@Fransexy

... and i'm quite sure that we could even get another 10% discount if we start to mass request this toy...

_________________
retired

@d0c

+1 for the powerstation.

@all

Or a port to a mac g5 dual core would be equal nice.

_________________
I'm an antique. Don't light my fuse

@all

Why do we need or want to port amigaOS4 to a discontinued ppc HW ???.

We as community had the chance when PPC was still in bussiness to port amigaos to but instead we waited and waited till PPC hw market to die than start porting to dead ppc hw.

port the AmigaOS4.x to X86 lets sort this HW problem once for all and start porting power apps like firefox/openoffice/flash or other usefull apps to amigaOS.

I DO NOT WANT TO BUY Macs from ebay or search net for second hand PPC HW all day but instead i would like to walk into my loca pcworld or other pc shop and buy one.

xbox/ps3 stays with ppc coz its just an game consoles not desktop OS`es and as i said that before amigaOS its not a game for ps3 to be ported.


To me Next Amiga should be on x86 to avoid HW side of things and start porting power apps and mainly to avoid hw side since we dont have £ =$=rich man or company to produce a special HW for small community .


we need to improve oS4 itself and we need to work on the software to catch up those lost years.

_________________
http://aros-exec.org/

http://serk118.blogspot.com/

@serk118

Its a tough one, but I feel similar in that a lot of time is wasting... I do hope the next few months Hyperion and ACube reveal a solid strategy in how they want to take the Amiga community forward and make it viable once again... But as the past suggests, very difficult indeed...

Q!

"i am home"

@matthey

Quote:

I have the money and I would buy a PowerStation tomorrow if AmigaOS 4.x came out on it even if just one processor was supported at first.

The same goes for me. If OS4 came out for PowerStation, I could buy it tomorrow

_________________
AmigaOS 4.1 FEu2 on Sam440ep-flex 800MHz 1GB RAM
C128, A500+, A1200, A1200/40, AmigaForever 2008+09+16, 5 x86/x64 boxes
Still waiting (or dreaming) for the Amiga revolution...
m4rko.com/AMIGA

@minator

It comes with a ATI X1650 Pro graphics card

_________________
Raptor BlackBird (awaiting a Miggy like OS)
Amiga 4000 3.9 / Pegasos II 4.1 F.E.

@serk118
I don't want x86. The only reason x86 is winning the CPU war is that it sells the most. An inferior design was made the fastest because of money. It's kind of like taking a tank and turning it into a drag strip queen. It can be done with enough money and horsepower but it's not elegant. Also, I can read PPC assembler somewhat which I can't x86. Reading assembler is still useful for debugging as well as optimizing. Windows is so slow and CPU hungry because most code is 50% slower than what is possible. AmigaOS is already on PPC and porting would take major resources away from improving. It would also introduce more bugs, bloat and confusion on which CPU flavors are needed. I don't mind spending $ on PPC hardware, but I want something fairly powerful for a desktop like this PoweStation. The SAM would make a nice platform for a laptop.

@all
SATA hard drives are plug and play with SAS controller so SAS HD is not necessary.

Last edited by matthey on 15-Aug-2009 at 06:09 PM.

@nubechecorre

If MorphOS is going to mac minis, then I bet Hyperion are doing something similar. I hazard a guess that it's arm cortex 8 or something.

Last edited by Al4 on 15-Aug-2009 at 06:28 PM.

Quote:

4 on-board USB 1.0 ports.

Seriously, is that a joke ?

USB 3.0 has just been released...

You pay a so called monster and cannot use USB 2.0 stuff... This is a joke. Wouldn't you prefere something designed for the desktop use ? on one side you don't have USB 2.0, and on the other side you have 2Gb ports... Seriously, this has been designed for the server market.

_________________
http://www.warpdesign.fr/

@minator

Quote:

I'd keep an FPGA on board but change it. Why? In a word: Minimig If there's a big enough FPGA on board with some form of I/O that can route the Gfx O/P to a PCIe card (probably just copy the frame buffer) you get a classic Amiga built in... Ok, it's only an A500 and wiring it up an emulated 68K might be er, fun but that's lots of old stuff you can run...

I dont see point. It seems Minimig is not selling at all and having an FPGA dsoldered on a board increases failure rate. Most users dont know what to do with FPGA nor can code in VHDL. It is just waste of money.

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

Quote:

It seems Minimig is not selling at all

Well, of course it doesn't... It is experimental, it hasn't been designed to be produced/sold.

An Amiga in a joystick, that you just plug to a power and tv, with something like 10-20 games included would sell though...

With the minimig you need to build it yourself, it isn't preinstalled with roms/games,... And since it's produced in such low quantity is quite expensive for what it does too. Not surprising it doesn't sell :)

Build an Amiga in a joystick and I'm sure you sell 50-100k pieces... not really big, but we haven't seen such numbers since the last CBM days in 92-94, so it wouldn't be that bad for a start :)

Back to the topic: I don't see what more than a PC the next Amiga could do... PC hardware seems fine... You can't do better anyway. You could do things in a different way, like copper, etc.. but seriously: what's the use today ?

