Why no NEW PPC accelerators?

[jiffydos]

Watching ebay for the last few months and seeing 10+ year old PPC cards selling for $500+ makes me wonder why no one, especially Jens, seems to even be talking about making a new one. Would the research and/or parts be that expensive that it would not make up for the money it could make?

I would spend $500 or even $750 on something brand new and warranted, but an old, slow 603e... not so much.

I would think it would be doable for someone to build a new PPC card, maybe based on the Sam440 or Efika processors/projects.

Any idea why not?

[lovrenco]

Maybe because you have new boards which work with recent version of Amiga OS (4.1).

PPC accelerators for classic work with version 4.0, design and production would be sky high overpriced and it would work on 20 years old MOBOs....

What do you think?

[jiffydos]

I think if such a huge chunk of the work has already been done to create these projects, it would make sense to give the "classic" world the chance too. I would spend $500/$600 on an "efika" accelerator for my A4000 before I'd spend that just on the standalone board. That way, I could run 68K code OR OS4.x PPC code.

I'd envision something the the old Sonnet Quad-Doubler where the original 68040 chip from the A4000 is inserted into the new accelerator to give the dual CPU experience similar to Blizzard cards.

[CountRaven]

Expensive to be created ...

On the other hand new boards are on the market now running 4.1. Your question would be more into the point if there was still no new hardware available.

But times have changed to the better -having in mind the standards of our almost dead platform-.

Still the point about giving classic a new chance is good and I am in it but do you think the people that would buy a new PPC accelerator would be enough to cover the cost of a new massive production of those cards?

Do not think so...

Even the new hardware has low specs and it is expensive to buy. But it is fun cause it is Amiga -or Amiga oriented- after all.

No one even dares to deliver a board with high specs based on today’s standards, letting it still be PPC -which is another dead technology but still rocks in Amiga world and it is ok for me and for all of us in the community-. I would love a PPC system in 2Ghz with 4GB RAM, lots of USB slots, gigabit ethernet support, blue ray etc.. and all those under the flag of AmigaOS / MorphOS.... But yeap it is still a dream...

Pegasos is the faster system so far, capable to run both OSes.

You would better wake up @ Genesi and release a super-duper-high-speed-high-end model.

We need more power.....

And MorphOS 3 everywhere :D

[SamuraiCrow]

@JiffyDOS

The problem with high-end accelerator cards are that you have to wonder after a while, whether you are plugging a card into an old computer or if you are plugging an old computer onto a new one.

The old bus controllers on the classics can't handle the bandwidth needed to use the more modern PPC chips that the SAM and the like have built in.

Even the latest Efika boards will not have PowerPC chips on them but instead will have ARM processors.

[persia]

The problem is quantity, the Amiga market is so small and the development costs so high that it just doesn't pay.

Look at SAM, a glorified mobile phone without the phone. You sell a couple thousand and that's it. Al your costs have to be recouped in those couple thousand sales, so the price is five to ten times what it should be.

It's a zero sum game, at best. You make enough money to exist and that's it. And don't make a mistake or you've lost a lot of money.

There is one clear solution, take AmigaDOS to X86 where ther's plent of cheap equipmwnt. I'd love to have an AmigaOS 4 EEEEEEEEEE pc. A few hundred dollars, neat amd compact. And AmigaOS 4 would fly in such an environment. A full machine at half the cost of a turn of the century SAM board alone.

[Comi]

Ask Elbox, and find out!
They preparing accelerators 8 years now 9!

[jpera]

I would be happy to see a new Amiga 3000/4000 68040 processor board with modern memory slots. It would bring my A4000 back to life whose A3640 broke down a few years ago.

[Linde]

Yeah, a new 68k accelerator would be nice. But I don't really see the point of a power pc. There are so few demos and softwares that support it anyway, and now we have fast new PPC only boards for that stuff.

[persia]

Let's say we could get the original 4000 board design (it apparently disappeared last century), this would save us design costs. This is a what, 4 layer board, how much would it cost to manufacture say 500 or 1000 of them? How much for the chips? Can we even get new parts nowadays?

