[Vampire] The AMIGA Future Is NOW! AROS!

[mikej]

Do you see any possibility to make an acc card for "real" amiga hardware using the RPi Compute module?

This was discussed. I have no interest in doing it, the accelerator market is somewhat saturated. It's much easier to interface the compute module directly to the FPGA as you can control timing. To fit in a 68K socket you would need a bridge device so you are losing the cost benefit. Perhaps a zynq with embedded CPU core would be interesting, but they are quite pricey.

"We know that we can put APOLLO in to an ASIC."

I can put anything into an ASIC. My day job is working for a fables ASIC house that actually makes chips (specialist high performance 28nm CPUs amusingly). I keep meaning to run the TG68K CPU through the Cadence synthesiser and get a timing estimate - it would probably go at least 500Mhz, maybe more.

That doesn't mean it's going to happen though. Licence costs for the tools, cells and memory compilers are very steep, as are the production costs. We make high price devices so there is a business case. I don't see a fast 68K processor selling for more than a few$ - it's got to compete against Atmel etc who make really quite nice ARM based SOCs.

There is test, characterisation and a million other things to consider apart from having just the RTL.

NXP still make 68K devices. While making a few hobby FPGA designs is not going to interest them, they will aggressively protect their IP if you try to make a business out of it.

The vampire core is fun and seems to work well, but I don't believe it's going to take over the world (unless ARM was to vanish tomorrow).
/Mike

[Iggy]

This was discussed. I have no interest in doing it, the accelerator market is somewhat saturated. It's much easier to interface the compute module directly to the FPGA as you can control timing. To fit in a 68K socket you would need a bridge device so you are losing the cost benefit. Perhaps a zynq with embedded CPU core would be interesting, but they are quite pricey.

"We know that we can put APOLLO in to an ASIC."

I can put anything into an ASIC. My day job is working for a fables ASIC house that actually makes chips (specialist high performance 28nm CPUs amusingly). I keep meaning to run the TG68K CPU through the Cadence synthesiser and get a timing estimate - it would probably go at least 500Mhz, maybe more.

That doesn't mean it's going to happen though. Licence costs for the tools, cells and memory compilers are very steep, as are the production costs. We make high price devices so there is a business case. I don't see a fast 68K processor selling for more than a few$ - it's got to compete against Atmel etc who make really quite nice ARM based SOCs.

There is test, characterisation and a million other things to consider apart from having just the RTL.

NXP still make 68K devices. While making a few hobby FPGA designs is not going to interest them, they will aggressively protect their IP if you try to make a business out of it.

The vampire core is fun and seems to work well, but I don't believe it's going to take over the world (unless ARM was to vanish tomorrow).
/Mike

500MHz? Sounds realistic with a little tweaking.
Possibly higher with cache and a bit more work.
But once its finalized in dedicated silicon, no more changes...;-(

[mikej]

500MHz? Sounds realistic with a little tweaking.
Possibly higher with cache and a bit more work.
But once its finalized in dedicated silicon, no more changes...;-(

You can park some wafers during production so you can do metal mask changes later without re-running the whole process. Let's you correct minor bugs for less cash.

The amount of verification done on ASIC designs is huge to avoid this kinda thing. ...

[Iggy]

You can park some wafers during production so you can do metal mask changes later without re-running the whole process. Let's you correct minor bugs for less cash.

The amount of verification done on ASIC designs is huge to avoid this kinda thing. ...

At least we are talking about  somewhat realistic speeds again.
And here is where your idea of an accurate recreation really helps.

But an '020 at a few hundred MHz still faces a pretty tough competition from the much more competent superscalar '060.

Ideally, a new 68K would be as compatible as the '020/'030 but have pipelining and scalar advantages that would speed up operations and make them more concurrent.

[psxphill]

as a comptent person you probably are aware that 68k mmu emulation wont work with jit.

It is perfectly possible to emulate an mmu in a jit.

You could probably even use the native mmu.

