FPGA Amiga

[Chucky]

yeah I know that we will never use it as a main platofrm.  maybe 2-3 test-things.

BUT! who WILL use AMMX if noone actuallty does software USING it?

and later with the AGA thing.  WHO will use the new stuff introduced?   it is not like there is any API using it anyway.   so it feels like doing stuff just to be able to say "HEY WE GOT COOL STUFF"  but..  noone can/will use it.  instead of implementing the stuff we had. FIRST  and then put in the new stuff?  it is so strange and SAD!  as we do NEED a new good FPGA replacement for our beloved machines.. it is getting harder for me to get hold of 060 cpus etc. so this would be needed.

sadly the wrong path.  well luckly there are other FPGA solutions.

Talk to gunnar etc. well  I haven't done it.  as I feel that there is absolutley not idea of doing that, being reading how he answers others with suggestions. he answers them as trump on a day in a bad mood.

Sad.  as the hardware is awesome.

[grond]

The order in which things are done does not depend on the date the reimplemented feature was first publically available (like doing MMU and FPU before AMMX because the 68k MMU and FPU predate MMX) but in which it makes sense from a technical point of view. E.g. if you create a new vector-FPU first, you can then use it to implement a 68k-compatible scalar FPU on top of it. So in this example you get the "who will ever use it"-FPU first and then the 68k-compatible.  If you first did a scalar FPU, you would have to scrap it when you start working on the vector-FPU. The same goes for AMMX and the 64bit mode: a part of these is the basic infrastructure on which the FPU and some bitfield instructions (which need an ALU wider than 32 bits!) are based.  This has been pointed out many times before but the same arguments repeat like the aforementioned broken record...

[Chucky]

the same and it is pointed out like a broken record that. WHO will do the programming?   just think of it: most sources of software is lost anyway.  this means software needs to be written from scratch.   will this happen? most likly not..

so why even bother adding new stuff instead of perserving what we got?

[grond]

so why even bother adding new stuff instead of perserving what we got?

Um, see above: because it is fun. Recreate what was already there: boring. Do an entirely new 68k CPU that implements what likely would have happened if Motorola had continued to develop the 68k CPU family: fun.

And even if nobody else uses new features, the Apollo Team members are using it.

[kolla]

To say that the Apollo Core mimics what would have happened if Motorola had continued is just nonsense. Motorola's largest customer base for 68k was telecommunications, aerospace and embedded.

[Niding]

the same and it is pointed out like a broken record that. WHO will do the programming?   just think of it: most sources of software is lost anyway.  this means software needs to be written from scratch.   will this happen? most likly not..

so why even bother adding new stuff instead of perserving what we got?

There are plenty of options for pure legacy hardware, like Whicher 500, ACA, TerribleFire, Ebay/AmiBay hardware etc. If thats your cup of tea, look in that direction.

Why should Apollo Core do the same? Given the relative decent backward compability (and gradually improving), you can even use Apollo and ignore the new features.
If they lack features, then you have the other alternatives.

Britelite has repeatedly given his reservations with the direction of the Apollo Core, but he isnt rejecting it in a pissed manner. His main concern is the possibility that a coder (demoscene or otherwise) develop keeping the legacy timings as the foundation, but the new features MIGHT have unforseen consequences.
Thats a fair critique/concern with regards to the Apollo Core. Wether or not its a valid concern as the compability increases, thats for the developers to decide. People release more and more tests of old demos that work just fine on the 2.7 and 3 Beta cores, showing quite decent improvements compared to 6-8 months back when Kioa was a slideshow with graphical bugs. Now it runs smoothly.

Work in progress and all that.

[psxphill]

Where did I write that?

You wrote it here:

http://www.amiga.org/forums/showthread.php?p=834828#post834828

If you read carefully what I write and not interpret something in it, I have never written anywhere that Vampire can be compared to up-to-date standard hardware.

I did read what you said carefully. If you think A doesn't like B because it's not C then you're implying you think B is C. I think you're completely wrong though, there are plenty of 4k demos on much more powerful systems released.

