@Kronos @Persia

I am fully aware of the history of the Amiga and was needed at the time for the Amiga as a consumer mass market product. I would not spend my 8 billion pounds to just walk in the same tracks as history already has done.

The server version would obviously not have been used for consumer purpose.

The experience gained by developing a high-end enterprise server AmigaOS could however later have been used for the consumer market Amigas, when the price and performance of the hardware had the right ratio.

With 8 billion pounds in my pocket, I would have made that choice.

@Smurfen

Quote:

Smurfen wrote: @babsimov I agree with many of your ideas. I would add two things. 1. I would have made the deal with Sun Microsystems to license Amiga OS for their Sun servers. If the cooperation had been good this would have laid the ground for Amiga as a server OS. It would also have required that the OS would have been forced to support a HAL, memory protection and multiprocessing. (multi core and thread). And also more focus on networking, which the Amiga was already good at in the early days.

Yes, i agree, i think it's a lost opportunity for Commodore at the time.

Quote:

2. I would have tried to prevent the Commodore PC department, and focused more on the Amiga OS and HW side.

I have said it "Commodore don't enter in the PC market and remain exclusivly on Amiga and C64. But there is no C128/+4/C16 and others. "

I strongly agree about no PC department at Commodore. Focus only on Amiga (software and hardware) and have a very good Marketing for it. All the things Commodore never do in reality.

Quote:

3. A bit contradictory but I would also have encouraged some ideas of Fleecy Moss, such as app store and a hardware agnostic mobile OS, that he thought about even before apple went there. Cheers

Don't know about it, maybe it's a good thing, but only if Commodore don't loose focus for promoting Amiga.

@Smurfen

Quote:

Smurfen wrote: @Kronos @Persia I am fully aware of the history of the Amiga and was needed at the time for the Amiga as a consumer mass market product. I would not spend my 8 billion pounds to just walk in the same tracks as history already has done. The server version would obviously not have been used for consumer purpose. The experience gained by developing a high-end enterprise server AmigaOS could however later have been used for the consumer market Amigas, when the price and performance of the hardware had the right ratio. With 8 billion pounds in my pocket, I would have made that choice.

I fully agree with the server option (remenber Envoy from Commodore ?) make it available early.

In my dream i like to see a very high end graphic Amiga Workstation (like SUN/Silicon Graphics) and of course see Amiga in every office instead of PC/MAC.

@Smurfen

I like the Amiga Server OS idea. Whether it is for Sun Microsystems or not. I would've loved to have seen this in the '90s. Screw Netware, screw NT Server, screw Linus Torvalds.

_________________
All the way, with 68k

@agami

Quote:

agami wrote: @Smurfen I like the Amiga Server OS idea. Whether it is for Sun Microsystems or not. I would've loved to have seen this in the '90s. Screw Netware, screw NT Server, screw Linus Torvalds.

Screw security, yeah!

@agami

Personally, I think the idea of an Amiga server is a little weird.
You take a machine designed with graphics and sound in mind, and you use it for applications that need neither and you don't take into account its lack of memory protection or security functions.
Ah...no.

@OneTimer1
Quote:

Screw security, yeah!

@iggy
Quote:

You take a machine designed with graphics and sound in mind, and you use it for applications that need neither and you don't take into account its lack of memory protection or security functions. Ah...no.

Why do people always assume that for any other purpose one would just take Amiga OS 3.x as-is and apply it to another use? I got £8B, I wouldn't just take the existing OS, bundle it with AmiTCP 4.x and call it Amiga Server OS.

I would spend a few million pounds de-emphasising user/desktop paradigms and emphasising backend server workload paradigms. Yes, I would have them add security and multi-user topology, and yes I would have them implement memory protection.

_________________
All the way, with 68k

@babsimov

Quite detailed and well thought out post there.

Quote:

Finance Jay Miner from the start in 1980 for his 68000 machine project, for a late 1983 release.

