Amiga fans could force merge Amiga Inc. and Hyperion...

[ffastback]

eslapion wrote:
Just a freaky stupid funny idea I'm throwing at you all.

Amiga Inc. paid less than 50'000$ to hyperion for OS 4.0. Does that mean anything to you people?

And the worst is... they actually made it. Or so they say.

To me, what it means is, both companies are very low budget companies and probably could be bought out for a relatively low price. Low being less than 500'000$ each.

If only a few hundreds, perhaps a few thousands of Amiga fans pooled a good amount of money together, it would be possible to buy out both companies, force merge them then force them to work together with one another and end this stupid battle.

But hey... that's just me... I'm a dreamer... an idealist... a crazy lunatic... a... whatever...

Too bad Amiga has a lot more cash than what you are talking about based on what we last knew, do searches on the Prokom Software SA investment they got (millions) and the very recent agreement with the city of Kent, Washington they entered into for naming rights for a new arena there (for again, millions).

How this is possible is anyone's guess, damn them for being a private company!  But they are not likely to sell for peanuts.  Maybe they are a tax write off for someone, some people think they are a money laundering op.  But no one can really say due to no access to the financials.  The only reason we know what we do at all is because Prokom is a public company and Kent, as a city goverment must be forthcoming with such things.

And if they don't have such money, they certainly don't want us to know about it, especially with the Kent deal in the newspapers.

bloodline wrote:

No, but it would be hard to get an 060 emualtion working in FPGA that would be anywhere near as fast as software emualtion of an 060 on the cheapest AMD or Intel CPU...

I find that rather surprising. Somehow, I would have expected the newer FPGAs that can run at ~ 200MHz to deliver some punch. I mean, real logic gates should be faster than software emulation.

-Edit- you wouldn't bother with an 060, actually... since it was from a software point of view, an 040 with a few missing instructions...

That's interesting.

[meega]

If some bits aren't working, then you might be able to run the chip faster.

Apples used 68k series processors that way.

[bloodline]

eslapion wrote:

bloodline wrote:

No, but it would be hard to get an 060 emualtion working in FPGA that would be anywhere near as fast as software emualtion of an 060 on the cheapest AMD or Intel CPU...

I find that rather surprising. Somehow, I would have expected the newer FPGAs that can run at ~ 200MHz to deliver some punch. I mean, real logic gates should be faster than software emulation.

You forget that the slowest x86 chips are multiple Mhz, designed specificaly for pumping instructions through them... and with an efficient JIT, the 68k instructions map closely to the native instruction set... An FPGA on the other hand is generic logic... nothing you can do will get it anywhere close to the billions of $ pumped into the Athlon64 or the Core2 Duo...

-Edit- you wouldn't bother with an 060, actually... since it was from a software point of view, an 040 with a few missing instructions...

That's interesting.

The 060 was little more than a process stepping (ok the introduction of a second integer pipeline is quite significant, etc)... but in the x86 world it probably wouldn't have warrented a new chip name/number...

[meega]

I don't know... 486, 486DX, 486DX2, and so on.

bloodline wrote:
You forget that the slowest x86 chips are multiple Mhz, designed specificaly for pumping instructions through them... and with an efficient JIT, the 68k instructions map closely to the native instruction set... An FPGA on the other hand is generic logic... nothing you can do will get it anywhere close to the billions of $ pumped into the Athlon64 or the Core2 Duo...

Using WinUAE with JIT here running on a dual core of 2.8GHz, I get an equivalent of a 175MHz 68040.

Emulation is still emulation, you do lose a lot of speed in software translation.

Somehow, considering the way other processors are emulated, I would expect a 200MHz FPGA to provide the equivalent of a 200MHz 68060... and produce much less heat and consume much less power than a modern day intel or AMD processor.

Since FPGAs use generic logic, you can configure them to reproduce any processors. I can understand there isn't much point in trying to reproduce a pentium processor running at 200MHz but in the case of 68k processors that can be a different story.

bloodline wrote:
The 060 was little more than a process stepping (ok the introduction of a second integer pipeline is quite significant, etc)... but in the x86 world it probably wouldn't have warrented a new chip name/number...