Running further roughshod over copyright, we could clone the ROMs. So now we have a machine that is AI's Lawyer's wet dream. Case, power supply. We're probably above the cost of used and it's still pre-turn of the century technology.

Now we still need to clone a PPC board...

Does the vapourware natami have any plans for a ppc board?

[bloodline]

persia wrote:
Let's say we could get the original 4000 board design (it apparently disappeared last century), this would save us design costs. This is a what, 4 layer board, how much would it cost to manufacture say 500 or 1000 of them? How much for the chips? Can we even get new parts nowadays?


The Board is not RoHS... so actually we've already fallen at the first hurdle...


Running further roughshod over copyright, we could clone the ROMs. So now we have a machine that is AI's Lawyer's wet dream. Case, power supply. We're probably above the cost of used and it's still pre-turn of the century technology.

Now we still need to clone a PPC board...

Does the vapourware natami have any plans for a ppc board?

Actually I believe it does, and a lunar landing module... though I think that comes after the planned cure-for-cancer feature!

[neofree]

We should do something simular to Apple.. (using an open source OS under the hood, proprietary GUI the user sees; hybrid approach)

Take Linux to the level we want it to be...

Linux could be the base on a modern hardware...

Then have an proprietary AmigaOS GUI layer that competes with Mac OS X (it already is very simular anyway..)..

Scrap AmigaDOS for Linux...

It's the only thing that makes sense to me to, if we want any chance of bringing Amiga to the modern world and actually possibly bringing it mainstream.. Linux already has leaps and bounds more potential of beating Apple or Microsoft, but it lacks the polished look and feel of the other 2. If someone could come out with a Linux distro that is as cool as Mac OS X, then we have something exciting and new everyone would want to be a part of...

But.. I'm coming down from the Magic Mushroom high now and facing reality.. I've given up on Commodore and Amiga ages ago... I now am a "Mac fanboy"... lol They provide the computer I want NOW..

Anymore I just use emu's for reliving the glory days.. A stock 64, a stock A500 are fine for me anymore.. If we can ever be realistic and agree enough to make Amiga return then yay... Otherwise enjoy your Retro computing and screw this wasted effort..

Thanks,

Neofree

[fatboy]

Sorry for my ignorance but what happened to the original PPC board designs from the various vendors.

Couldn't someone re-manufacture original Phase5 boards etc and pay the patent /copyright holder a royalty???

Surely the parts are still out there and as development cost wouldn't be required, just parts and manufacturing, couldn't original PPC boards for A1200/A4000 be re-produced at reasonable cost???

I'm sure that there are loads of people happy to run OS3.9 with a desktop A1200 PPC board.....I would!...or just an '060' board?

[neofree]

fatboy:

Do you have any idea what it costs to produce a multi-layer/SMT board with parts that aren't readily available?

As I said.. this is all wasted effort.. So OS4 and OS4.1 have become realized.. For what? What am I going to do on this platform? run a few programs that are inferior to their Mac, Windows, and Linux counterparts? I understand the Will... but unless the current owners of Amiga have some real intellegence in today's marketplace, it's all doomed...

BTW. You realize that Amiga is the Spanish word for Girlfriend... I don't see this name even marketable in the modern world... People would question why we are calling our computers a Girlfriend.. Don't we want real Girlfriends?

Commodore is the better name...

Thanks,

Neofree

[fatboy]

Surely most of the parts are still fairly standard(SMT components).

With the global economy being so bad, I'm sure a lot of the large Chinese manufacturing plants will be crying out for ANY sort of orders?

is a £249-£299 card not possible?

[neofree]

Problem is the chips, not the boards....

The CPU's, *custom chips*, etc are not available...

Sure you could find comparable products, design replacements, whatever... And the resulting boards would probably cost $2500+ a piece because of lack of demand.. Or cost $600 and suddenly their business fails. Hmm.. It just doesn't work.. You can order a computer for $200 on Tigerdirect.com because of *volume*..

[Triton199]

what we need is amiga software that can run on commodity hardware. this whole "special hardware to run the software" game is a dead end and couldnt possibly be profitable in any significant way. sure desiging a new accelerator card with modern hardware (FPGAs newer processors etc)could be done but the price would have to be high (500-1000$ range) to even come close to breaking even

[itix]

If new PPC accelerator board was developed it would run no software until someone ports OS4 or MorphOS for it.

