@Hans

It still would not make ram disk or 3D run faster

Quote:

It would also be nice to see tests that don't involve OpenGL (we know the issues there) and/or old compiles of games/apps that don't take advantage of new features in Amiga OS 4.1.

Any ideas?

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

@maurensen

Even on OSX and FreeBSD you can use KDE but nobody will tell you are a Linux clone :)

Quote:

excuse me Shinky, I don't want to seem harsh, but why reinvent the wheel?

So true... and has been true since 2001/2 where OS4 wasn't event developped... Question would have been usefull back then...

_________________
http://www.warpdesign.fr/

@Leo

Still not tired of repeating the same things over and over ?

@itix

Quote:

itix wrote: @Hans It still would not make ram disk or 3D run faster

How the heck are we supposed to find performance bottlenecks without eliminating other known sources of slowdown? The IDE bug is known; it's fixed; so let's remove it from the test results. It's called isolating the problem. Why is this so hard to for you guys to understand? Or are you still gloating?

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.

@Seiya

Quote:

Seiya wrote: Quote: wegster wrote: @Seiya This looks like total 'selective data.' Plus: 1. A1/OS4 wasn't tested except for a small handful of cases. WHY? 2. No links to source of the programs used, nor params, are mentioned. 3. No system configuration data is given (clean boot, no WBStartup, vs 'any old apps running,' etc ) 4. dnetc certainly runs on A1/OS4, and with JIT - why nothing there? 5. Some of your tests are obviously done with JIT disabled. (or not done at all) Why? Incomplete data with benchmarks or only 'selective results' really isn't of much *real* use. Can you answer the above or clarify? yes :) this test is made some years ago when i have asked to italian amiga community to make some benchmark within AmigaOne and Pegasos. And then i have asked to make a 68k benchmark on OS4, MorphOS to compare with Amithlon. the benchmark file i used is here. A1/OS4 wasn't full tested because in italy for my experience, not all amiga users like very much made a benchmark. you have to consider these test like a first OS4 and MOS benchmark when OS4 was born and the first test against MorphOS. we say that this bench is an amithlon bench vesersu A1 and MOS, only few users has acepted to make these test and i thanks again these people benchmarking with 68k test againt amithlon has bring to us a first duel within OS4 and MOS with they limit siince that time. after, when OS4 and MOS was more mature, i hope to have more modern test, but amiga community has a strange relationship with benchmark. The tests also try that Amitlhon was much faster than WinUAE and much faster (on 68k application) than OS4 and MOS. i like to see now, a 68k bench with Sam440 and Efika wth my test. it will be very interesting for me to see P2@350 with amithlon versus Sam and Efika.

Thanks for the honest clarifications, things were looking a bit...fishy there, but that helps.

Anyone with *current* OS4 and MOS care to duplicate his tests? (or do you?)

As it is....this really isn't representative of much 'today,' but I do appreciate your clarifications and links, thanks.

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

@ShInKurO

Quote:

ShInKurO wrote: @Mr_Capehill Quote: Mr_Capehill wrote: @ShInKurO So have you already analyzed how much changes are needed and how horrible patch would be the outcome? eg. how much MUI4 code, or other MOS-only stuff (Reggae?) is used? IMHO it would painful to start with the very first Ambient and then trying to update it to the latest. I prefer to begin from old sources, in which we should be add OS4/OS3/AROS changes to hook macros and other OS4/OS3/AROS stuff to compile it on OS4/OS3/AROS, and after try to merge this old sources with new one with svn to see where are changes and eventually to adapt them to MUI3.8+. Reggae is not very problem because if I remember right there are some conditional code to make Ambient free from Reggae... in any case Morphos team is always avaible to replies, in contrast to what other people told about them... Quote: I don't know anything about Ambient but I'm just a little sceptical here. Open source and Amiga rarely match because there are so few developers and the ones left are occupied by other projects. The point is this: if we could take Ambient and make it executable on OS4, OS3 and AROS I hope a good group of people will become involved to project of open source desktop for AmigaOSes... we just only to organize ourself... programs like Yam didn't born portable, but with work of few people they became what they are now. In contrast with these old programs, Ambient sources are much more modern and modular, so they are simpler to understand than these other ones... The real problem is into amigans mind: they have some beliefs and obsolete thoughts which make them mindclose...

I agree with the sentiment. Personally, I see nothing wrong with Ambient as a 'workbench replacement,' for those that might prefer that. After all, MUI itself was an add on or 'replacement' of sorts itself, no? On my *nix systems, I've run anything from TWM, the various rebrands of CDE, fvwm, windowmaker, AfterStep...you name it. There is nothing wrong with 'choice,' especially if in some cases, it's of benefit to *all* in some manner. I'd also like to see MUI4 available on OS4.

Having said that, I tend to agree with Capehill on this - I expect it to be a not so insignificant effort, especially with Itix's comments, let alone then finding a way to backport from an ancient version of Ambient to current. Not that it can't be done, or that I wouldn't try it (I would, and do have a Peg2/MOS as well as my A1), just that it's unlikely to be as relatively easy as it seems you think.

Then again, while there may still be a few(thankfully only a few) vocal err, < insert some filtered word of choice here >, on both 'sides,' still 'fighting' over anything they can, or trying to use this thread as nothing more than 'I told you so' etc..I also have never seen a whole lot of harm in OS4Emu, although I'd prefer a common API layer across all Amiga-like platforms, at least for porting purposes. That should surely be easier than the NPR for FireFox, etc...

*shrug* Good luck with the Ambient port, though

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

@Pecosbil

first, winuae don't use multicore so he has a 2,4 Ghz cpu, you 4 Ghz.
my test on Athlon with 2 Ghz.