Last edited by Leo on 15-Aug-2009 at 10:47 PM.

_________________
http://www.warpdesign.fr/

@matthey
I'll bite.

Quote:

I don't want x86. The only reason x86 is winning the CPU war is that it sells the most. An inferior design was made the fastest because of money. It's kind of like taking a tank and turning it into a drag strip queen. It can be done with enough money and horsepower but it's not elegant.

PowerPC's register data movements between VMX and GPR is not elegant. PowerPC also has a higher overhead for functional calls(1).

On the money issue. During the development of K7 Athlon, AMD was relatively a small corporation i.e. smaller than Motorola/Freescale. What AMD have is talent from Digital Equipment Corporation (DEC). Alpha EV6 bus technology was recycled for K7 Athlon's FSB.

The original drag strip queen was DEC's Alpha i.e. it has the raw compute performance and clockspeed.

Reference
1. http://www.barefeats.com/doom3.html

Quote:

Also, I can read PPC assembler somewhat which I can't x86. Reading assembler is still useful for debugging as well as optimizing.

On optimisations, X86 has industry leading compliers. X64(X86-64) has clean-up alot of X86 ISA issues.

Quote:

Windows is so slow and CPU hungry because most code is 50% slower than what is possible. AmigaOS is already on PPC and porting would take major resources away from improving. It would also introduce more bugs, bloat and confusion on which CPU flavors are needed. I don't mind spending $ on PPC hardware, but I want something fairly powerful for a desktop like this PoweStation. The SAM would make a nice platform for a laptop.

On Windows issue, there's AROS X86 i.e. IcAROS 1.1.3 distro.
This post was typed on IcAROS 1.1.3's OWB.

Last edited by Hammer on 15-Aug-2009 at 11:41 PM.
Last edited by Hammer on 15-Aug-2009 at 11:30 PM.
Last edited by Hammer on 15-Aug-2009 at 11:29 PM.
Last edited by Hammer on 15-Aug-2009 at 11:19 PM.
Last edited by Hammer on 15-Aug-2009 at 11:12 PM.
Last edited by Hammer on 15-Aug-2009 at 11:03 PM.
Last edited by Hammer on 15-Aug-2009 at 10:56 PM.

_________________
Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB
Amiga 1200 (Rev 1D1, KS 3.2, PiStorm32lite/RPi 4B 4GB/Emu68)
Amiga 500 (Rev 6A, KS 3.2, PiStorm/RPi 3a/Emu68)

+1 for the power station

_________________

@ALL

+1 for the power station... and of course +1 for the LimeNoteBook PPC's

_________________
Quote:

Easy Pocket Money, Freelancers & Experts Online

MiniMig FPGA, Sam440 Flex 733Mhz PPC, Amiga OS 4.1 Update 2, MorphOS 2.4, Other - AmiKit + Cloanto Amiga Forever 2008 + E-UAE, AmigaSYS

@minator

Probably just being grumpy, but I don't think there will be another "Amiga" as anybody recalls it. Some argue that Amiga's already wandered off its path, splitting off into the various flavors. The thing that set the Amiga apart was the hardware it has, which has been surpassed in capability, though the Operating System has proven to be advanced enough to keep up with modern "innovations". Interesting how many of the design ideas prevalant in the Amiga started to take off a few years after C= went bust. (Engineers still need jobs, and I'm sure they found them in the x86/accessory markets) Ideally, MOS/OS4/Aros could co-operate towards a common OS, OS4 and MOS setting the API, and Aros getting things over to X86. (Just an example) This would allow cross-platform compatibility on the Amiga side at the very least, and would provide a single OS that would work on a wider variety of hardware. (Hell, go for cellphones and PDA's, too, not to sound too much like the Amiga Anywhere flop)
I guess I'm trying to say each "flavor" has strengths and weaknesses, and would benefit from combining talent for a new OS, which seems to be the remaining jewel left of Amiga that hasn't been passed by. A web browser w/ CSS, Javascript, Flash, etc... would be nice, however remarkable progress has been done in this area of late and shouldn't be amiss too much longer. (I've gotten NetSurf to run on AOS 3.9 on native hardware, even. Was shocked. until the speed was what I would expect of a 50mhz 060)
Also, advantages in running an Amiga need to be there, too. An x86 port that's cheaper than Windows, easier than Linux, and lightweight would give a small edge, even if dealing with the X86 architecture. Plus, crude or not, it does work and have horsepower. Native Amiga apps. should be more a priority than Linux ports, otherwise ppl will just go for a "user-friendly" linux distro. (Linspire, Suse, etc..)
A port of Wine would be helpful to make transferring software easier, and perhaps a similar system on other operating systems to convert Amiga API's back over to Windows or Linux API's, though that could be commercially shooting oneself in the foot. (UAE is great for classic stuff, but modern apps, the likes of which my 1200 enamored butt can't use, won't work) The AROS guys especially might have luck in this, with familiarity with x86 hardware, plus UAE could possibly help.

