Click Here
home features news forums classifieds faqs links search
6071 members 
Amiga Q&A /  Free for All /  Emulation /  Gaming / (Latest Posts)
Login

Nickname

Password

Lost Password?

Don't have an account yet?
Register now!

Support Amigaworld.net
Your support is needed and is appreciated as Amigaworld.net is primarily dependent upon the support of its users.
Donate

Menu
Main sections
» Home
» Features
» News
» Forums
» Classifieds
» Links
» Downloads
Extras
» OS4 Zone
» IRC Network
» AmigaWorld Radio
» Newsfeed
» Top Members
» Amiga Dealers
Information
» About Us
» FAQs
» Advertise
» Polls
» Terms of Service
» Search

IRC Channel
Server: irc.amigaworld.net
Ports: 1024,5555, 6665-6669
SSL port: 6697
Channel: #Amigaworld
Channel Policy and Guidelines

Who's Online
10 crawler(s) on-line.
 63 guest(s) on-line.
 1 member(s) on-line.


 zipper

You are an anonymous user.
Register Now!
 zipper:  2 mins ago
 kriz:  25 mins ago
 Karlos:  30 mins ago
 redfox:  59 mins ago
 DiscreetFX:  1 hr 1 min ago
 danwood:  1 hr 20 mins ago
 NutsAboutAmiga:  1 hr 24 mins ago
 K-L:  1 hr 30 mins ago
 Kronos:  1 hr 31 mins ago
 kiFla:  1 hr 52 mins ago

/  Forum Index
   /  Amiga OS4.x \ Workbench 4.x
      /  Should Linux kernel power future AOS solution ?
Register To Post

Goto page ( Previous Page 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 )
PosterThread
OlafS25 
Re: Should Linux kernel power future AOS solution ?
Posted on 9-Nov-2013 18:27:38
#361 ]
Elite Member
Joined: 12-May-2010
Posts: 6363
From: Unknown

@Signal

Where do you know?

Hope you do not take it too seriously. Noone at Hyperion or a-eon will do what we want so it is not important. Some will take one direction, others a different one. And if people are happy with it then be it. But I can at least have a different view and hope I do not insult others by expressing it.

 Status: Offline
Profile     Report this post  
Raffaele 
Re: Should Linux kernel power future AOS solution ?
Posted on 10-Nov-2013 22:57:01
#362 ]
Super Member
Joined: 7-Dec-2005
Posts: 1906
From: Naples, Italy

@cdimauro

Quote:

cdimauro wrote:

It doesn't mean that ALL PowerPC "emulations" will be slow. Take a look at the Rosetta results: http://www.macworld.com/article/1049163/intelimacs.html

"ROSETTA iMac Core Duo/2GHz iMac Core Duo/1.83GHz iMac G5/2.1GHz
Adobe Photoshop CS2: Suite 2:50 3:02 1:16
Microsoft Word Scroll: Scroll 1:58 2:05 0:57"

As you can see, an OLD 2Ghz Core Duo runs at about HALF (not 1/13!) the speed of a 2.1Ghz G5.

Consider that:
- they are old (January 2006!) machines;
- both the o.s. (OS X Tiger) and Rosetta ran in 32-bit mode (x86).

And that now we have:
- double the clock;
- more efficient micro-architectures (Nehalem and successors);
- the new CPUs with new instructions which can help emulating big-endian machines (thanks to the very welcome MOVBE) and Altivec (AVX is a much more efficient and flexible SIMD unit/ISA; the coming AVX-512 will be MUCH better);
- 64-bit architectures and software, which can GREATLY enhance both the emulation and the quality of JIT generated code.

As a very rough and conservative estimation, I think that the performance can be 3 times better.

So, didn't you like a 3Ghz G5 performance? That's for a newer Rosetta version, OK, but I think that a new project can achieve good results, so having a good user experience.


