AGA integration in Apollo core : work in progress

[Amiga1992]

Originally Posted by matthey

This "compatibility" is the reason why you and everyone is here though. It is the common ground which keeps the Amiga community together. Lose a portion of the compatibility and you lose a portion of the Amiga community.

I am here because I love 68k Amiga and I have plenty of them that work just fine. I don't need a new compatible machine. I have enough Amigas. Future reliability is mentioned? I can maybe see that, but that doesn't seem to be Apollo or Vampire. So why force them to? I thought there was also Minimig and FPGA Arcade with aims nearer to compatibility I believe.

So a project started a way and was veered another way. So what? Make your own if you are so affected. Support a project more suited to your vision and goals. WHat's teh point of destructive comments!

Vampire I see now as a chance to do something new. You say "lose a portion of the compatibility and you lose a portion of the Amiga community"? Isn't that already what "Amiga NG" is? All the PPC stuff? Again, keeping the thought that PPC Amiga, Amiga NG, whatchamacallit, is all part of the same word as 68K Amiga or even Vampire future, is ridiculous. They are not. They are not compatible, they don't crossover, people interested in one are usually not interested in the other.

[matthey]

Originally Posted by Akira

I am here because I love 68k Amiga and I have plenty of them that work just fine. I don't need a new compatible machine. I have enough Amigas. Future reliability is mentioned? I can maybe see that, but that doesn't seem to be Apollo or Vampire. So why force them to? I thought there was also Minimig and FPGA Arcade with aims nearer to compatibility I believe.

The Apollo Core has the wrong future (hyper optimized for affordable FPGA today only). The Minimig, FPGA Arcade and Mist have no future (they are retro only with no upgrade path).

Originally Posted by Akira

So a project started a way and was veered another way. So what? Make your own if you are so affected. Support a project more suited to your vision and goals. WHat's teh point of destructive comments!

My comments are not destructive but sometimes the truth is painful. I was contacted by Gunnar of the Apollo Team and asked to take part in his project. We at one time had a somewhat similar vision but Gunnar's vision changed. He became obsessed with performance in an affordable FPGA. I learned enough to be able to see the potential of the 68k and his mistakes. I wish there was a project to take us all the way instead of half the way but most of the Amiga community doesn't even understand.

Originally Posted by Akira

Vampire I see now as a chance to do something new. You say "lose a portion of the compatibility and you lose a portion of the Amiga community"? Isn't that already what "Amiga NG" is? All the PPC stuff? Again, keeping the thought that PPC Amiga, Amiga NG, whatchamacallit, is all part of the same word as 68K Amiga or even Vampire future, is ridiculous. They are not. They are not compatible, they don't crossover, people interested in one are usually not interested in the other.

At one time, further developing the 68k would have been very difficult and expensive. PPC was one of the better choices then as it was big endian and supported as an upgrade path by Motorola. AmigaOS 4 and MorphOS didn't change that much when they were ported which is why they have most of the same limitations as AmigaOS 3. They did lose part of the Amiga user base as they became less compatible (they are both decently compatible so far with 68k AmigaOS friendly executables). They can drop the limitations and move further from their foundation as they port to ARM or x86_64 trying to replace users they lose along the way or they could also come back to the 68k and develop it properly and end the Amiga fragmentation. I am considering the PPC AmigaOS as part of the Amiga family but it is not as compatible as I like and PPC is unattractive today. Amiga users are voting as customers are choosing the Vampire over the NG systems. Does this mean Gunnar has it all right and I'm wrong? Not at all. He has no clue how many customers he is missing out on if he had an FPU, backward AmigaOS compatibility which didn't need patching, better compiler support, better documentation, etc. He has no clue how popular a standalone Amiga that costs half as much as the Vampire accelerator would be either but that is highly unlikely to happen with Gunnar's plan for the future. Be warned if you were hoping for an FPGA upgrade path and buy the Vampire for what it offers today only.

[E-Penguin]

Oh no, not again.

Can I just say "yay for AGA on the Vampire!" and be grateful for the work done by the Apollo team?

Optimising for cheap FPGA is perfectly reasonable when one is building a card with a cheap FPGA.

I seriously don't get all the hate. I bought one, it is amazing as it is. If something better comes along, I'll buy that. The Vampire is an amazing achievement but unfortunately it seems that the (a lot of the) Amiga scene is more about bitching than appreciation.

There is no way in hell i could upgrade my a4000 affordably to even come close to how great my A600 is with a vampire. AGA integration and 15khz sans scandoubler or an obscure LCD is icing on the cake. I now actually use an Amiga daily rather than sit them around like a dusty museum.

