:

GLQuake
Quake2
DVPlayer (running on OS4emu)

I edited the first benchmark post to show the test conditions and results.

@Fab

Quote:

I managed to get a few more tests done: ... CUT

Great tests... it seems MorphOS is the winner today.

Surely, OS4.x is very good but, obviously, is younger and these results were somewhat expected.


PS - Now stop to lose time with these nice but trivial things (and the related bragging... ) and hurry to help finishing the Mac mini port !

_________________
-------------------------------
AmigaOS, the last hope...

@Fab

Thanks.

@COBRA: Maybe this explains some of the test results. It seems that AmigaOS4 on PegasosII is not in the same way optimized like the versions for AmigaOne (or maybe SAM).

_________________
AmigaOne X5000/40, 2.2 Ghz, 4 GB RAM, Radeon R9 280X, M-Audio Revolution 5.1, 240 GB SSD

@Fab

It's a pity you can't add Linux MPlayer as comparison

_________________
The only spanish amiga news web page/club: CUAZ