I agree here. I think in coming after Apple in 1985 something was lost. Like it was a year late. Apple looked like the trailblazer. The Amiga could have come off as an Apple rip off being so close.

Quote:

- A 2 colors 31 kHz hires mod with an inexpensive monochrome monitor for it.

Do you remember the A2024? It could do 1024x1024 monochrome. In PAL. 1024x800 NTSC.

Quote:

The AmigaOS would be what originally planned (CAOS)

So users don't confuse it with Workbench.

Quote:

But there is no C128/+4/C16 and others.

I liked those machines. By comparison the C16 palette was almost four times the size of the Amiga.

Quote:

The new chipset ACS (Advanced Amiga Chipset) keeps the OCS for compatibility with 6 bits.

ACS. Haha. What about ECS? Ranger even? I think it makes sense to up the bitplane depth to 8 like was done in AGA. The BitMap structure supported it.

Quote:

But, a new 256 colors mode (chunky) appears with a blitter and copper chunky 32-bit. The palette evolve from 4096 to 2621444. Of course there is Chunky HAM8.

You mean 32-bit copper writes? Also, it sounds like SVGA, 256 colours with 18-bit precision. And your HAM8 would differ slightly to AGA, which was based on 24-bit palette, but still only modified 6-bit of colour.

I also think a 16-bit true color mode would help having OCS palette precision and being faster to process. Especially on pre 68020 CPUs.

You didn't mention dual playfield. How does that fit in? At OCS level that's two 8 color bitmaps.

Quote:

1988 work start on AAA chipset for 1990.

What about sound? I didn't see it.

@Hypex

Quote:

Quote: Hypex wrote: @babsimov A 2 colors 31 kHz hires mod with an inexpensive monochrome monitor for it. Do you remember the A2024? It could do 1024x1024 monochrome. In PAL. 1024x800 NTSC.

Yes i remember, but it's not cheap. What I proposed was a 640x512 color mode 2 in 31 kHz and a bit expensive monitor to go with.

Of course later the A2024 option could have been available in addition.

Quote:

But there is no C128/+4/C16 and others. I liked those machines. By comparison the C16 palette was almost four times the size of the Amiga.

Are you sure ? We talk of this computer :
https://en.wikipedia.org/wiki/Commodore_16

They say it have a 128 colors palette, the Amiga have 4096 colors. And, if i remember correctly, The C16 is not C64 compatible.

The key word, focus on computers that sell well, the C64 and PET range. And all other ressources on the Amiga which should become the spearhead of the brand and ultimately the only platform for Commodore.

Quote:

The new chipset ACS (Advanced Amiga Chipset) keeps the OCS for compatibility with 6 bits. ACS. Haha. What about ECS? Ranger even? I think it makes sense to up the bitplane depth to 8 like was done in AGA. The BitMap structure supported it.

ECS is a hack, AGA too (instead we go to AAA for high end).

With ACS chipset, we keeps OCS for compatibility and move to 8 bit chunky for the future. No need to go through 8 bits planar, no future, as we have seen (it was assumed that we go back with the knowledge of what happened and mistakes done)

About the ranger chipset, it's nice, but 128 colors is not enough to compete with incoming VGA, and it remain planar.
Even Jay Miner acknowledged that he should have gone directly to Chunky for the Amiga with hindsight.


Quote:

But, a new 256 colors mode (chunky) appears with a blitter and copper chunky 32-bit. The palette evolve from 4096 to 2621444. Of course there is Chunky HAM8. You mean 32-bit copper writes? Also, it sounds like SVGA, 256 colours with 18-bit precision. And your HAM8 would differ slightly to AGA, which was based on 24-bit palette, but still only modified 6-bit of colour. I also think a 16-bit true color mode would help having OCS palette precision and being faster to process. Especially on pre 68020 CPUs.

