"New" amiga hardware

[matthey]

Jens S for Individual is certainly NOT on apollo team! lmao. These guys are working daily on this code from what I hear.

Jens Künzer - Apollo Team
Jens Schönfeld - Individual computers

There has been some friendly constructive communication between Jens Schönfeld and Gunnar of the Apollo Team on the German forum http://www.a1k.org/. I am not aware of any agreements but future cooperation is not out of the question. JensS could use more powerful 68k fpga processors for Amiga accelerators, his Chameleon experiment and Clone-A prototypes use Altera fpgas which the Apollo Team prefers and I believe they live fairly close to each other in Germany. I believe JensS was planning on ARM or MIPS for the main CPU in Clone-A before the latest communication with Gunnar. JensS seemed to be surprised at how much more powerful of a 68k fpga CPU is possible than the TG68. Any reluctance he has may disappear with a successful Phoenix core though. A few users of the fpga Arcade and Chameleon are already asking if they will be able to use the Phoenix core. That's only going to increase if the Phoenix core lives up to it's hype.

[matthey]

I thought Clone-A was a dead project, but I never heard of any reason save a vague reference to some legal issues, as to why it was cancelled. Are you implying that Clone -A is still in development?

I believe the Clone-A concept is still alive. The Indivision and Chameleon projects are offshoots of the developed technology. Clone-A is not necessarily another fpga board. JensS has kept much of his ideas secret so they are not stolen but we have seen hints. He seemed to be leaning toward ARM or MIPS for the CPU (probably with software emulation of other processors) and it likely would be Amiga compatible (as well as other systems), probably with an fpga (for chipset emulations?). He has talked about targeting the masses and not just the narrow Amiga niche. It could have specific uses targeted at particular modern users and could come in many forms (board, tablet, laptop, portable gaming system). I can't help but think that it would be somewhat similar to the fpgaArcade but with differentiating advantages of some kind. There is no guarantee that it will ever see the light of day but much like the Natami, it's premature to call it dead.

[matthey]

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.

The MIPS result in SysInfo is rubbish and can't be compared to a 68060. The Phoenix fpga CPU is doing some work though . There was a fast memory bug fixed recently by Majsta that may help the results. I don't know why the system libraries still appear in "SLOW RAM". The caches may not show up in SysInfo as they are transparently handled in Phoenix because they are Writethrough with bus snooping instead of Copyback caches. Writethrough caches with bus snooping don't require the caches to be flushed by the OS so it may not be necessary to inform the OS that they exist. SysInfo detects a 68000 which had no caches or instructions to add them also. Gunnar could clarify the question about caches (and fast memory). It was a good question which I can't answer except to tell you not to trust SysInfo.

[matthey]

small update, small tuning with 5 Mips more

Chip memory access is up significantly too. Nice!

[matthey]

Wouldn't SysSpeed give more accurate results than SysInfo? Would be nice to be able to make more meaningful comparisons

SysSpeed has already been suggested as well as Doom .

Is that you in your avatar? If so, who of the two are you? Just had to ask

That's him. He's the one with the big gun(s) but maybe his girl has some too .

The bringup web page has been updated now.

http://www.apollo-core.com/bringup/