[bloodline]

Triton199 wrote:
what we need is amiga software that can run on commodity hardware. this whole "special hardware to run the software" game is a dead end and couldnt possibly be profitable in any significant way. sure desiging a new accelerator card with modern hardware (FPGAs newer processors etc)could be done but the price would have to be high (500-1000$ range) to even come close to breaking even

Yeah, imagine if we had an AmigaOS that ran on standard commodity hardware... You know, if I had started work on such a system 10 years ago I bet we would have had something by now... If such a system existed i'd call it something like AROS or something...

[Jose]

A new PPC board allows you to have both the old and new on the same machine. I'd love to have one in my A4000T which is nowdays mostly unused...

There were new PPC accelerator designs at the time Phase5 went down but nobody seems to know, or those who know don't say it ;), where they ended up.
The designs were for a G3 board for the A1200 and a G4 for the A3/4000's (called the CyberstormG4 IIRC). I was eagerly waiting for them.

[stefcep2]

Wasn't done because it was easier to keep the amiga propriatory- hardware and software- going ppc rather than x86. All the ppc manufacturers feared competing against generic x86 hardware. Performance-wise there was no advantage really, it was about making a buck out of the loyal Amiga community, most of whom were so anti-MS and anti-intel they didn't see any of the BS that was coming from the ppc crowd. Look at Apple, x86 has given them a future.The Amiga went nowhere

[neofree]

AROS is kinda interesting.. Honestly I've never tried it.. I've never wanted to dedicate an entire PC to Amiga, because that seemed like such a waste of a machine. It's not as if I could just create a virtual machine with Windows or Linux running in AROS...

The only thing the Amiga really has left is the philosophy and style... Combined with Commodore's it was the coolest thing going in the EIGHTIES... Nineties they tried to keep up but I could so often see it was behind... This entire decade is a dissapointment if you're holding your breath for the Amiga to wow you with state of the art technology and style...

That's what Commodore and Amiga were... State of the Art, and Style... Today we're certainly not State of the Art.. So it's no longer Commodore/Amiga... It's dead... Sad.. but true..

That said..Long Live Retro-Computing... Even if it just winds up mostly being emu...

Thanks,

Neofree

[SamuraiCrow]

@stefcap2

From Amiga, Inc.'s perspective the move to PPC was supposed to be temporary as they were going to try to take the platform to many different processors with their ill-fated AmigaDE environment.

AROS has done something similar and is still behind both the AmigaOS 4.x and MorphOS systems since as an open-source system, they couldn't get NDAs to access the hardware and driver designs of today's hardware.

AnubisOS may make a better showing since it will be able to use the Linux device drivers available for $0 including OpenGL. Nonetheless, AnubisOS supports PPC as well as Intel and soon, ARM.

As far as Apple having a future with x86, it is only the decline of Windows Vista and the success of the iPod and iPhone that left the door open for them. Also, their BSD underpinnings are quite good and if you have the money, MacOSX already spans PPC, Intel, and on the handhelds, ARM.

I think if Amiga wants to ride the coattails of Apple for a while until a better opportunity arises, LLVM would make a much better cross-platform development technology than Amiga Anywhere ever was. It is also open-source so it can be ported to the entire Amiga platform collection, including the free variants. This will allow new hardware to be used as soon as the compiler toolchain is ported. Thus Amiga will not be stuck with the same hardware forever.

If you detect a bias toward cross-platform development within the Amiga operating system market in my posts, you're only partly right. I think the best is yet to come and none of today's processors offer enough to keep the Amiga going.

[persia]

I'd leave out Anubis, it's just another funny name without even an alpha version. What it may or may not do is a matter of conjecture and navel gazing, all it can currently do is eat your chickens...

[SamuraiCrow]

I think it's supposed to just be a hosted AROS with better drivers. How hard to write can that be?