I am not a technician and I can't follow on that land there, especially in English (it's not my language). But my readings here and there on the internet about the Amiga, on forums, allowed me to understand that having keep the 16 bits blitter and copper from OCS in AGA and not to move to Chunky were the main mistakes. And, of course, not to have release AGA instead of the ECS (1990), although AAA would have been better for High end, and that's what I did, in my dream timeline.
The ACS is a cheap way to counter the VGA that appears in 1987, forever keep the Amiga ahead of others.

Quote:

You didn't mention dual playfield. How does that fit in? At OCS level that's two 8 color bitmaps.

Yes i know, because i'm not a technician.
But of course ACS might include something like this, maybe have a 16 bit chunky mode from the start, so we could have two 8 bit chunky dual playfield. As not a technician, maybe it's not possible, i don't know.

Quote:

1988 work start on AAA chipset for 1990.

What about sound? I didn't see it.[/quote]

AAA include the Mary sound chip (8 channel 16 bit)

http://www.amigahistory.plus.com/amigaaaa.html
https://en.wikipedia.org/wiki/Amiga_Advanced_Architecture_chipset
http://www.thule.no/haynie/

But it's expensive for entry level computer.
So, for 1990 entry level computer, I retain the ACS. Because it's been amortized from sells of high end from 1987 to 1990. It's powerful enough and, now, inexpensive for entry level.
I have not talked about the sound for ACS. Because I thought keep the Paula chip. Until 1991/1992 it was still a sound chip entirely correct. But it is conceivable that ACS include two Paula chips, for 8 channels 8 bit or 4 channel 14-bit. It seems a good compromise to better AAA monica chip for high end.

Maybe at 1990, i would had an optional zorro 3 board with DSP 3210 for the ACS range. This would ensure that the DSP is more common on the Amiga and therefore more likely to be used by developers. And it add the 16 bit sound for ACS generation.

Here here information on the DSP3210 and its integration into the Amiga, scheduled for 3000+ in Commodore at the time and what led to this the not so.

I think i made an edit to my original post to include these modification and information about the sound with ACS (the two paula thing).

EDIT :

Maybe at 1990, i would had an optional zorro 3 board with DSP 3210 for the ACS range. This would ensure that the DSP is more common on the Amiga and therefore more likely to be used by developers. And it add the 16 bit sound for ACS generation.

The overhaul ot the ACS range at 1992 make the DSP card standard. But i realize that the 700 board need a redesign to add it in internal (the case is a 500 case, remember. Or there is an internal connector for the DSP from de start in the 700 (better option i think).

Here informations on the DSP3210 and its integration into the Amiga, scheduled for 3000+ in Commodore at the time and what led to this the not so.

http://www.lemonamiga.com/forum/viewtopic.php?t=12482&start=15&sid=338786b0555e8580097e825f1d1884d8

I think i made an edit to my original post to include these modification and information about the sound with ACS (the two paula thing).

The original post edit is done, with some new details.

Last edited by babsimov on 06-Sep-2015 at 09:08 PM.
Last edited by babsimov on 06-Sep-2015 at 08:29 PM.
Last edited by babsimov on 06-Sep-2015 at 03:29 PM.

I would team up with a company called Nvidia (founded 1993) to develop graphics cards for the Amiga. And start moving the whole platform towards commodity hardware.

@persia

"...commodity hardware"

There's some sanity.

As time has shown you can't compete with commodity hardware. No-one, regardless of how much a bohemath ever has and its crazy to think you could.
These companies spend billions focusing on developing what end up the best that technology at any particular time can achieve.

Also, the problem the Amiga and commodore had wasn't engineering hardware, they had good engineers whose fingers where firmly on the pulse of what was required to keep Amiga at the forefront much more closely than a bunch of delusional fans 20 years later..... the problem was upper management who were clueless and also shackled the engineers.

@fishy_fis

Quote:

fishy_fis wrote: As time has shown you can't compete with commodity hardware. No-one, regardless of how much a bohemath ever has and its crazy to think you could. These companies spend billions focusing on developing what end up the best that technology at any particular time can achieve. Also, the problem the Amiga and commodore had wasn't engineering hardware, they had good engineers whose fingers where firmly on the pulse of what was required to keep Amiga at the forefront much more closely than a bunch of delusional fans 20 years later..... the problem was upper management who were clueless and also shackled the engineers.