Then ask Apple for Rosetta PPC emulator sources and be luck if they release it to you...


Quote:

Quote:


We all know your passion for Classic Amigas,

You're wrong: my passion is for the Amigas. I don't know of any "Classic" Amigas.


It seemed me so...

Quote:

There'll be a transition period to have almost all applications ported, such as happened to MacOS X x86.


It has being happening... OWB and Netsurf for example are replacing glorious old AWeb and IBrowse. Hollywood is replacing with steroids what Scala was...

Quote:

Quote:
Drivers are the last of our problem actually.

First step is to integrate well new Operating Systems with the existing hardware that at this moment is still not supported entirely. Look at AmigaOS and its lack of USB 2 support, for example.


And isn't the USB 2 support... a DRIVER implementation? It took 7 years to have it, and you have the bravery to say that the drivers aren't the last of the problems now?


Yes I have the bravery. It is not my problem USB2 driver support.
I have MorphOS and into MOS we have it since version 2.1...

It left the fact that AmigaOS could have it since 7 years ago, and they were uncapable to integrate it in the OS...
It is called bad support of the hardware device called USB 2. It is a problem of integrate correct routines of USB2 into the existing driver, not a lack of drivers itself.

Did you understand the difference?

Hence thus we need that all existing devices and videocards should be "integrated" well into the OS, complete for example of 3D engines like Gallium or whatever else...
But first we need all OS being well enginereed and running well...

Quote:

Quote:

Only when all things will starting working flawlessly and the drivers and the 3D engines will be integrated fully in AmigaOS, resulting in no bugs, then and only then we could think of changing actual architecture, and migrating to X86 and ARM, or else we will encounter again the same problems even being in possess of all kind of drivers available for other systems...

OK, so may be my grandchildren will have a chance to see it.


Yes. Maybe.

Teach them of Amiga.

Last edited by Raffaele on 10-Nov-2013 at 11:59 PM.
Last edited by Raffaele on 10-Nov-2013 at 10:57 PM.

_________________
"When the Amiga came out, everyone [at Apple] was scared as hell." (J.L. Gassée, former CEO of Apple France and chief of devs of Mac II-fx, interviewed by Amazing Computing, Nov 1996).

 Status: Offline
Profile     Report this post  
Raffaele 
Re: Should Linux kernel power future AOS solution ?
Posted on 11-Nov-2013 0:19:59
#363 ]
Super Member
Joined: 7-Dec-2005
Posts: 1906
From: Naples, Italy

@OlafS25

Quote:

OlafS25 wrote:
@Raffaele

I have done some google. QBox is certainly a interesting idea but there is no indication that there is any work done on it in recent times (the MorphOS team is small too and they have updated MorphOS (ABox) and added a lot of hardware). So you talk about something that is at the moment propably years away (if they decide to port MorphOS on different hardware). You can dream of a lot of things but at the moment it is just theory.

ARIX will be reality in a couple of days and people can judge it then. Projects like AEROS (and a future version of AROS Vision) that are using Linux for driver support and applications are real too. Linux has a huge driver base and that alone is a huge benefit. Even if MorphOS has a good kernel (even if it would be better than the Linux kernel) they still have to develop every driver and that is a huge disadvantage.



As You noticed I am complaining it too...

Version 0.4 of MorphOS was demoed in 1999 (?)
Since that they had developed ABox only... and putting all hardware drivers in ABox Amiga legacy affected box, and not in microkernel complete support QBox...

At least Trevor could take a look and decide if buying Quark Microkernel from MorphOS team and provide also AmigaOS of a modern Microkernel environment...

Once both system (or even three systems including AROS) will share the same microkernel, then it will be easy to develop common drivers as again the Amiga software developers will be united again.

It is a matter of money, and Trevor has it, and he is free to decide if purchasing Quark Microkernel from MorphOS Team or perhaps not purchasing it...