[giturox]

Development and in particular production of custom hardware is getting more and more expensive. A 2GHz design f.ex. would require careful considerations regarding the PCB-layout and components that require a know-how beyond most hobby-engineers. A softcore processor for a FPGA can never reach the speeds of a hardcore processor, which is also why Xilinx provides one or more hardcore processors for their toprange FPGA's.

I personally think that there already is a reasonable solution: UAE. With the addition of the amazing JIT-compiler the speed is very nice on a modern x86 PC. This enables me to run ImageFX 4.5 at a decent speed, with some operations to all practical intents and purposes being just as fast as some commercial solutions under Windoze. So a very nice solution would be to keep improving UAE, so it supports more and more features of modern hardware (this already happening). This support for new hardware would require extension of the classic M68k AmigaOS. Here there is an opportunity for either Hyperion or AROS.

Why is this useful? When I switch on a device, I don't care if it is x86, PPC, or ARM, nor do I care if it is MacOS or Windoze, or Amiga. What i do care about is my data and the applications that allows me to use my data. Here UAE allows me to run Amiga applications on a multitude of hardware, where I don't have to care about the host operating sytem or host hardware. I doubleclick an icon and the software runs.

Since the M68k support in UAE is made of software, extensions could be made to support 64bit addressing modes. Such an extension would secure the usefulness of the Amiga emulation for years to come. The switch from 16 to 32 bit took a few years, where as the switch from 32 to 64 bit took almost 20 years. And the only real reason for this switch is to a be able to handle large quantities of memory and disk-space.

So the Amiga may have a future as a hosted OS, but only of the development of applications for M68k keeps flowing. As of current the community is splintered among AmigaOs 4.x, MorphOs, AROS and Classic M68k AmigaOS. The only uniting factor seems to be ???? (well i don't know)

PS: I still plan to buy stuff for my classic A1200's (subway) and I also hope to collect enough cash to buy a Sam440 with AmigaOS4.1.

[Piru]

<neo>Déjà vu (http://www.amiga.org/forums/showthread.php?t=36726)</neo>



Not exactly the same as the topic start here, but this thread soon fell into the same patterns. So I'd urge people to take a look at the old thread first before posting more of the same old.

[yssing]

What could be cool, instrad of a new PPC acc.. would be of say. Sam440 was "reshaped" to fit inside an A1200/600/500 case. New modern HW, but definetly keeping the "old" amiga feeling.

[alexh]

Doesn't sound cool to me.

Classic Amiga's are cool.

SAM440 is just a slow, unsupported, obscure platform with nothing of any merit to run on it.

[yssing]

then something else. but atm. its only the SAM440 that is actually being produced.. I don't care if its ourdated and slow, compared to a modern day 3. gig quad core Pentium PC.

And the SAM is still a lot faster than any classic amiga.

[Rob]

Does the vapourware natami have any plans for a ppc board?

That's one feature that won't creep in, though I've read that the specifications for the CPU slot will be released.

[Fats]

neofree wrote:
AROS is kinda interesting.. Honestly I've never tried it.. I've never wanted to dedicate an entire PC to Amiga, because that seemed like such a waste of a machine. It's not as if I could just create a virtual machine with Windows or Linux running in AROS...


But you can do it the other way: running AROS in a virtual machine on Windows or Linux. See VmwAROS (http://www.vmwaros.org) for more information.

greets,
Staf.

[Faerytale]

Why do people buy old 500$+PPC-cards for the amiga?
Because they want to run OS4!?

OS4 runs like a snail on all PPC cards out there.
If you gonna run OS4 you need new hardware.

Windows XP runs on Pentium90Mhz but is useful?

Sp why are people buying PPC for those redicolous Ebay prices?

Classic is better to stay classic. The new OS4 needs new hardware. Thats it!

[coldfire]

They buy those expensive boards because they want to run the most powerful amiga they can. I want the fastest accelerator in my amiga I can get. emulation is like sex while wearing a condom....I just can't seem to get off on it.

coldfire

[RRunner]

They buy those expensive boards because they want to run the most powerful amiga they can. I want the fastest accelerator in my amiga I can get. emulation is like sex while wearing a condom....I just can't seem to get off on it.

coldfire

Hahahaha! :D