I agree, specialy about the Commodore management (mismanagement i mean :) )

@fishy_fis

"...delusional fans 20 years later...."

hey!

@mcbone

Quote:

mcbone wrote: Q . If you could go back in time with eight billion pounds world you buy out commodore 64 and Amiga and if you had enough money left and see apple up for sale would you buy it think you could use some of hardware and software and technology to help Amiga out more into the future

No.

_________________
Troll - n., A disenfranchised former potential customer who remains interested enough to stay informed and express critical opinions.
opp., the vast majority who voted silently with their feet.

@Boot_WB

hello could you tell me why you world not go back to buy commodore

_________________
maybe i am dyslexia

An Apple a day keep bill gates away

@mcbone

- I can think of far more pressing concerns than rescuing Commodore from their own destiny;
- I have no desire to run a large company;
- outside forces being the same, I suspect any successful parallel reality Amiga operating system would yield to the same market forces that afflict the main OSes: advertising, background processes and frameworks hogging resources. - losing the aspects of the system I appreciate (user control mainly) in the process;
- Ethics - With an eye to the ethics of time travel, I couldn't live with the responsibility and temptation. LOTR, as a popular example, gives some interesting insights into the ethical dilemma of unilateral Power: Galadriel's temptation by Frodo being the prime example.

I'm not usually one for absolutes, but in this case my 'No' is immutable. :)

_________________
Troll - n., A disenfranchised former potential customer who remains interested enough to stay informed and express critical opinions.
opp., the vast majority who voted silently with their feet.

@OneTimer1

Quote:

And if it was done right, the PC64 would have been more C64 compatible than the C128 in its C128 Mode.

I don't see the need since it already had C64 mode. But to me it makes sense to improve the OS and grqphics rather than stay in a rut becauese C64 owners wanted it compatible with their hardware banging games.

Quote:

Besides, the time of the mid 80ies was the time when the C64 sold most machines and even the C128 was sold over 5 million times without loss.

Maybe but 16-bit was modern then and even the 32-bit 68020 was out. Plus graphically it was inferior to machines with 256 colour palettes. Which evne makes the Amiga look bad.

Quote:

Oh maybe you missed it I mentioned the a 8086 (16bit) card as an alternative to Z80

Yes but Commodore ran the kernel on a MOS. Don't quite get why they decided to add a Z80 to boot it for CP/M. Did they lack software? And I see no need to drag Intel into this. Don't need another Commmdore Colt. It's a bit like adding a PPC to an Amiga. It can't run native code the machine was designed for. Not a good upgrade path.

Quote:

You forgot (or deleted) that I was talking about an A500 successor an A1200 like machine.

And that would make it better? An Amiga in a PC case looks like a PC. But why do that when the Amiga should stand apart?

Quote:

It might be an unknown to you, but i wrote:

Yes I saw PC case. So Amiga KB and mouse into PC case. Compatible with PC ports? What about Amiga joystick ports?

Quote:

True, but some CPUs had the wrong byte orders and others (MIPS, Sparc) where much more exotic than PPC, lacked performance or where discontinued earlier like the PA-Risc or Alpha.

From what I read the PA-Risc survived until 2005 so certainly enough time for an upgrade path.

Quote:

All the links connecting the Hombre with the Amiga on wikipedia are leading to a quote from Dave Haynie, totally denying a relation with Amiga:

That quote removes AmigaOS but not Amiga. In any case if such a machine was released there could have been doubt about it being a real Amiga since the hardware would have little relation to the original Amiga. And as has become the trend Amiga would have just been a label.

Sorry for my late addition, but this needed quite a bit of thought...

