MiniMig: One Random CRAZY Thought.

[EDanaII]

@ AJCopland,

You mean something like an FPGA based graphics card that implemented the necessary parts of the Alice and Lisa (for AGA) or Denise and Agnus (for OCS/ECS) to reproduce the Amigas graphics in hardware.

Essentially, yes. :-) But a card designed in FPGA, somebody correct me if I'm wrong, doesn't have to stay in FPGA. Just a thought.

I think that'd probably be harder than doing the MiniMig as you'd effectively be pulling out and exposing the raw interchip interfaces and then trying to put a clean interface over them so that they could be called from the "other" OS over some kind of bus. PCI/PCIe most likely.

Of course. I've never stated otherwise.

Consider, though, that some people on this thread, and elsewhere I'm sure, have been advocating an open-sourced Amiga. What's amusing for me about all this, thanks to Dennis and AROS, is that we already HAVE an open-sourced Amiga. :-) And all I'm doing is advocating one of the directions that it should take.

Ed.

[AndrewBell]

little wrote:
Problem is that would have been a cool when PCs were not fast enough to emulate the amiga. Nowadays it would not add anything you can't already have with winuae. Same would apply to a playstation one emulator on a card, you can already do that in software, why do you need a PCI card also?

There are still advantages to hardware solutions. Having the custon chips in silicon means less for the host processor to do, which in turn means faster 68k emulation.

Plus, the best emulation programmers alive will never get a VGA card to pump out a 15kHz PAL signal. The MiniMig also has real 9 pin joystick/mouse ports. The only way to use Amiga joysticks with UAE is via a Catweasel card. Imagine both products combined into one.
________
BUY GLASS BONGS

[Belial6]

I would think that the best way to do an add in card would be to have a complete MiniMig with all ports passable through the PC's Bus (Be that PCI, ePCI, or whatever)  This way you could use your 9-Pin joysticks, or have your PC route the equivalent commands through the bus so that the MiniMig thinks that they are real 9-pin joysticks.  The same with keyboards and mice.  Give the PC access to a frame buffer on the MiniMig Card, and PC software could route it to a window on the host PC.  This would give you all the benefits of the suggested video card without the problems associated with trying to get hardware banging software to work on non-integrated MiniMig chips.  It could also allow for a Virtual MiniMig bus that would make adding any PC peripheral just (just??? who am I kidding?) a matter of writing drivers.

Of course, I wouldn't even consider buying/building such a card until I had at least one or two stand alone MiniMigs in the house.

[freqmax]

I think what's sought after by the OP is essentially an Amiga graphics card that can be used on modern hardware. There is already a project that may be altered for that purpose:
http://en.wikipedia.org/wiki/Open_Graphics_Project

It even got 2x DVI :-)

@jkonstan:
I guess I was looking for thumb rules. Most Minimig boards will be 2-layer at most.

[little]

In order to insure that the SDRAM works, the answer is complex and may be out of scope for a simple post ... If you really want to know more about this, I can write up more after I finish assembling my MiniMIG1.1

Since nobody answered to this post either people find this knowledge fascinating but completely out of our grasp (like me) or already know about it (I am hoping there are a few in this category, otherwise we will be stuck with 2mb or worse, with a hardware design that overtime becames imposible to implement because those chips can no longer be obtained).

[asian1]

> FPGA and PowerPC:

Project Blackdog: Xilinx Virtex II Pro on USB:

http://www.projectblackdog.com/

Celoxica plan to release Xilinx Virtex II Pro FPGA for AMD Torrenza motherboards. Perhaps similar FPGA product will be available for Intel Geneseo.

"AMD’s Torrenza Initiative Begins to Materialize."

http://www.xbitlabs.com/news/cpu/display/20070924084002.html

Compaq Plan to release server with HTX slots. Perhaps someday PowerPC / Power CPU can use HTX / Hypertanspor socket.

[koaftder]

freqmax wrote:
I think what's sought after by the OP is essentially an Amiga graphics card that can be used on modern hardware. There is already a project that may be altered for that purpose:
http://en.wikipedia.org/wiki/Open_Graphics_Project

It even got 2x DVI :-)

@jkonstan:
I guess I was looking for thumb rules. Most Minimig boards will be 2-layer at most.

Oh man, what a waste of time. gfx cards are soooo cheap these days, whats the point? And AMD is making all the radeon specs OS anyway. Why, oh why, reinvent the 10 dollar wheel?

