@WolfToTheMoon

Quote:

The next stage in Amiga's life was, IMHO, not to be on a desktop computer but on something that could rival Psions and Palms...

I really dont like that idea.

I think the Home Computer market should of been the focus, IBM & Microsoft had corned Big Business and Schools, Apple had Publishing and Design tied up. Both where not really aiming at making a Home computer.

Amiga had a small chance at the high end in Multimedia department thanks to Toaster, so I think thats where the high end Amiga should of focus on.

But I really think after 1994 there was a gap in the market for a low cost home computer. Many family couldn't afford PC and Macs, (like mine) PCs kinda eventually filled this market with the rise of computer market that sold old PC business machines and cheap second hand cheap PC started flooding the market. But it wasnt really until I say year 2000 ish could you pick up a fairly good second hand Pc for £400, the price of a A1200.

Then their the overlooked other market like india, China, asian markets that also where after lower cost computers, Amiga could of had a bigger impact in these countries.

_________________
AmigaNG, YouTube, LeaveReality Studio

@amigang

Quote:

I really dont like that idea. I think the Home Computer market should of been the focus, IBM & Microsoft had corned Big Business and Schools, Apple had Publishing and Design tied up. Both where not really aiming at making a Home computer. Amiga had a small chance at the high end in Multimedia department thanks to Toaster, so I think thats where the high end Amiga should of focus on. But I really think after 1994 there was a gap in the market for a low cost home computer. Many family couldn't afford PC and Macs, (like mine) PCs kinda eventually filled this market with the rise of computer market that sold old PC business machines and cheap second hand cheap PC started flooding the market. But it wasnt really until I say year 2000 ish could you pick up a fairly good second hand Pc for £400, the price of a A1200. Then their the overlooked other market like india, China, asian markets that also where after lower cost computers, Amiga could of had a bigger impact in these countries.

By 1993 the jig was up. The biggest problem is Motorola. They abandoning 68K was really unfortunate for Amiga. After that, there is no viable path forward, you have to port to a more viable ISA... and that would happen in the worst possible moment, exactly the time in which even Apple nearly went bankrupt - because the PC market was tooo overpowering.

But there's a lot of space in the emerging mobile market. Things like PDAs, small netbooks and similar... that would have been a great place for AmigaOS to carve a niche for itself.

_________________

@WolfToTheMoon

The RISC camp was fragmented instead of combining their efforts into a single goal.

X86 world assimilated the best aspects from SGI (e.g. 3D accelerated graphics, SMP) and DEC Alpha e.g. high clock speed.

Wintel's road map is to assimilate both SGI and DEC Alpha/VMS technical knowledge and biological (brain drain) characteristics.

1. NVIDIA is the chief SGI dismantler. Direct3D is being influenced by OpenGL.

NVIDIA assimilates SGI's graphics unit.
NVIDIA agreed to license SGI's 3-D graphics patent portfolio.
https://www.eetimes.com/sgi-graphics-team-moves-to-nvidia/
Date: October 1999.


2. For VMS vs Windows NT, DEC issued a legal threat against Microsoft. The result is Microsoft paid a settlement of $105 million to DEC. https://www.techmonitor.ai/technology/dec_forced_microsoft_into_alliance_with_legal_threat

Dave Cutler recycled DEC's abandoned Mica operating system code for Windows NT.



3. For Alpha vs Pentium Pro/Pentium II, DEC sued Intel, result: out of court settlement with the dismantling of DEC with the combined might of Intel and Compaq.

https://money.cnn.com/1997/05/13/companies/digital/

https://money.cnn.com/1997/10/27/deals/settle/
Intel, Digital settle lawsuit. Intel buys DEC's cutting-edge semiconductor manufacturing division. Some Digital semiconductor employees will shift to Intel as part of the settlement.

Intel assimilated "big iron" DEC Alpha's IP.

Last edited by Hammer on 24-Aug-2024 at 11:53 AM.

_________________
Amiga 1200 (rev 1D1, KS 3.2, PiStorm32/RPi CM4/Emu68)
Amiga 500 (rev 6A, ECS, KS 3.2, PiStorm/RPi 4B/Emu68)
Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB

@Hammer

Quote:

The RISC camp was fragmented instead of combining their efforts into a single goal.

