Quote:

First, OWB 1.11 uses a WebKit snapshot from 5th november, so it might be more recent (and optimized?) in some parts than your Safari browser.

Also, I can't tell if Safari for OSX / PPC uses the two PPC cores, but in any case, Safari probably rather wins on the gfx render speed (check the detailed histogram in peacekeeper results). OWB MorphOS indeed uses a software cairo surface, and it can easily be beaten by anything else that is somewhat accelerated, like OSX CoreGraphics or so.

And like said previously, WebKit's JIT is only implemented for ARM and X86/X64. So OSX has no advantage there.

Fab

If there was full hw accel with Cairo in morphos would it be as fast as chromium? I really think that this project should be a huge priority for Morphos atm...

Er... Cairo hardware acceleration is not the Holy Grail. And i don't see why it would be such a "huge priority" either. It's not like it's unusable as it is. Some things would be faster with it, and some also slower, actually.

Anyway, Chrome's speed is mostly about javascript, which is unrelated to cairo. So if we want speed, better ask WebKit guys to write a PPC implementation of their javascript JIT.

[ Edited by Fab on 2010/11/15 2:16 ]

> I am on a business trip in San Fransico when I get back to my studio
> in LA I will elaborate more.

Did you find the time yet?

Quote:

Erm... Why do I see you now with a can of gasoline and a matchbox?

I thought magnetic put very clear that he no longer remembers which supposed HD video he saw with that unexplicable low CPU usage.

It was all a mistake, or not?

> Why do I see you now with a can of gasoline and a matchbox?

Because you didn't follow the thread properly? ;-)

> I thought magnetic put very clear that he no longer remembers which
> supposed HD video he saw with that unexplicable low CPU usage.

Yes, that's why he said he wants to check his browser history as soon as he has returned to LA.

> It was all a mistake, or not?

That's what I've suspected from the start. Only a link to that video can clarify.

Hey Andreas

Apparently unlike you I have a life, and my top priority isnt posting on a forum to please YOU. I will report my findings when I set my peg up again as I had to do repair on a4000t. And do yourself a favour and dont get confrontational.

> my top priority isnt posting on a forum to please YOU.

Apparently you find the time to post on a forum to not please me ;-)

> I will report my findings when I set my peg up again

I'm looking forward to it and hope you won't mind me reminding you now and then, just in case the busy man that you are forgets about it.

>> Did you find the time yet?

> dont get confrontational.

It's obvious that you read something else than my harmless question (quoted above for reference). Whatever "confrontational" it is that you think you just replied to, let me assure you that it's just in your head.

Quote:

But I did. For example, now it's the time to bring up that muppets photo again.

Quote:

And that's what I suspected from the start too: You knew it had to be a mistake, but insist in having confirmation for the obvious.

Have you ever thought if that's worthwile? Especially, if you see it might piss people off...

> But I did.

I don't think so, else you wouldn't have thought that my simple question was something that had already been more or less answered by magnetic when in fact he held out the prospect of providing the link as soon as he fires up his Pegasos again.

> You knew it had to be a mistake

No, as I said I *suspect* it. See there for reference:

https://morph.zone/modules/newbb_plus/viewtopic.php?topic_id=7522&forum=11&post_id=78919#78919

> Have you ever thought if that's worthwile?

Yes, I did. Conclusion: Asking simple questions costs me much less time than discussing with you whether it's worthwile or not to ask them ;-)

> Especially, if you see it might piss people off...

A simple harmless question pissing someone off indicates that there's more to that reaction than just the factual matter. But that's not my problem, really.

> if we want speed, better ask WebKit guys to write
> a PPC implementation of their javascript JIT.

Nothing about Nitro/JaegerMonkey unfortunately, but there seems to be progress on Nanojit/TraceMonkey for PPC:

"With big thanks to Edwin Smith at Adobe who put up with my constant queries, I'm proud to announce that we're making progress on doing the obvious: expanding the PowerPC nanojit so it can be used as part of TenFourFox. Ed helped me out with some of the fiddly aspects, and we're now full speed ahead on implementing the LIR instructions needed to fully support the length and breadth of JavaScript that the Mozilla framework requires. Already it is passing most of the LIR assembler tests, and when it passes all of the supported ones, I'll start work on integrating it with the main browser."
http://tenfourfox.blogspot.com/2010/12/enter-nanojit.html

"The nanojit has been up and running for a few days on my personal dev build and is pretty stable, but the results are currently mixed, and I'm going to do a full post on that when I have some hard numbers to present. The long and short of it is, for many tasks the nanojit is a screamer, but for certain tasks it is actually slower (because remember, the trace and emit step has overhead of its own, and any savings the JIT makes seem to be outweighed by that overhead under certain conditions). The argument is which benchmark to believe, but I'm still working on optimizations to the nanojit, it still seems to be an overall win"
http://tenfourfox.blogspot.com/2010/12/got-it-flaunt-it.html

"The plan for the nanojit then is this: it will come out as part of beta 9 [...]. It has been tested mostly on G5 and G4/7450, though I also ran it on a G4/7400 and it seemed okay. [...] See if it's faster for your workload. See if it's something you can improve upon. It's a start."
http://tenfourfox.blogspot.com/2011/01/how-bench-was-run-and-what-it-got-us.html