Assuming that nobody is going to ask where I got the 8 billion pounds, I'd go back to about 1982 and invest in the early Amiga Inc. I'd let them develop the OCS chipset with just a few tweaks, 2 channels of 16-bit audio instead of 4 channels of 8-bit for example, and adding a 6-bit chunky graphics mode.

At the same time, I'd set up a small CPU development team and begin developing a new cpu. I'd get them to develop a CPU like the hitachi super-H SH-2, i.e. a 32-bit RISC chip but with a 16-bit instruction (like the ARM-Thumb instruction set, which would mean that the 32-bit chip would work well on a 16-bit memory bus).

Also, I'd create a S/W engineering team to create a C compiler for said new CPU and create a new Operating System for the new chip, possibly buying out TripOS and porting it to C for a starting point.

Over the next 3 years these 3 strands would be increasingly integrated in order to come up with a final product in 1985 which would be licenced to Commodore at low cost and any other companies that wanted to produce it (Atari and Acorn maybe ). This would probably be the hardest part of this plan, to convince companies to licence and not buy the new computer platform, but hey, this is just fantasy right...

Anyway, the end goal of this would be to have multiple companies release A500 (and A2000) type machines in 1985 and for the competition between the companies to keep them competitive. The minimum machine specification would be set by Amiga-Inc (512K Ram, 880k floppy, 32-bit RISC CPU with 16-bit instruction set and memory interface, O/S on an internal ROM cartridge) and licencees would be free to add whatever extra features they like in an O/S friendly manner (Midi interfaces, audio input, HDD interfaces etc.). Speaking of which, hitting the hardware directly would not be possible and developers would have to use Amiga-OS libraries to access the functionality of the computer. Sure, this would prevent developers from pushing the machine to it's limits, but with more frequent hardware updates this would not be required. To spur software development, the Amiga SDK would be freely available. I'd also try to create an open-source community around the Amiga by paying developers to create open-source initial versions of essential software, basic word processors, text editors, graphics editors, emulators, sample games etc.

For further development I'd keep eveloping the hardware with new major versions of hardware every 3 years. So, major relases in 1988, 1991, 1994 etc. with the high-end machines leading the way and the consumer machines following 6 months to a year later.

In 1988 (for the high-end machines, 1989 for the low-end) I'd like to see a base machine with an upgraded and faster 16-bit chipset utilising VRAM, say with 512K of VRAM and 1024k of Fast Ram (For a 1.5MB total), this should allow more bitplanes to be shown, certainly up to 8-bit and maybe 12-bit, and it should also allow VGA monitor-output. For audio, either more channels, or adding compressed audio support to the hardware (having the hardware play ADPCM audio directly should be easy enough). A 1.76 MB floppy drive would be mandated. The CPU would have some DSP functions added, as well as an MMU, it would also be available with a 32-bit fast-memory interface for the high-end machines. Also moving to a 5-stage pipeline and better process technology should allow for faster speeds which would also mean that caches would need to be added.

After the 1988 release and with the new MMU in the CPU, The O/S would be gradually be replaced with an incompatible but similar memory-protected O/S, but booting up in the old O/S would always be possible.

Then in 1991/1992 the range would move to 32-bit everywhere, palettes would be 24-bit and the graphics chipset would add texture shading/mapping. 1.76Mb Floppy drives are still included and CD-Rom drives would be mandatory, all machines would come with a IDE interface and space for a HDD. The CPU would have a vector unit (128-bit but double pumped, i.e. it would operate on 2 32-bit floating points at once, but the instruction set would be for operating on 4 32-bit floats at once).


Anyway, that's my basic plan for the Amiga taking over the world with the help of a time-machine, please don't respond and say... "But that's not an Amiga anymore because...." The Amiga wasn't perfect, but with a time machine it could be made better

@raddude9

Quote:

At the same time, I'd set up a small CPU development team and begin developing a new cpu.

With that much money in hand you could have bought MIPS, which started about the same time as Amiga.

_________________
"Unix is supposed to fix that." -- Jay Miner