New Amiga motherboards from Jens

[matthey]

Originally Posted by kolla

So the 68020 softcores used on various other FPGA systems are superior then, since they can actually run software such as OS3.9?

Superior for what purpose (68020 compatibility, high performance, AmigaOS 3.9 support)?

Originally Posted by traal

It's tough to be taken seriously in this crowd, isn't it?

Crowds which are knowledgeable and well informed are always going to be tough. Cycle exact simulation/emulation was more important for the 68000 (and Atari) where CPU timing loops were sometimes used but this was less of a problem on the Amiga which was often synced to display timing. Most Amiga software developed after the 68020 came out is more robust and tolerant of CPU differences.

Originally Posted by demolition

I am certainly looking forward to trying this accelerator in my 600, and I certainly do not expect it to be 100% compatible with any existing CPUs, although it would be nice if it was. To get a major speed bump, you need to accept some compatibility problems, like it was the case when 040/060 Amigas came out. Much of the old stuff did not work any longer and had to be rewritten/patched. Although maybe not to begin with but eventually, I would close to 100% compatibility with the most common SW like WB 3.1-3.9, libraries and system friendly software.

The Apollo core is potentially (after enough bugs are fixed) *more* compatible with the 68020/68030 than the 68040 and 68060 where the much larger copyback caches could cause problems. The Apollo core with modern memory bandwidths is able to use more compatible writethrough caches with bus sniffing which is very tolerant of self-modifying code and does not require manual cache flushing. Gunnar was originally going to make some incompatible (to the 68020) changes to the ISA which I opposed after code analysis showed inadequate benefits but he claims to have improved 68020 compatibility since. The 64 bit result integer multiply and divide instructions and MOVEP instruction no longer require trapping which is more compatible than the 68060 in some cases.

The biggest incompatibility of the Apollo core currently may be lack of standard (1200 and big box Amiga) connectors on accelerators. These connectors should allow FPGA accelerators to work with Jen's new motherboard. The most important incompatibility may be the strong, independent and proud personalities of Jens and Gunnar which seems to be having interfacing and communications problems.

[jbenam]

We seriously need some mods to split this thread - it's getting out of control

[kolla]

Originally Posted by matthey

Superior for what purpose (68020 compatibility, high performance, AmigaOS 3.9 support)?

Compatibility, for running Amiga software that requires a 68020+.

Most Amiga software developed after the 68020 came out is more robust and tolerant of CPU differences.

Yet, among all the videos on youtube of the Apollo cores being test, 68020+ software is very rare, all I can recall has been doom clones.

The Apollo core is potentially (after enough bugs are fixed) *more* compatible with the 68020/68030 than the 68040 and 68060 where the much larger copyback caches could cause problems. The Apollo core with modern memory bandwidths is able to use more compatible writethrough caches with bus sniffing which is very tolerant of self-modifying code and does not require manual cache flushing. Gunnar was originally going to make some incompatible (to the 68020) changes to the ISA which I opposed after code analysis showed inadequate benefits but he claims to have improved 68020 compatibility since. The 64 bit result integer multiply and divide instructions and MOVEP instruction no longer require trapping which is more compatible than the 68060 in some cases.

The biggest incompatibility of the Apollo core currently may be lack of standard (1200 and big box Amiga) connectors on accelerators. These connectors should allow FPGA accelerators to work with Jen's new motherboard. The most important incompatibility may be the strong, independent and proud personalities of Jens and Gunnar which seems to be having interfacing and communications problems.

"... potentially ..." "... may ..."

I don't understand, with all the talk about compatibility, and all the youtube videos, why is it so damn hard to produce some video that demonstrates the compatibility?!

[grond]

Originally Posted by kolla

I don't understand, with all the talk about compatibility, and all the youtube videos, why is it so damn hard to produce some video that demonstrates the compatibility?!

But you do understand the difference between a work-in-progress and a finished product?

[wawa]

Originally Posted by matthey

The most important incompatibility may be the strong, independent and proud personalities of Jens and Gunnar which seems to be having interfacing and communications problems.

so is this a handicap? i thought it is a requirement to drag along a project. and as long as these personalities are able to reconsider it may even work its way.