In a nutshell, if the differing "flavors" merge, Amiga won't necessarily nip Microsoft in the butt, however it would result in something that much more remarkable in an operating system. I know I'm daydreaming, and several others here (and other forums) have stated the same (or opposite in some cases), but I'm in a weird mood right now.

As far as a computer that had the same initial impact as the classic Amiga's, something incredibly ground-breaking and unexpected would have to happen. The biggest thing with the Amiga was being a computer with the capabilities it had when x86 stuff could beep and show 16 colors of text, and Macintosh was capable of 2 colors with a 1-button mouse. The multitasking abilities and (more or less) advanced OS (1.x wasn't all that great, 2.x and up were much better) weren't that big of a deal considering the graphics and audio capabilities were more in-tune for a gaming machine, but unlike a console, one that could grow with the user as they bored of the joystick. Another computer that would be as advanced in capability compared to todays systems is rather difficult for me to imagine, and hopefully something bearing the name Amiga will realize it.

Last edited by kvasir on 16-Aug-2009 at 01:01 AM.

_________________

@Hammer

Quote:

Hammer wrote: PowerPC's register data movements between VMX and GPR is not elegant. PowerPC also has a higher overhead for functional calls(1).

There shouldn't be much data movement between the integer unit, FPU unit and VMX unit anyway. The idea is to load them up with data and let them run in parallel. They have to sync up while transferring data between units and that is going to hurt the performance of both units affected. I would rather have better performance when they are not synchronized though.

The overhead for function calls is OS dependent. The old MAC OS used to use A-line CPU traps/exceptions that had about twice the overhead of the Amiga library calls. Hyperion has reduced the library call overhead in 4.x on the PPC over the Amiga 68k. Even if the branch and return instructions cost more on the PPC, the additional general purpose registers available for register passing more than makes up for it. Registers don't have to be swapped around before or after calling as they are truly general purpose. AmigaOS 4.x makes very good use of these additional general purpose registers. GCC now supports efficient inlining of PPC code. The compiler not being as good for PPC had much to do with the test you listed. The code was written and optimized on the x86 for the x86, also. The G5 is not the pinnacle of PPC development. It's still quite good considering how much less resources went into producing it than any of the modern x86 processors.

Quote:

On the money issue. During the development of K7 Athlon, AMD was relatively a small corporation i.e. smaller than Motorola/Freescale. What AMD have is talent from Digital Equipment Corporation (DEC). Alpha EV6 bus technology was recycled for K7 Athlon's FSB.

I wasn't talking about the money the company had but the money the x86 market has because of the x86 and winblows duopoly. AMD wanted a piece of that duopoly.

Quote:

On optimisations, X86 has industry leading compliers. X64(X86-64) has clean-up alot of X86 ISA issues.

Again a result of the duopoly. Nothing can compete. A new CPU design that was 10x as good/fast could come out today and it would fail, Intel would buy it and the new technology would be kludged into the x86 architecture. The result would be 2x as fast.

Quote:

On Windows issue, there's AROS X86 i.e. IcAROS 1.1.3 distro.

I'll agreed there. If anyone wants x86 there is AROS already. I have no problem with AROS and UAE based Amigas. I even help them but I don't want an x86 Amiga ;).

Last edited by matthey on 16-Aug-2009 at 03:14 AM.

Except for altivec/sse stuff, I don't see where you would need to use ASM.

And it's something normal users don't care anyway.

For a normal user, it could be x86, ppc, 68k, arm, 6502,... it shouldn't matter.

Actually, there is one difference from a user point of view: x86 would be faster.

Last edited by Leo on 16-Aug-2009 at 03:41 AM.

_________________
http://www.warpdesign.fr/

@Leo

It has USB 2.0

_________________
Raptor BlackBird (awaiting a Miggy like OS)
Amiga 4000 3.9 / Pegasos II 4.1 F.E.

Quote:

TheKorn wrote: @Leo It has USB 2.0

On a card only. The MB supports what OS4+ has now.

Chris

@Hammer

Quote:

PowerPC's register data movements between VMX and GPR is not elegant. PowerPC also has a higher overhead for functional calls(1).

I always thought the way x86 had to dump the FPU registers on the stack because the x86 vector unit used them was inelegant. I assume that's been fixed!

Being register based and with SysV using registers for first 8 parmeters PowerPC should be good at functions. Of course, the SysV ABI could introduce overhead by trackng the stack. But AFAIK these can be avoided with optimisation flags with GCC.

Or we should use VBCC!

Quote:

On optimisations, X86 has industry leading compliers. X64(X86-64) has clean-up alot of X86 ISA issues.

It's been thay way for a while. But, what do you expect for 68k and PowerPC, using a free compiler? It ain't going to be the best! I wonder if Apple games and code would have ran faster if they used VBCC.