Apollo Team announces the Vampire V4

[Amiga1992]

Originally Posted by TuKo

IO headers are meant for that. You could have floppy, serial, etc. It's all up to developpers.

There's 30 pins of I/O headers. Surely you can't have every I/O option an Amiga has on 30 pins.

[Samurai_Crow]

Originally Posted by Akira

There's 30 pins of I/O headers. Surely you can't have every I/O option an Amiga has on 30 pins.

Just use serial connections at many times the clock speed.

[Amiga1992]

And what would that achieve?
The main thing missing here and that I don't think is trivial to add, is the Amiga video port, so genlocks are out of the question. One would also need the rare 23 pin D-sub connector.

[Samurai_Crow]

Real time video grabbing is easier and could be based on the Apollo team's Parasite board prototype.

[Sinphaltimus]

Originally Posted by Akira

And what would that achieve?
The main thing missing here and that I don't think is trivial to add, is the Amiga video port, so genlocks are out of the question. One would also need the rare 23 pin D-sub connector.

These?

https://www.google.com/shopping/prod...C10Q8wIIygIwCA

[daxb]

The question is how the I/O headers work and what is possible. Hardwired? Configurable? Special software driver needed?, aso.

[Amiga1992]

Originally Posted by Samurai_Crow

Real time video grabbing is easier and could be based on the Apollo team's Parasite board prototype.

That isn't the point, the point is being locked between compatibility with legacy peripherals and modernity. This is the "cross" the Vampire is carrying for some reason.
Yes sure, you can do video capture somehow else. But what if I want to use a device I already want? I need the 1200 slave I/O

[Djole501]

Originally Posted by Akira

That isn't the point, the point is being locked between compatibility with legacy peripherals and modernity. This is the "cross" the Vampire is carrying for some reason.
Yes sure, you can do video capture somehow else. But what if I want to use a device I already want? I need the 1200 slave I/O

The board specs have been posted, there is no changing possible anymore. It is what it is, i dont think there are a lot of people still using genlocks on Amiga. You should keep your original Amigas for that.

[Retrofan]

I don't know who is involved in the project yet.

Is Majsta going to make them?

Any other like Kipper or Amiga Kit?

How advanced is the development, do you have an estimated date for a working A1200 one?

[IanP]

Originally Posted by Akira

And what would that achieve?
The main thing missing here and that I don't think is trivial to add, is the Amiga video port, so genlocks are out of the question. One would also need the rare 23 pin D-sub connector.

Genlocks? It's 2017 and video has gone digital, if you want to still use an Amiga genlock from the 1980s or 90s then the right thing to use it with is a classic Amiga, not a brand new FPGA standalone system.

[NovaCoder]

http://amitopia.com/vampire-v4-info-...e-card-itself/

Pics!

[Megol]

Originally Posted by grelbfarlk

One perspective it's now basically something like this:

http://www.terasic.com.tw/cgi-bin/pa...yNo=167&No=941

Sure the V4 has more LE's, but that board is $90. So you're buying €100 worth of hardware with a €250 software license. This doesn't apply to the magic of adding plug in compatibility to the A500/A2000/CDTV. So why not take a cheap off the shelf board and slap the core on it and call it a day. It makes buying a retail copy of Windows seem cheap.

Actually that was the original plan - use a cheap off the shelf board with an adapter board connecting it to the Amiga hardware.

There's a reason that plan didn't work out.

You implications are either dishonest or misguided. Did you note that the FPGA board you listed also listed the manufacturers of the main components - something extremely unusual? Why? Because that board is subventioned by them. They are also manufactured in large series (compared to Vampire boards) and thus have lower overheads and lower component costs. That Altera/Intel use it as a demonstration/development board reasonably means the FPGA cost is much lower than what the Apollo team can even dream about.

The price is listed in US dollars which means it doesn't include sales tax/value added taxes which (unless I'm mistaken) the Vampire prices do. If that is true you should subtract 19-25% from the Vampire price to get something comparable. Paying for the Terasic board doesn't contribute to a continued development of hardware, software and soft core unlike doing the same for the Vampire board.

Now you can of course think the Apollo people should go back to the plan of using off the shelf boards with adapters. Up until fairly recently the team was still looking for people to join, why not offer your services if interested in doing the work? We need more people like Majsta in this world IMHO...

[kolla]

Originally Posted by Megol

The price is listed in US dollars which means it doesn't include sales tax/value added taxes which (unless I'm mistaken) the Vampire prices do.

I doubt it, it certainly was/is not the case for the current Vampire cards, I paid 25% taxes on top of card+shipping. USD 269 + 35 + 76 (25% VAT) + custom service which is like 10 or so? Anywas, close to USD 400 each.

[Amiga1992]

Originally Posted by IanP

Genlocks? It's 2017 and video has gone digital, if you want to still use an Amiga genlock from the 1980s or 90s then the right thing to use it with is a classic Amiga, not a brand new FPGA standalone system.

Yeah, that is exactly my point.
I don't see any advantage in using a 680x0 platform in dong these things. If we are talking "amiga", there's far more modern tools working on PPC setups than there are 68k ones. I feel like tying all this new tech to 680x0 is a bit weird.
But I'll stop now because clearly, this thing is not for me, and I don't want to stir shit up. I just don't see this as a possible "step up" from a classic Amiga, it's another branch progression, like PPC was at its moment.