[ Edited by Andreas_Wolf on 2011/1/2 1:18 ]
[ Edited by Andreas_Wolf on 2011/1/7 18:14 ]

> Nanojit/TraceMonkey for PPC

And more WebKit related also Google's V8 used in Chrome/Chromium:

https://github.com/ic/v8-powerpc

Edit: New project attempting to build Chromium including V8 for PPC:
http://code.google.com/p/chromium-leopard/

Edit2: Another new project for V8 on PPC:
https://github.com/PowerLinux/v8

Edit3: Another one by IBM:
https://github.com/ibmruntimes/v8ppc

[ Edited by Andreas_Wolf 20.09.2016 - 03:24 ]

Update:

> there seems to be progress on Nanojit/TraceMonkey for PPC

It got released, the first PPC web browser with a JS JIT:

"After working more with the nanojit, I have decided not to make it the default, but it is there, and you can use it."
http://tenfourfox.blogspot.com/2011/01/beta-9-now-available.html

"The first draft of the PowerPC-specific nanojit is now available, helping to accelerate certain JavaScript operations. This portion of code is specific to TenFourFox and is still under development, so it is not enabled by default."
http://code.google.com/p/tenfourfox/wiki/ReleaseNotesBeta9

"As of beta 9, 10.4Fx uses a custom PowerPC-enabled nanojit for JavaScript acceleration through TraceMonkey. Currently 10.4Fx is the only Mozilla-based browser to use this module, as it was developed for it; it is being contributed back to the Mozilla codebase for other Mozilla source consumers."
http://code.google.com/p/tenfourfox/wiki/TechnicalDifferences

"Thanks to user PoLiYa who wrote in to tell me about his fabulous score on SunSpider with his PowerBook G4/1.5GHz and the nanojit turned on: a whopping 2866ms, down from around 6400ms. [...] I got out my iBook/1.33GHz and repeated the test. [...] On the G4, the nanojit not only worked, it easily cut SunSpider scores in half, and Dromaeo was also about double the speed. User agg23 over at 68KMLA tried it on his G3, and also saw a similar speed improvement, from 16000ms down to around 8900. [...] the plan for beta 10 is to ship the G3 and G4 versions (7400 testing is pending) with the nanojit turned on for content (web pages). If this works, it may also be enabled for chrome (i.e., the browser's own JavaScript, and add-on JavaScript)"
http://tenfourfox.blogspot.com/2011/01/lies-damn-lies-statistics-and-g5-beta-9.html

Nice comparison chart:
http://www.floodgap.com/software/tenfourfox/graf.jpg

[ Edited by Andreas_Wolf on 2011/1/18 11:36 ]
[ Edited by Andreas_Wolf on 2011/1/22 0:49 ]
[ Edited by Andreas_Wolf on 2011/2/8 3:21 ]

Update:

Some information from the author of TenFourFox:

http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=33201&forum=15#601209

Quote:

"If any MorphOS developers read this thread, I would love to get MorphOS running on my idle iBook G4s. If it runs on G4 Mac minis, I can't imagine why it would be difficult to get it to run on iBooks as the hardware is nearly identical, and I'd love a portable Miggy."

Fab, got any time left to contact this guy? Or is this JavaScript PowerPC JIT unpluggable into OWB?

I fear it wouldn't be trivial for someone not familiar with javascriptcore integration in WebKit. This should be asked to WebKit developers, rather.

> Time to re-open the renaming discussion!

Kas1e two weeks ago:

"Himon, which is a port of mui-OWB browser from morphos to amigaos4."
http://jurassic-blog.de/?p=99

Kas1e a week ago:

"Fab already have a name , he will change it for 1.13 or so."
http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=32847&forum=32&start=1020#611815

Fab three days ago:

"I wouldn't allow it to have a different name for the OS4 port, anyway. That's part of my license."
http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=33605&forum=32&start=100#612258


Edit: added another quote

[ Edited by Andreas_Wolf 29.04.2011 - 22:15 ]

Or we could just take one of the many suggestions by OS4 users.

> Or we could just take one of the many suggestions by OS4 users.

We? It's Fab's decision alone I'd say. And if I combine Kas1e's and Fab's statements I quoted above it seems the decision has long been arrived at Himon. I could be mistaken, though.

Fab's never shown any inclination to change the name. Frankly, except for the fact that it doesn't reflect Fab's continuous work on it, I'd be happy with...OWB.

Quote:

You are saying someone just spilled the beans about the new name?

Quote:

On the contrary, he has suggested that a renaming would occur in a close future, like for version 1.13 or so.

Which makes perfect sense, since it hasn't had much to do with the OWB project for a very long time, and a lot of people obviously gets confused about the name, thinking this would be the same as what's available for OS4 etc, when it in practice really is a new browser based on webkit.

> Fab's never shown any inclination to change the name.

Fab in August 2010:

"if you find a nice one, maybe i'll think about changing it."
https://morph.zone/modules/news/article.php?item_id=1787&comment_id=4825&mode=thread

> You are saying someone just spilled the beans about the new name?

I'm suspecting that 'someone' spilled the beans about the new name two weeks ago.

It's not about the name, it's about how much this browser can do.. Take a look on AmigaOS name.. It has something but can much less than MorphOS can.