What Amiga product do *YOU* would want. Please read.

[takemehomegrandma]

@ redrumloa

Are you actually thinking of starting a new company to develop new hardware for the "classic" Amiga market? Wow! Are you thinking of living on this, or will it be more "a work of love"?

A bold mission nevertheless, since the few classics still not in the closet are probably dying in an increasing pase.

And just out of curiousity, who are "we"?

redrumloa wrote:

I'll start first.
-> I'd like a new PPC accelerator for classic Amigas at a reasonable price.

We are actually evaluating it, but to be completely honest it's not likely to happen. The cost of R&D would probably never get recouped, even if OS4 became a smash hit on classic boxes:-(

It could perhaps be made "the Eyetech" way; buying a batch of pre-made boards from outside the community and sell them here? Wasn't the SharkPPC (at least the first "versions") very similar to a certain Mac PCI based CPU upgrade? The software part might be a nightmare though ...

And now YOU. What would YOU like to see? We are thinking mainly hardwarewise. Would you like to see a cheaper RGB->VGA adapter? An extinct product ressurected? What? I'd really like to hear alot of opinions. This is mainly aimed at hardware, not software.

The Amiga was traditionally very strong on Video/TV production. It worked both on PAL and NTSC, and in several resolutions. The custom graphics chips could be synced from an external source, so it was very popular to use it with genlocks. The custom hardware made things very smooth and flicker free, without heavy CPU usage.

Both the BoXeR and the Commodore One uses/would use FPGA's to accomplish solutions very similar to the Amiga custom chips, with coppers, blitters, hardware sprites, etc (of course, the C-1 also has to take the C64 in account).

Both the Pegasos and the A1 hardware goes the mainstream PC/Mac way, with all standard components. That is great for a cost/price POV, but it does not make the hardware stand out.

I would like to see a new "Amiga" custom chipset. It could be made with FPGA's, fast and easy (ehrm, sort of at least). The features of the Xilinx Virtex-II Pro FPGA's looks kind of interesting IMO, as an option they can even include PPC CPU cores on the chip (up to four of them).

Forget about advanced 3D, there is no way (and no reason) to compete with the 3D chips developers. This is about powerful 2D with consumer TV's and video production in mind, with improved custom chips features. Perhaps some basic 3D functionality that can be used for 2D planes manipulation (for zooming, twisting, rotating, picture in picture, etc). It should be flexible and powerful in screen resolutions and modes (all VGA/SVGA resolutions/modes included, all TV modes, and it should go as far as HDTV), it should use 32 bit graphic (alpha channel in hardware), it should offer multiple displays, it should have a blitter, a copper, "windowing mode" (smooth HW scrolling of a small view of a much bigger screen), etc. It should offer HW antialiasing. It should offer at least one video in channel, with built in genlocking functionality.

As a start you could produce a graphic card with this chipset (or with sound too, a "media" card) in PCI/AGP. This could be used outside the Amiga world too, in PC's and Mac's for video editing/broadcast solutions. It should offer high quality video output, multiple displays, lots of connector options (including component video connectors). If the chipset has a built in CPU (like the virtex), it could run an OS internally (AROS with custom low level, hardware banging drivers in asm?) which handles everything internally "the Amiga way", independent of the OS on the main board.

This chipset could also be used in STB's and other consumer electronics/home entertainment products.

A new "Amiga" motherboard with a new custom chipset would be great too. It should be PPC of course, and it would also have PCI/AGP for expansions (and 3D cards for the ones who wants one of these).

But I'm just dreaming away here! :-D

It will be interesting to see what (if anything) may come out of your plans.

[takemehomegrandma]

redrumloa wrote:

since the few classics still not in the closet are probably dying in an increasing pase.

This is our thoughts. Why are the classics getting neglected? The is no good new product out there, or the price is unreasonable.

Well, the classic Amigas are dying of age one after another. No new classic amigas has been produced for ages. Perhaps this is where to start, producing a new "Amiga"? It will desperately need updated specifications though. Perhaps the old BoXeR design could be acquired to get a flying start? ;-)

[takemehomegrandma]

downix wrote:

Forget about advanced 3D, there is no way (and no reason) to compete with the 3D chips developers.