_________________
"When the Amiga came out, everyone [at Apple] was scared as hell." (J.L. Gassée, former CEO of Apple France and chief of devs of Mac II-fx, interviewed by Amazing Computing, Nov 1996).

 Status: Offline
Profile     Report this post  
cdimauro 
Re: Should Linux kernel power future AOS solution ?
Posted on 11-Nov-2013 17:39:18
#364 ]
Elite Member
Joined: 29-Oct-2012
Posts: 3650
From: Germany

Quote:
Raffaele wrote:
@cdimauro

Then ask Apple for Rosetta PPC emulator sources and be luck if they release it to you...

First and as I already said, Apple has no interest on selling Rosetta.
Second, Rosetta was initially developed by a small company. So, no, it wasn't a project by a big company.
Third, Rosetta isn't the only existing PowerPC binary translator, and even if it was, nobody stops other people di write a new one, even more efficient.

Rosetta was just an example that there's the possibility to run the (old) PowerPC code on x86 machines at very good speed, with room for further improvements (especially with x86).

Investing money on a Rosetta-like project will be much better than continue wasting them over an architecture ad the dead end.

Quote:
Quote:
You're wrong: my passion is for the Amigas. I don't know of any "Classic" Amigas.


It seemed me so...

I don't see why, since, as I stated many times, no "Classic" Amiga machines existed. In my more 10 years of experience with the Amigas, there was nothing called "Classic Amiga".

But may be you're too young and missed that period: this could explain why you're talking about non-existing Amiga machines such as the "Classic" that you mentioned...

Quote:
It has being happening... OWB and Netsurf for example are replacing glorious old AWeb and IBrowse. Hollywood is replacing with steroids what Scala was...

I was talking about existing project just recompiled or ported to the new platform. Please read well before answering, in order to acquire the meaning of the discussion.

Quote:
Yes I have the bravery. It is not my problem USB2 driver support.
I have MorphOS and into MOS we have it since version 2.1...

It left the fact that AmigaOS could have it since 7 years ago, and they were uncapable to integrate it in the OS...
It is called bad support of the hardware device called USB 2. It is a problem of integrate correct routines of USB2 into the existing driver, not a lack of drivers itself.

Did you understand the difference?

Why don't you read before writing? In the end you're agreeing with me! Drivers ARE important, and the USB 2 stack experience with OS4 proved it.

Quote:
Hence thus we need that all existing devices and videocards should be "integrated" well into the OS, complete for example of 3D engines like Gallium or whatever else...
But first we need all OS being well enginereed and running well...

As I already stated, may be my grandchildren will see them. May be...

Quote:
Yes. Maybe.

Teach them of Amiga.

Did it.

Quote:
Raffaele wrote:
As You noticed I am complaining it too...

Version 0.4 of MorphOS was demoed in 1999 (?)
Since that they had developed ABox only... and putting all hardware drivers in ABox Amiga legacy affected box, and not in microkernel complete support QBox...

So QBox is a dead platform itself.

Quote:
At least Trevor could take a look and decide if buying Quark Microkernel from MorphOS team and provide also AmigaOS of a modern Microkernel environment...

Once both system (or even three systems including AROS) will share the same microkernel, then it will be easy to develop common drivers as again the Amiga software developers will be united again.

It is a matter of money, and Trevor has it, and he is free to decide if purchasing Quark Microkernel from MorphOS Team or perhaps not purchasing it...

It doesn't make sense, as many already told you in the other thread.

Anyway, I don't see why you should use a micro-kernel (or kernel) putting another micro-kernel (Exec) on top of it: just let Exec run alone!

 Status: Offline
Profile     Report this post  
Goto page ( Previous Page 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 )

[ home ][ about us ][ privacy ] [ forums ][ classifieds ] [ links ][ news archive ] [ link to us ][ user account ]
Copyright (C) 2000 - 2019 Amigaworld.net.
Amigaworld.net was originally founded by David Doyle