If that is so then is there a way for me just to swap the 040 on my Mercury for a 060?

It seems to me that the 060 has more pins.

[bloodline]

meega wrote:
I don't know... 486, 486DX, 486DX2, 486DX4, and so on.

In 68k terms, the 68060 probably should have been called the 68050... since it was an improved 68040 rather than a radical improvement in the architecture...

Note:

68000 -> 68010 small improvement
68010 -> 68020 radical improvement
68020 -> 68030 small improvement
68030 -> 68040 radical improvement
68040 -> 68060 small improvement    -Odd one out, but then also the last of the line, so doesn't really matter...

[cecilia]

Methuselas wrote:
Personally, I think we should force Matt to wear nothing but those mil-spec, boing-ball knickers he's got.


They're HOT!  :lol:


I bet Cecilia would agree with me.  :crazy:

:roflmao:

Now THERE is an idea!!!   :idea:

[bloodline]

eslapion wrote:

bloodline wrote:
The 060 was little more than a process stepping (ok the introduction of a second integer pipeline is quite significant, etc)... but in the x86 world it probably wouldn't have warrented a new chip name/number...

If that is so then is there a way for me just to swap the 040 on my Mercury for a 060?

It seems to me that the 060 has more pins.

Yes, the 060 and 040 are pin compatible... but the 060 use 3.3volts rather than the 040 5volts.. so make sure you adjust the regualtor.

[bloodline]

eslapion wrote:

bloodline wrote:
You forget that the slowest x86 chips are multiple Mhz, designed specificaly for pumping instructions through them... and with an efficient JIT, the 68k instructions map closely to the native instruction set... An FPGA on the other hand is generic logic... nothing you can do will get it anywhere close to the billions of $ pumped into the Athlon64 or the Core2 Duo...

Using WinUAE with JIT here running on a dual core of 2.8GHz, I get an equivalent of a 175MHz 68040.

Not sure where you get that figure from... but WinUAE does a hell of a lot more than just emulate the 68k... run Amithlon or BasiliskII to see a true idea of just how fast Bernie's JIT can be... though it could do with an update for the x86-64 architecture... that would give it a boost, as it has 8 more registers.

Emulation is still emulation, you do lose a lot of speed in software translation.

Not as much as you think.

Somehow, considering the way other processors are emulated, I would expect a 200MHz FPGA to provide the equivalent of a 200MHz 68060... and produce much less heat and consume much less power than a modern day intel or AMD processor.

The 060 was something like 1 and a half million gates... you can't get an FPGA big enough yet... AFAIK.

The Core2 Duo is something like 250million gates...

Since FPGAs use generic logic, you can configure them to reproduce any processors. I can understand there isn't much point in trying to reproduce a pentium processor running at 200MHz but in the case of 68k processors that can be a different story.

The FPGA is designed for flexibility, not speed. Central Processing units are massive and complex devices... that is why it takes massive companies with huge budgets to develop them.

bloodline wrote:
Yes, the 060 and 040 are pin compatible... but the 060 use 3.3volts rather than the 040 5volts.. so make sure you adjust the regualtor.

Hummm... never seen such a thing as an adjustable voltage regulator on a Mercury accelerator...

[meega]

bloodline wrote:
In 68k terms, the 68060 probably should have been called the 68050...

I know, but it looked better as 68060 when competing with the ever increasing x86 numbers of the day.

Besides, the x86's were ramping the clock rates rather than noticeably increasing efficiency.

[bloodline]

meega wrote:

bloodline wrote:
In 68k terms, the 68060 probably should have been called the 68050...

I know, but it looked better as 68060 when competing with the ever increasing x86 numbers of the day.

Well the x86 development was jumping forward leaps and bounds by then... the 68060 was EOL and all of Motorola's money was going in to the PPC project... so I guess the processor number didn't really matter.

[meega]

A little ot, when were the first 68010's produced, and when did the first 68020's appear?

Edit: with apologies for the ninja edit above in the post that you just quoted.

[bloodline]

meega wrote:
A little ot, when were the first 68010's produced, and when did the first 68020's appear?

Edit: with apologies for the ninja edit above in the post that you just quoted.

68010 was relelased in 1982... the 68020 was released in 1984, does that help?