Au Contraire.

3D chip vendors have stopped innovating, and the quality of products is starting to show.  My now 4-year old Kyro II renders scenes that look better than my brand-new Radeon.

The 3D chip developers are aiming primarly for the gaming industry now, and they are in die hard competition. It's a space race (and their budgets are closing in to the budgets used by the former Soviet Union and USA in that race ;-)).

Surely they (both, there are only two now) have a roadmap for a couple of years ahead. But innovation is not only about technology. You have to make it profitable too, to sustain in the cut throat competition and secure future funds for development. It's a race. Business innovation has been the key here for a while IMO.

You can also license a core, such as the PowerVR or 3DLabs to include in such a chipset.  So you get a brand-new chipset while not the R&D effort needed to develop a 3D.

So the best of both worlds.

That would be the way to go! :-)

[takemehomegrandma]

JoannaK wrote:
Takemehomegrandma:

Unfortunately Xilinx Virtex chips costs a way more than existing GFX chips do.. So making Gfx using Virtex woudl be not only technically a lot to ask, but also economical suiside..

I figured that. I know that you have looked into it (and I haven't). :-)

I have no idea how these things works, but shouldn't it be possible to use these FPGA's as a small scale prototype, and then move ahead and go to a real "silicon producer" in the far east and place an order? Yes/No? AFAIK, even this IBM PPC core is available for licence for custom projects (like in the case of this Xilinx product). That would of course require a large scale production run (which in turn might require a lot of different products based on this component, and/or external partners/OEM customers), and budget far beyond a "garage company", but even a garage company can secure some VC funds if they manage to present their ideas in the right way to the right people. After all, both Microsoft and Apple started out this way (more or less in a garage)!

I don't know, IF it would be possible to use on of those as a replacement to whole machine.. afterall.. it has more CPU power than any Classix Turbo... :-D

That's a cool idea! :-)

The CPU part won't be *that* powerful, but what are custom chips for? ;-)

[takemehomegrandma]

Turambar wrote:
How about an RTG amiga (like draco) on a mini-itx or smaller board with on board graphics and sound.

techiebk wrote:
I do not know if it would be possible oraffordable, but a new mobo would be nice. Something with an 060 and 4 or 5 PCI slots. I would think that this would eliminate a lot of the bottlenecks that you get with a1200 - 060 accelerator combos.

My price point would be approx. $300USD. For that price I would like to see a mobo with an 060, four PCI  slots, three DIMM slots, and basic connectors (serial, parallel, IDE).  I do not know if it would be necessaru to have all the classic vid & sound chips, perhaps a custom boot disk would allow you to go straight to a vid card.

Aren't you guys really thinking slightly in terms of a Pegasos with a 68060 instead of a PPC? So a 060 card for the pegasos would be the product you want? That would of course limit you to the old Amiga3.x OS's forever (with heavy patches), it's performance would be inferiour, and you would not be able to use the improved MorphOS/OS4 ...

[takemehomegrandma]

takemehomegrandma wrote:

downix wrote:

Forget about advanced 3D, there is no way (and no reason) to compete with the 3D chips developers.

Au Contraire.

3D chip vendors have stopped innovating, and the quality of products is starting to show.  My now 4-year old Kyro II renders scenes that look better than my brand-new Radeon.

The 3D chip developers are aiming primarly for the gaming industry now, and they are in die hard competition. It's a space race (and their budgets are closing in to the budgets used by the former Soviet Union and USA in that race ;-)).

Surely they (both, there are only two now) have a roadmap for a couple of years ahead. But innovation is not only about technology. You have to make it profitable too, to sustain in the cut throat competition and secure future funds for development. It's a race. Business innovation has been the key here for a while IMO.

An addition: It's not only about 3D hardware, it's also about software. Today, Microsoft is a key here. They are setting the standards, and actually, the functionality of the current top of the line gaming hardware is ahead of the software support. The software side needs to catch up. In the mean time, the chipset manufacturers focus their efforts of gaining high scores in various performance benchmark tests. That's what it's all about for now (until the software actually starts using the new functions in the hardware), big numbers from tests for marketing purposes.