Amithlon is much faster than winuae, so a 2 Ghz cpu on Amitlhon is much faster than 2 Ghz on WinUAE.

If you try amithlon on 4 Ghz cpu, you will have more key/s :)

my test is an amithlon bench, but include some test A1 vs Peg1/2 on real test 68k and PPC.

_________________

@Hans

Quote:

Any chance of seeing it updated? Where is chip these days anyway?

He's still around, just he's been less active lately.

all you failed to see that in seiya's test theres also native ppc benchmarks besides the 68k ones and the ppc results beat everything?

@ShInKurO

Quote:

@maurensen Even on OSX and FreeBSD you can use KDE but nobody will tell you are a Linux clone :)

touchè

Last edited by maurensen on 21-Feb-2009 at 01:06 PM.

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

Just to know: what version of Ambient are we talking about? Is v2 opensource too? Because on Sourceforge i canf find v1.4 only.

Edit: just to make things clearer: i don't know if Ambient's version numbering follows MOS' one and if it should be at v2 too
Btw, yesterday i looked on Sourceforge and the most recent nightly build was really old.

Last edited by Nibunnoichi on 21-Feb-2009 at 04:18 PM.

_________________
Proud Amigan since 1987
Owner of various Commodore and a SAM440ep\OS4.1FE
See them on http://retro.furinkan.org/

@Nibunnoichi

builds may be old, but what matters is what on CVS.

@Hans

Quote:

How the heck are we supposed to find performance bottlenecks without eliminating other known sources of slowdown? The IDE bug is known; it's fixed; so let's remove it from the test results. It's called isolating the problem. Why is this so hard to for you guys to understand? Or are you still gloating?

This discussion is about benchmarks. But again:

Quote:

It would also be nice to see tests that don't involve OpenGL (we know the issues there) and/or old compiles of games/apps that don't take advantage of new features in Amiga OS 4.1.

I am interested in benchmarks where OS4 could have advantage over MorphOS. While they may or may not suffer from the IDE bug OS4.1 could still outperform MorphOS on the same hardware.

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

@Hans

>The IDE bug is known; it's fixed; so let's remove it from the test results. It's called >isolating the problem. Why is this so hard to for you guys to understand? Or are you >still gloating?

if you think it can give a problem, that the bench only need read 800 kb from harddrive, then the maxtransfer size of filesystem maybe can change to avoid the switch toPIO mode.

On my old classic amiga and Quantum Fireball hd-drive i need set max transfer to 32 kb because on larger values give errors.there is no slowdown see, but of course today drives are faster.

on modern systems when set max transfer to 32 kb give near no slowdown.

On winuae on AMD64 3000+, i reach with diskspeed a blocksize of 16 kb 99% of tranfers speed my hd reach with blocksize of 256 kb.ca 80 mb.

Last edited by bernd_afa on 23-Feb-2009 at 06:26 PM.
Last edited by bernd_afa on 22-Feb-2009 at 04:48 PM.

@all
Just bring up that old thread because of os4.1 update 1.

So, right now i checked Quake3 (that capehill's port). The results
almost the same as for Fab before (even, it looks like it's now a bit slower on 1-2 fps). So, the results:

640x480 window mode / on 1440x960x32bit screen: 21.5fps
640x480 window mode / on 1440x960x16bit screen: 24.5fps
640x480 full-screen: 23 fps

Thats all on the aos4.1 update1 / peg2 /1ghz/1ghz /radeon9250 and all tests with:
demotime 1
demo four

Last edited by kas1e on 21-Jan-2010 at 11:57 AM.

_________________
Join us to improve dopus5!
zerohero's mirror of os4/os3 crosscompiler suites

@kas1e
It is the game that must be updated not the OS, NeoGeo games still play like crap on xMame on SAM440, but just load them in gnGeo and they blast through at silky smooth 60FPS no frame-skipping.
Same HW, different performance due to different software (not a different OS).

_________________
SamFlex Complete 800Mhz System + AmigaOS 4.1 Update 4
Amiga 2000 DKB 2MB ChipRam GVP G-Force040 Picasso 2 OS3.9 BB2
AmigaCD 32

@DAX

By some reassons i am 99% sure, that is because of warp3d drivers, and not because of port. I can be wrong on that 1% of course, but we already checked other warp3d apps (i am personally) and can say, that they faster on morphos everytime in in twice. That is related to everything, old wapeout, my diskmag (where i can check count of fps myself to see how speedy rendering is happenes) and some other games and tests.

So, i am on 99% sure that is OS problems.

I know how you like os4, but let's be real, our 3d system must be updated.

_________________
Join us to improve dopus5!
zerohero's mirror of os4/os3 crosscompiler suites

@kas1e
Quote:

our 3d system must be updated

No arguing with that!

I am confident that the X1000 will bring new possibilities (such as PCI-E 16x) and that those possibilities will prompt Hyperion or people like Hans De Ruiter (which is already working at drivers) to update the 3D "back-bone" of AmigaOS.

_________________
SamFlex Complete 800Mhz System + AmigaOS 4.1 Update 4
Amiga 2000 DKB 2MB ChipRam GVP G-Force040 Picasso 2 OS3.9 BB2
AmigaCD 32

@DAX

Yep. X1000 can be interesting :) Btw, check screen1 and screen2 of quake3 in action on my peg2 (on one 43 fps, on other one 18). That result (~25fps, it the middle one, for all the demo-play).

_________________
Join us to improve dopus5!
zerohero's mirror of os4/os3 crosscompiler suites