[gregthecanuck]

Ideally, a new 68K would be as compatible as the '020/'030 but have pipelining and scalar advantages that would speed up operations and make them more concurrent.

Hi Iggy -

This is what the Apollo core does. I am sure you know that but just in case you don't, there is a good write-up on the Apollo core wiki in the instruction fusing, bonding and other goodness. I believe most instructions are down to 1 clock.

Cheers!

[wawrzon]

It is perfectly possible to emulate an mmu in a jit.

You could probably even use the native mmu.

thats not what toni says, im afraid.

[psxphill]

Ideally, a new 68K would be as compatible as the '020/'030 but have pipelining and scalar advantages that would speed up operations and make them more concurrent.

The whole "compatible as the 020/030" issue was pretty much solved in the 1990's. I'd rather not have another project poisoned.

If someone is going to do something like a 68060, then doing a 68060 clone would be ideal.

[wawrzon]

according to what ive been told again, 020 represents the widest possible instruction set of the whole 68k family. all the others cpus except few small exceptions use the subset of instructions available on 020 (and 6888x, counting fpu in). apollo identifies itself as 040 but provides all the extensions there have been to 020 instruction set.

so, if this is true, apollo core is functionally the most compatible alround 68k implementation there is.

[Iggy]

according to what ive been told again, 020 represents the widest possible instruction set of the whole 68k family. all the others cpus except few small exceptions use the subset of instructions available on 020 (and 6888x, counting fpu in). apollo identifies itself as 040 but provides all the extensions there have been to 020 instruction set.

so, if this is true, apollo core is functionally the most compatible alround 68k implementation there is.

"020 represents the widest possible instruction set of the whole 68k family"

Fair statement.
The '030 isn't really that different.

"and 6888x, counting fpu in"?

FPU is now functional?

[IanP]

according to what ive been told again, 020 represents the widest possible instruction set of the whole 68k family. all the others cpus except few small exceptions use the subset of instructions available on 020 (and 6888x, counting fpu in). apollo identifies itself as 040 but provides all the extensions there have been to 020 instruction set.

so, if this is true, apollo core is functionally the most compatible alround 68k implementation there is.

I think old software mis-identifies the Apollo as an 040, I've also seen old software mis-identify it as an 060 recently where-as some new software can correctly identify it as an Apollo "68080". So it would be wrong to say the Apollo identifies itself as an 040.

[wawrzon]

what i wanted to say actually is that i suspect, whatever the core is being identified as, the software supposedly will work, as all instructions are available.

[wawrzon]

a member of apollo team has posted a video of aros running on vampire expanded amiga.
maybe not that much to see but the degree of responsivity and the 68k cpu recognition is probably what was intended to be demonstrated.
https://youtu.be/V83Az5BrTow
config, apparently: V500 in an A2000 + ZorroII PicassoIV (1024x768/16).

[Iggy]

128MB of addressable memory (chip+fast), SOLD!

Pack me one of those mofos up, we now have enough power to run AROS on an A2000.

[psxphill]

thats not what toni says, im afraid.

What does toni say?

Because toni has a lot of experience with winuae and not a hypothetical jit that runs natively on x64. So I can't tell whether you misunderstood what he was saying, or whether to disagree with him.

what i wanted to say actually is that i suspect, whatever the core is being identified as, the software supposedly will work, as all instructions are available.

Unless it needs the instructions to trap.

[polyp2000]

a member of apollo team has posted a video of aros running on vampire expanded amiga.
maybe not that much to see but the degree of responsivity and the 68k cpu recognition is probably what was intended to be demonstrated.
https://youtu.be/V83Az5BrTow
config, apparently: V500 in an A2000 + ZorroII PicassoIV (1024x768/16).

That really was a beautiful sight! This is what happens when people cooperate and share ideas. On one hand we have the open source AROS and on the other side the pure joy to follow technology coming out of Team Apollo. This is real progress , this is what all that work was for and the more open we get the more the promise of an exciting future starts to shine. Let us continue to build on these partnerships.