[kolla]

Originally Posted by grond

But you do understand the difference between a work-in-progress and a finished product?

These terms do not apply for computer products, everything is work in progress, nothing is ever really finished products. But, from what you are saying, can one assume that all features we have been told about for Apollo core, so far are just "goals" and not actually implemented?

[grond]

Originally Posted by kolla

But, from what you are saying, can one assume that all features we have been told about for Apollo core, so far are just "goals" and not actually implemented?

Partly. The FPU is at present only half implemented and bitfield instructions are missing. This breaks compatibility with a lot of programs. OTOH the transparent caches mentioned above are there and so are some 020 instructions that were missing on the 040 and 060. And some features are present but still show unexpected behaviour under rare circumstances that need to be figured out, e.g. superscalarity, branch prediction and instruction fusing. This is what we call "bugs" and what the current work focuses on. A work-in-progress, progressing fine.

BTW: your question is a splendid example of what I call "binary thinking": it's not completed, this must mean nothing has been done yet.

[kolla]

Not really, just that all the features that are "promised" are not at all guaranteed to become reality, a little bit of moderation when announcing features would be very much in order. The way Gunnar announces things, it sounds like everything is already implemented, which typically is not at all the case. Also, a little acknowledgement for what other people have accomplished would not hurt.

[wawa]

Originally Posted by grond

Partly. The FPU is at present only half implemented and bitfield instructions are missing. This breaks compatibility with a lot of programs. OTOH the transparent caches mentioned above are there and so are some 020 instructions that were missing on the 040 and 060. And some features are present but still show unexpected behaviour under rare circumstances that need to be figured out, e.g. superscalarity, branch prediction and instruction fusing. This is what we call "bugs" and what the current work focuses on. A work-in-progress, progressing fine.

BTW: your question is a splendid example of what I call "binary thinking": it's not completed, this must mean nothing has been done yet.

gunnar?

[grond]

Originally Posted by wawa

gunnar?

Nope. But we could meet for a coffee, I'm also in Berlin.

[eXeler0]

So.. any news about Beta-dates / pre-orders etc? Or if hardware is finalized?

[iggybeans]

Originally Posted by kolla

...The way Gunnar announces things, it sounds like everything is already implemented, which typically is not at all the case. ...

He can get carried away.
It didn't help the Natami project when he would do that.

Does anyone else feel like PMing "Just finish the fuggin' CPU core, dude"?

[eXeler0]

Originally Posted by iggybeans

He can get carried away.
It didn't help the Natami project when he would do that.

Does anyone else feel like PMing "Just finish the fuggin' CPU core, dude"?

Remember how this thread used to be about Jens' Amiga Reloaded mobo? ;-)

[Schoenfeld]

Originally Posted by eXeler0

So.. any news about Beta-dates / pre-orders etc? Or if hardware is finalized?

No change to...:
http://wiki.icomp.de/wiki/Amiga_relo...C_target_price

Jens

[iggybeans]

Originally Posted by eXeler0

Remember how this thread used to be about Jens' Amiga Reloaded mobo? ;-)

Sorry, someone mentioned Gunnar and I saw red for a moment.
Yes, by all means, back to a project with a stated timeline and an likely end result.

[spudje]

Just wondering what's the state of the mentioned network addon? It refers to an ACA500 network addon, but as far as I know there is none available?

[NovaCoder]

Jens,

I have a request, not sure if it's already been covered in this thread.

How about adding a MIDI output? It should be cheap and easy to do and would be very useful.

[eXeler0]

Originally Posted by NovaCoder

Jens,

I have a request, not sure if it's already been covered in this thread.

How about adding a MIDI output? It should be cheap and easy to do and would be very useful.

Interesting idea, but seeing his previous arguments, he's not a big fan of adding stuff only *some* ppl will appreciate. (But hey, it worked pretty well for Atari back in the day ;-)

[SKOLMAN_MWS]

CDTV has MIDI.

[demolition]

Most people will not care about MIDI and it is relatively simple to build such an interface and connect to the parallel port.