[freqmax]

koaftder wrote:

freqmax wrote:
I think what's sought after by the OP is essentially an Amiga graphics card that can be used on modern hardware. There is already a project that may be altered for that purpose:
http://en.wikipedia.org/wiki/Open_Graphics_Project

It even got 2x DVI :-)

@jkonstan:
I guess I was looking for thumb rules. Most Minimig boards will be 2-layer at most.

Oh man, what a waste of time. gfx cards are soooo cheap these days, whats the point? And AMD is making all the radeon specs OS anyway. Why, oh why, reinvent the 10 dollar wheel?

That project was born when graphics manufactors refused to give documentation that allowed drivers to be written for free operating systems like FreeBSD, NetBSD, Linux etc..
This is especially true when it comes to 3D usage.

ATI and NVidia binary blob drivers that are supposed to solve this problem for end users are poorly written, and doesn't work alright with software upgrades or non-x86 architectures.

I have heard that Intel and ATI are more documentation friendly at the momement. But NVidia is a particular rotten egg in this department. So NVidia hardware are ditched by many that want unix on their machines because of their refusal to provide documentation to driver writers.

The OGP card also gives the possibility to experiment with new computer graphics hardware algorithms.

[nBit7]

   In order to insure that the SDRAM works, the answer is complex and may be out of scope for a simple post ... If you really want to know more about this, I can write up more after I finish assembling my MiniMIG1.1

Since nobody answered to this post either people find this knowledge fascinating but completely out of our grasp (like me) or already know about it (I am hoping there are a few in this category, otherwise we will be stuck with 2mb or worse, with a hardware design that overtime becames imposible to implement because those chips can no longer be obtained).

Unfortunately Signal Integrity issues are not a simple problem.  At frequencies above 50MHz (actually its the rise times that count) you have to start looking to RF principles.    Every track on the PCB starts looking like a transmission line that has to be properly terminated.  Maxwell's equations and Impedance matching are the order of the day rather than ohms law and thevenin.  This does make SATA and DDR2 a challenge for engineer let alone a hobbyist. Most electronic engineers are still taught nothing of this subject at UNI.  

[EDanaII]

@ freqmax

freqmax wrote:
I think what's sought after by the OP is essentially an Amiga graphics card that can be used on modern hardware. There is already a project that may be altered for that purpose:
Open Graphics Project

Yep, that's essentially what I'm arguing.

Nice link, BTW, I had no idea such a project existed, but I'm not surprised either. :-)

Ed.

[koaftder]

EDanaII wrote:
@ freqmax

freqmax wrote:
I think what's sought after by the OP is essentially an Amiga graphics card that can be used on modern hardware. There is already a project that may be altered for that purpose:
Open Graphics Project

Yep, that's essentially what I'm arguing.

Nice link, BTW, I had no idea such a project existed, but I'm not surprised either. :-)

Ed.

I don't see how having an expansion card that offloads some of the horse power of generating video has an advantage over just doing a software only solution in this case. With either software or hardware, the end goal is to replicate the graphics functionality of the amiga. It doesn't matter if the functionality is hardware or software, the result is the same. In over a decade of evolutionary advances in software emulation we still don't have perfect emulation. The latest hardware emulation, the minimig, is not a perfect emulation of the a500.

I just don't see the advantage of an imperfect and expensive hardware emulation of graphics over an imperfect and free software emulation of graphics...

Hardware solutions make most sense when what you are trying to accomplish is impossible or prohibitively expensive in software.

[EDanaII]

Of course it's expensive, koafter. Welcome to the Wonderful World of Amiga. :-) Even the MiniMig is going to be _relatively_ expensive for what it has to offer.

My point in arguing this has to do with modularity. It has to do with the direction I think this should go. If this takes off, rather than developing an A1200 and AGA -- not that I'm against this -- I simply would like to see it used in a modular fashion that parallels the state of computers today.

Ed.

[Hattig]

Nice idea, the Amiga Hardware on a card, with WinUAE altered to use it, whilst still running 68k emulation itself. Of course you would need to route all the output back into the host computer (audio, graphics, etc) to integrate into its output - unless the card had its own graphics and audio I/Os.

An alternative idea in the opposite direction is to extend the MiniMig hardware implementation to add some chunky graphics modes, and write a Picasso96 driver that uses them. It would only be useful for Workbench, etc, but if you could have some graphics modes like 1280x720, 1920x1080 it would be neat on a modern TV. Probably would need a MiniMig implementation with SDRAM rather than SRAM, to use the extra capacity, or be stuck with 4 colour hi-res modes or divisors of that mode (640x360, 960x540). Also the 68k might struggle to manage that on its own...