[spacey]

this looks fantastic - can't wait!

[Predseda]

[Sinphaltimus]

More AGA video Demos please. Applications next.

[OlafSch]

@matthey

we know that you do not like gunnars vision because you keep telling it for months. Gunnar propably not even reads the different forums so he cannot read what you post here and even if he has decided what he wants to do so he will not change direction because you do not like it. The only effect you have that you annoy people and (in worst case) harms vampire/apollo because people perhaps not buy it because of your downtalking, but that is not very propable because too many already have and use it and are happy and even write that on forum discussions. I cannot understand why you do it, be bored, waste of time?

And BTW "truth" is relative, everybody has own set of truths. Your "truth" is not absolute, only dictators or certain politicians own "truth" as absolute but I assume you are neither dictator nor politician even if you sound sometimes like one :-)

[OlafSch]

Originally Posted by Akira

Yeah I get that, it has heritage planted on the Amiga of course, but I think I am ready to see it get life of its own before the Amiga drowns it with it .
Just by recent developments, like it overriding the chipset and not using the Amiga's videoport, it's getting more and more "out of the womb", and I can't wait for that to happen. The nature of the project has taken me from really wanting a Vampire inside my A600 to totally uninterested in using a Vampire with Apollo core inside any Amiga I have.

if your only interest is a accellerator with some extra ram then vampire is not for you

[grond]

Originally Posted by OlafSch

if your only interest is a accellerator with some extra ram then vampire is not for you

If I understood him right, he is now interested in the Vampire stand-alone.

[Predseda]

I really want a Vampire for ONE OF MY A1200s. So beside a vintage accelerator powered classic Amiga I want to have another with FPGA inside. They easily can live side by side and I am looking forward to it.

[E-Penguin]

Originally Posted by Predseda

beside a vintage accelerator powered classic Amiga I want to have another with FPGA inside. They easily can live side by side and I am looking forward to it.

This ^^

[daxb]

I think the most important problem is that people don`t understand or can`t. So it is easy to say "yay". At least some or most people with knowledge about Amiga understand.

Unfortunately, matthey is right. At least if it comes to the current state of Apollo and I fear it will not change in the near future. There is no hate. Just constructive criticism.

Topic:
Because it makes absolute sense and rumors say that there will be a dis- enable function for SAGA, it gets one. On the other hand it is possible that it will not come. Let us surprise.

Perhaps. But it reminds me of coworkers who got so hung up on the 'correct' way to develop projects that nothing ever happened and ultimately the whole business failed.

[Lord Aga]

Originally Posted by matthey

Because you are supposed to accept whatever the Apollo Team (Gunnar) decides is best for you.

Are they busting people's kneecaps if they don't accept it ?

[Amiga1992]

Originally Posted by OlafSch

if your only interest is a accellerator with some extra ram then vampire is not for you

Originally Posted by grond

If I understood him right, he is now interested in the Vampire stand-alone.

Right. I am not interested in the Vampire as an accelerator. I am interested in its future as something better and standalone.

Anyway, Matthey, it is impossible to have a chat with you, really.
Just be sure that your "truth" is also not THE truth, and by acting that way, you are no better than the Gunnar you describe and criticize.

To clarify, I am not part of Apollo team, I don't even know who Gunnar is or what he said. But it's really annoying to keep seeing these discussions with basically the same approach that is being criticized.

And the cherry on top is saying Amiga users "do not know/do not understand". Sure, we might not know about processor design or whatever, we don't have to, but we sure know what we want as the end consumers of whatever it is being produced. If you don't care about what the users want, then don't release stuff out to them. Keep it to yourself and make your dreams come true for your own self-satisfaction.

[Sinphaltimus]

I like dreaming of the day mine gets delivered. Watching progress videos. More AGA videos please.

They have me hook line and sinker. I haven't seen anything I didn't like. No one is posting videos demonstrating why I wouldn't have wanted a vampire board. Every video I see reinforces my interest. More please.

[Overflow]

[ Show youtube player ]

Amikit

[ Show youtube player ]

[Sinphaltimus]

Yay!

[modrobert]

Originally Posted by matthey

The Apollo Core has the wrong future (hyper optimized for affordable FPGA today only). The Minimig, FPGA Arcade and Mist have no future (they are retro only with no upgrade path).

Think I understand what you are getting at, correct me if I'm wrong.

By avoiding FPGA vendor primitives and focusing on implementing all in plain VHDL/Verilog you open the door to future ASIC masks which potentially can give at least ten times the performance of the FPGA used now.

Then we are talking GHz range in performance, having 68k up there with modern PPC in performance, perhaps even surpass. I can relate to this dream and understand your concerns.