Demo coders code for plain A500 or A1200, not for expanded big boxes or new highend 68k hardware. It would be not fun to them

Actually demo coders are consistently inconsistent about what hardware they code for, for a long time there were demos released that would only run full frame rate on WinUAE on a fast PC. But you're right about Apollo not being fun. I'm hoping to get an Ultimate 64 soon, it's less powerful as FPGA systems go but Gideon is a great guy.

When has ANYTHING Amiga NOT been like joining a cult???  Lol :roflmao:

If commodore were trying to run a cult then they failed miserably, but then they would.

[kolla]

Why should Apollo Core do the same?

Because they want Apollo Core to be the de-facto "base line" for Amiga hardware.

[kolla]

Britelite has repeatedly given his reservations with the direction of the Apollo Core, but he isnt rejecting it in a pissed manner.

And the other way around? What does Gunnar have to say about the concerns and reservations of developers, demo coders etc. because that's what really matters.

[trekiej]

Why was Natami a failure?

[TrashyMG]

And the other way around? What does Gunnar have to say about the concerns and reservations of developers, demo coders etc. because that's what really matters.

Can you list these developers that are so concerned?

[AJCopland]

Why was Natami a failure?

bearing in mind that all of this is my personal recollection of things and I was only a little involved with some software and ideas etc...

Because of a number of reasons. Thomas Hirsch felt it was time to announce NatAmi and take on some help after MiniMig appeared. However it was a lot of pressure upon what had been a fun hobby project for him and I think that he lost interest.

Also he kept it all very close to his chest so people involved in NatAmi never got to help out with the hardware except for investigating an FPGA based CPU that they called the N68070/N68050 at the time. That was the only group part of the project. NatAmi itself was 100% Thomas.

As NatAmi got closer to working it conversely also put more pressure on Thomas and he decided to go quiet for a time. It looks like that's now become permanent sadly. I liked what he'd done and hoped that he'd eventually release it with a real 68060.

The people who had worked on the N68070 CPU didn't want it to disappear, they liked what they'd done. So it became the Apollo Core.

Majasta had released his V1 accelerator for A600 using the tg68k.c core but it had various issues.

The both teamed up and between them built the newer Vampire A600 accelerators with the Apollo Core replacing the tg68k which is where we are today.

TLDR; I think the pressure from the community and it's usual "supportive" behaviour drove away Thomas's interest in finishing the Natami.

Andy

[moogaloonie]

Really, did people complain when 030/040 accelerators or new Amigas were introduced into the market?  The same arguments can be made - "we are fracturing our userbase!"

Uhhh yeah, there actually was some of that.  The accelerators were much more common in the US for 3D rendering than in the other markets where the Amiga was a dominant games platform, with said games usually being hard coded for the base 68k.  I even remember a game that would have required I remove my internal memory expansion just to run.  And when AGA was released a lot of development continued to focus on OCS/ECS because of the larger installed base.  Western users' requests to C= for MMUs and FPUs as standard were met with groans from European users who merely wanted the cost of the base machine to come down even further. Encouraging OS compliant software and RTG was always an attempt to keep the platform from fracturing along the various spec boundaries. Fear of fracturing the platform could be partially blamed for the long delay between AGA's completion and its eventual release or might have even factored in the reluctance and eventual failure to make the AT&T DSP board a commercial product. But I certainly do not remember there ever being a time when a potentially significant improvement to the Amiga's capabilities was met with an universally warm reception, whether it was an OS revision, a CPU evolution or even a display enhancement.

[trekiej]

Sorry to hear about Natami.

[psxphill]

But I certainly do not remember there ever being a time when a potentially significant improvement to the Amiga's capabilities was met with an universally warm reception, whether it was an OS revision, a CPU evolution or even a display enhancement.

commodore's marketing as woeful as it was, is better than gunnar.

[kolla]

groans from European users

I believe you here mean British users. Just consider where most CPU boards and other high end hardware for Amiga were made, where Amiga clones were made, where most productivity software was made etc.