"New" amiga hardware

[wawrzon]

how about that manner of comments:
http://repo.or.cz/w/AROS.git
i think its pretty informative, when its comes to the progress and what is being worked on. the code itself must not necessarily be public.

[wawrzon]

Update:
new progress video uploaded
http://www.apollo-core.com/bringup/

Is this way of updating the progress what people want?

wery well, but couldnt you finally post screenshots of the current amiga benchmarks?
as it looks like the core runs amiga games just fine, im assuming this is 68000 software. what is the area to work on now, are there still missing instructions?

[wawrzon]

i think it could be good at some point to outsource/distribute the testing to save time of the core developers.

[wawrzon]

chip ram access seems normal as expected, the frame rate on indy though looks pretty low?

[wawrzon]

The result of 3.5 MB/sec is actually the best possible result on a 16bit AMIGA.
I'm not sure if this is normal to reach.
What do other cards reach on 16bit AMIGAs?

ive never measured that myself, but yes, i think its as much as you can expect to get.

[wawrzon]

Quote:
Originally Posted by Thomas Richter  View Post
My idea would be that you provide some "unimplemented integer instruction" exception and an "unimplemented addressing mode" exception (whoops, there are already!)
This is of course another option and would work even today.
Good idea.

i think its self explanatory that it should be done as thomas proposes, the system should be usable at all times, even if non existent instructions lead to exceptions that are handled by a cpu lib instead crash. it will also be convenient for user bug reports.
what concerns apollo target it remains to be seen how popular it will get, but it will either fragment the scene or the new instructions will have to bi added to the appropriate existent cpu libs to achieve forward compatibility of all systems. either way most existent software is compiled against existent targets, and this will overweight any alternative very likely forever.

[wawrzon]

That a CPU trows an excpetion for unimplemented instructions is normal.
Phoenix does this already.
And old 68000 cores did this also.


Nevertheless on AMIGA really no one cared about this.
In theory you could catch all FPU instructions  anfd run programs using the FPU this way on any AMIGA.
Did anyone do this?

im not sure anymore but i always sort of expected math libs to provide soft fpu emulation, at least that is how this should have been done i guess. of course this doesnt actually offer anything, the fpu version of software remain to be recommended for systems with fpu but the system remains stable and running, and this is important. okay maybe its enough to throw exception and exit the task, except it is an important one and takes down the system.

[wawrzon]

On my A500T + RTG + 060 I barely could get 2.1 MB/s on Chip at best.

060 cards are one known inconvenient example.

[wawrzon]

Yes this is a debug/development build only.
I think in the end about 10% better is reasonable to assume.

This result is the same for the Vampire 600 and the Vampire 500.
The Apollo/Phoenix card is a lot faster though....

gunnar as i see it you achieve 060+ class speed on a vampire card!¿:O
this really aint bad.
as far i see the caches dont get recognized, or arent there any? if so i guess it could still have some serious impact, but i wonder if there is any space left.

anyway im most curious how your accelerator will get off..

[wawrzon]

@ppcamiga1

It really really depends on what you mean by "Amiga".

1. If Amiga is a computer running AmigaOS or something very similar then yes, AmigaOnes are Amiga, as well as PPC Macs running MorphOS or AROS running on PC.

2. If Amiga is a computer with a series of special chips, then nothing after Commodore/Escom and some of the clones can be called an Amiga. It's doubtful even if the Draco could be called one in that case.

3. If Amiga is a computer running AmigaOS or an OS based on it's source code then AmigaOne is an Amiga and not MorphOS nor AROS.

4. If Amiga is a culture rather than a specific technical solution, then everything trying to be Amiga-like is Amiga.

Strictly speaking, I think (2) is right, but I live by the rule that (4) is right because it's much more fun that way

macro system, the company behind draco has never tried to disguise it as amiga. it was a compatible computer in its own right meant for particular application. i dont even now how did they licensed and adopted the kickstart but it must have been done properly.

at least i don know anyone referring to his draco as amiga. in comparison amigaone owners constantly insist that they computers are true and official amigas, even though they are actually not allowed to be called that.

[wawrzon]

AmigaOne is Amiga and AmigaOne is better Amiga than crap which Commodore made after 1990.

in what way its better than our "crap"?
because there is no technical invention in those "amigaones" that would make them unique, just only the standard pc components, only either broken by design or totally outdated at release?

[wawrzon]

the page is so trashy. i kinda like it.

[wawrzon]

Another small tuning.
Now enabling LINKSTACK

great. though keep in mind to cgeck if performance on some other end does not suffer, because for example the chip speed is now back to the lower value.

[wawrzon]

if i read gunnar correctly the vampire fpga is too small to include an fpu, but they have it ready for apollo softcore that means for they own card.

[wawrzon]

interim apollo (vampire) bustest results:
http://www.apollo-core.com/bringup/bustest2.jpg