Jump to content
IGNORED

Amiga on a chip FPGA by Jeri Ellsworth


Math You

Recommended Posts

I just found this video about an Amiga FPGA 'Computer on a chip'. It was designed by Jeri Ellesworth who designed the Commodore 'Plug & Play' TV joystick game.

 

http://www.youtube.com/watch?v=5uaDzF99a80

 

Interesting thing about the Amiga is that the faster ones could also emulate the Atari 800 and Atari ST using software (and probably a lot of other things aswell).

 

I really hope this gets made one day. It would be a fitting tribute to Jay Miners work.

  • Like 1
Link to comment
Share on other sites

For what it is, the modern day Amiga evolution is just too expensive.

 

An FPGA solution allowing a setup for under $100 or so for legacy support e.g. 68030 ~ 40 MHz and AGA would have been great.

 

That won't happen until FPGA costs come down. Small FPGAs, like 500K gates or less (Xilinx Spartan 3, Altera Cyclone II/III) are not too bad at about $25 to $35 per FPGA, but when you start getting in to the 1M gate devices, or the newer faster FPGAs, the costs start to rise exponentially.

 

A 32-bit chip like to 030 requires a lot of gates, and the other custom ICs in the Amiga would also require a ton of gates. I don't even know if I would try it in an FPGA smaller than 1M gates. Also, the dedicated ICs in the Amiga had their own local RAM, which for the FPGA means multiple wide address and data busses, which means and FPGA with lots of I/O, which means a fine-pitch BGA package, which means 3/3mill trace/space, which means minimum 4-layer boards and complicated layout, which means more assembly costs and fewer options for having small quantities of boards made, etc. It goes on and on, and is not as simple as saying "that FPGA is big enough and only costs $45 at Digikey..."

 

Low demand hobby projects are always going to cost more and cannot be compared to "consumer electronics" that are being made in batches of 50,000 over a 5,000,000 production run. Sometimes I am absolutely amazed at the cost (meaning how "cheap") some electronic products cost, especially when I start thinking about what it would cost me to produce the same product. Hell, as a hobbyist, you can't even get a 2"x2" circuit board made for less than $15, and compared to some electronic gadgets I have bought lately, that circuit board alone blows the whole budget!

 

I guess my point is, you can't be a classic computer hobbyist and be "cheap". The computers are old and limited and the gear costs money. If hobbyists really want to merge new technology with the classic computer stuff, they have to be willing to pay for it otherwise no one is going to do it. No one has the time or money to do FPGA design, development, production, etc. for free or "on the cheap".

 

I would estimate that, for one person to reproduce an Amiga 4000 on an FPGA (working on just the HDL via a devboard) would take at *least* a year of dedicated "hobby hours", i.e. time in the evenings and weekends. And that is if they are really good at HDL and not just starting out. Once that is done, at least another 6 months to get an "off devboard" working prototype, and a final 6 months to set up a small production run.

 

So, assuming the HDL requires a $80 FPGA alone, plus board and production costs pushing the bill of materials to $150, what would the hobbyist market be willing to pay that person for 2 years of their life? IMO, anyone who makes something like that should get at least double the materials cost, so in this case selling for $300 would not be unreasonable, but my feeling is people would balk at that price (even though it is fair). We are disillusioned with the cheap cost of modern throw-away electronics and fail to realize that those economics cannot be applied to everything, *especially* classic computers and hobby electronics.

 

I think Jeri probably stopped due to economics. If she was being paid by a company, they probably realized the market was too small or unwilling to pay a price that would let the company make a profit. Or simply the requirements were too high, or the project too large for one person. You have to remember, all these calssic computers were built by *teams* of very smart people, and for one person to sit down and reproduce that product from datasheets is a monumental task. IMO the Amiga is border line as being too big for one person.

 

Well, that turn into quite a rant. Sorry for letting my opinions run away with me.

  • Like 7
Link to comment
Share on other sites

The MCC-216 http://arcaderetroga...param=cid%3D%26 can do it for $150 and it can be an Amiga, C64, Apple II and Atari 2600.

The MCC-216 only emulates the 68000. I could certainly do that at a higher MHz but emulating a 68030 & math coprocessor would take a lot more real estate for added instructions, cache memory, etc...

Whether or not it would still fit on the MCC-216 I don't know.

Minimig will run on a DE1 development board.

Link to comment
Share on other sites

The MCC-216 http://arcaderetroga...param=cid%3D%26 can do it for $150 and it can be an Amiga, C64, Apple II and Atari 2600.

The MCC-216 only emulates the 68000. I could certainly do that at a higher MHz but emulating a 68030 & math coprocessor would take a lot more real estate for added instructions, cache memory, etc...

Whether or not it would still fit on the MCC-216 I don't know.

Minimig will run on a DE1 development board.

 

Or a Turbo Chameleon... =-)

Link to comment
Share on other sites

I'm selling my MCC-216 and plan to get a Turbo Chameleon for myself. Both products have beta support for Amiga but the Turbo has read/write support on disk images.

Did you ever run Sysinfo speed test on it? I always wondered how fast is it.

Can you have OS3.1 on it?

Link to comment
Share on other sites

  • 3 years later...

Math You wrote:

 

> I think Jeri's version was going to be an official Commodore product with a similar price tag as a Plug & Play TV game.

 

Unknown about the price tag but Mammoth Toys turned down the prototype.

> It's a shame it never got made.

 

Very true. :(

 

FCUG celebrating 33 years,

Robert Bernardo

Fresno Commodore User Group

http://www.dickestel.com/fcug.htm

July 18-19 Commodore Vegas Expo v11 -

http://www.portcommodore.com/commvex

Link to comment
Share on other sites

matthew180 wrote:

 

> I think Jeri probably stopped due to economics.

 

Mammoth Toys stopped the project.

 

> If she was being paid by a company, they probably realized the market was too small or unwilling to pay a price that would

> let the company make a profit.

 

The market was there, but there were other factors involved, too.

 

> Or simply the requirements were too high, or the project too large for one person. You have to remember, all these calssic

> computers were built by *teams* of very smart people, and for one person to sit down and reproduce that product from

> datasheets is a monumental task. IMO the Amiga is border line as being too big for one person.

 

As shown by her Amiga DTV prototype, it could be done by one person.

 

FCUG celebrating 33 years,

Robert Bernardo

Fresno Commodore User Group

http://www.dickestel.com/fcug.htm

July 18-19 Commodore Vegas Expo v11 -

http://www.portcommodore.com/commvex

Edited by RobertB
Link to comment
Share on other sites

  • 3 months later...
  • 2 years later...

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...