[Signman]

Originally Posted by kolla

I doubt it, it certainly was/is not the case for the current Vampire cards, I paid 25% taxes on top of card+shipping. USD 269 + 35 + 76 (25% VAT) + custom service which is like 10 or so? Anywas, close to USD 400 each.

Me too. Another $65 US on top.

[Signman]

Originally Posted by Akira

Yeah, that is exactly my point.
I don't see any advantage in using a 680x0 platform in dong these things. If we are talking "amiga", there's far more modern tools working on PPC setups than there are 68k ones. I feel like tying all this new tech to 680x0 is a bit weird.
But I'll stop now because clearly, this thing is not for me, and I don't want to stir shit up. I just don't see this as a possible "step up" from a classic Amiga, it's another branch progression, like PPC was at its moment.

Well if this isn't up, I would quit looking in that direction.
This is a very old platform and it is all that one can expect.

[Ffin]

Originally Posted by Akira

That isn't the point, the point is being locked between compatibility with legacy peripherals and modernity. This is the "cross" the Vampire is carrying for some reason.
Yes sure, you can do video capture somehow else. But what if I want to use a device I already want? I need the 1200 slave I/O

Look, you clearly don't want one, just leave the thread already!

[idrougge]

Originally Posted by jbenam

No Amiga is an Amiga without a floppy drive!

Well, it's not an Amiga…

[grelbfarlk]

Originally Posted by Megol

Actually that was the original plan - use a cheap off the shelf board with an adapter board connecting it to the Amiga hardware.

There's a reason that plan didn't work out.

You implications are either dishonest or misguided. Did you note that the FPGA board you listed also listed the manufacturers of the main components - something extremely unusual? Why? Because that board is subventioned by them. They are also manufactured in large series (compared to Vampire boards) and thus have lower overheads and lower component costs. That Altera/Intel use it as a demonstration/development board reasonably means the FPGA cost is much lower than what the Apollo team can even dream about.

The price is listed in US dollars which means it doesn't include sales tax/value added taxes which (unless I'm mistaken) the Vampire prices do. If that is true you should subtract 19-25% from the Vampire price to get something comparable. Paying for the Terasic board doesn't contribute to a continued development of hardware, software and soft core unlike doing the same for the Vampire board.

Now you can of course think the Apollo people should go back to the plan of using off the shelf boards with adapters. Up until fairly recently the team was still looking for people to join, why not offer your services if interested in doing the work? We need more people like Majsta in this world IMHO...

I agree there is a reason that plan didn't work out. It was actually a member of the Apollo team that pointed out the Terasic board to me originally-not that this particular board is really the one to get anyway with half of the core being a 925MHz ARM.

I agree with everything you've said, a common off the shelf board would lower prices and have a common piece of hardware to swap out if you suspected you had problems with your board. I wonder what possibly could be the motivation of building something very similar and compatible to a cheap, easily attainable, off the shelf board and selling it for 3-4 times as much. I realize low volumes=higher prices and having professional assembly also costs more, those are all arguments for using a COTS board.

Last I heard about two years ago the original board (an Arrow Be Micro(?) )they were basing the V500 off of worked, but Chris just hadn't gotten the memory controller working. At which point the idea of a common off the shelf board evaporated as Majsta had finished building his initial prototypes.

As I said in my previous post to this, it is unknown what the price and performance points are so this is all just idle speculation until that stuff is known.

[matthey]

Originally Posted by grelbfarlk

I agree with everything you've said, a common off the shelf board would lower prices and have a common piece of hardware to swap out if you suspected you had problems with your board. I wonder what possibly could be the motivation of building something very similar and compatible to a cheap, easily attainable, off the shelf board and selling it for 3-4 times as much. I realize low volumes=higher prices and having professional assembly also costs more, those are all arguments for using a COTS board.

There are both advantages and disadvantages of a off the shelf board.

+ cost
- the board is probably not exactly what is wanted
- the board could be changed or discontinued at any time

Originally Posted by grelbfarlk

Last I heard about two years ago the original board (an Arrow Be Micro(?) )they were basing the V500 off of worked, but Chris just hadn't gotten the memory controller working. At which point the idea of a common off the shelf board evaporated as Majsta had finished building his initial prototypes.

Yes. The Sandwich accelerator original plan was to use an Arrow BeMicro CV board. It would be better if 2 boards were not needed but the Arrow board and most off the shelf FPGA boards do not have HDMI.

There is a FPGA board which does have HDMI. The Turtle board is meant to be an FPGA version of the Raspberry Pi.

http://www.cnx-software.com/2017/03/...uperh-sh2-soc/

It may be better suited as a FPGA Arcade upgrade as it uses a Xilinx FPGA but it would technically be possible to put an Apollo Core in it. They created their own board to get a larger FPGA for the open core (SuperH) J-Core CPU project. The SuperH was a simplified hybrid RISC derivative of the 68k developed by Hitachi after they licensed (and produced) the 68000 from Motorola (there was a license dispute). The patents are expiring for SuperH much as they did for the 68k CPU so they are using the ISA for a open core. The guy behind the project is Jeff Dionne who is the uClinux co-creator which began for a CPU32. He is a 68k fan which is part of the reason why he chose the SuperH (also because of code density even though I tried to show him how much the 68k can beat SuperH). He is making various versions which he hopes will scale for embedded applications. These guys try to abstract the current FPGA details and have tools which automatically prepare for an ASIC of which they plan on doing several. It's pretty much the opposite philosophy of a certain other FPGA core which is designed for FPGA only. Another missed opportunity for the 68k and Amiga. RIP.