No, it's really simple. Price, performance, backwards compatibility/huge software base. That's it.

None of the RISC based systems had all of those attributes.

_________________

@amigang

Quote:

I think the Home Computer market should of been the focus, IBM & Microsoft had corned Big Business and Schools, Apple had Publishing and Design tied up. Both where not really aiming at making a Home computer.

IBM's high-resolution text UI DisplayWrite lost against high-resolution text UI WordStar and Word Perfect. Microsoft recycles MacOS GUI-influenced MS Word against high-resolution text UI word processing establishment. MS used similar tactics against Lotus 123.

#Metoo wouldn't be enough when Amiga's Word Perfect 4.x and 5.x ports are from MS-DOS versions.

_________________
Amiga 1200 (rev 1D1, KS 3.2, PiStorm32/RPi CM4/Emu68)
Amiga 500 (rev 6A, ECS, KS 3.2, PiStorm/RPi 4B/Emu68)
Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB

@WolfToTheMoon

Quote:

WolfToTheMoon wrote: @Hammer Quote: The RISC camp was fragmented instead of combining their efforts into a single goal. No, it's really simple. Price, performance, backwards compatibility/huge software base. That's it. None of the RISC based systems had all of those attributes.

Big iron RISC's 1 IPC argument was cancelled by CISC's 1 IPC with code density.

_________________
Amiga 1200 (rev 1D1, KS 3.2, PiStorm32/RPi CM4/Emu68)
Amiga 500 (rev 6A, ECS, KS 3.2, PiStorm/RPi 4B/Emu68)
Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB

Quote:

WolfToTheMoon wrote: Quote: The RISC camp was fragmented instead of combining their efforts into a single goal. No, it's really simple. Price, performance, backwards compatibility/huge software base.

If you start constructing a RISC CPU you want to implement something special, something that could make it faster than other architectures: Register Windowing, Conditional Command Execution, Parallel Memory Transfers, ...

This architectures can't be compatible, this will always lead to fragmented markets. It could be solved via API compatibility (POSIX), choosing a widely accepted common OS (WindowsNT, or by using a fast virtual machine (Java).

But nothing really worked against (backwards) compatibility on the x86 market, even Intel had its problem with their own RISC or VLIW Architectures.

---

On the Amiga, a RISC system would have needed to target the gaming market, where compatibility doesn't matter, or it would have needed a very powerful (expensive) system emulating the 68k/AGA platform.
Fun part, a x86 system may have been an alternative but none of us (including me) really wanted that.

@WolfToTheMoon

Quote:

By 1993 the jig was up. The biggest problem is Motorola. They abandoning 68K was really unfortunate for Amiga. After that, there is no viable path forward,

Maybe, but I actually think 680xx chip would still have been fine to use for the period of 1993 to 2000. The problem was going to be more getting the cost/ price down.

Escom Walker, with its 68030, 2mb chip, 4mb fast ram, cd rom drive, I actually think was a ok spec. If they could got that out for less than £500, (£400 would of been really good) than I think it would of done well.

I mean even with my base Amiga A1200 I was able to do my course work for school in 90s.

I remember reading also in the final years of escom they where going to put together a A1200 surfer pack, that was to a low cost all in one internet package for £600 and it got quite a bit of attention for being one of the lowest cost solution for getting a computer that could connect.

But your right sooner or later Amiga was going to have to make the jump to another cpu.

_________________
AmigaNG, YouTube, LeaveReality Studio

@amigang

Quote:

Maybe, but I actually think 680xx chip would still have been fine to use for the period of 1993 to 2000. The problem was going to be more getting the cost/ price down. Escom Walker, with its 68030, 2mb chip, 4mb fast ram, cd rom drive, I actually think was a ok spec. If they could got that out for less than £500, (£400 would of been really good) than I think it would of done well. I mean even with my base Amiga A1200 I was able to do my course work for school in 90s. I remember reading also in the final years of escom they where going to put together a A1200 surfer pack, that was to a low cost all in one internet package for £600 and it got quite a bit of attention for being one of the lowest cost solution for getting a computer that could connect. But your right sooner or later Amiga was going to have to make the jump to another cpu.

In eastern europe, you could probably continue selling an A500 variant for the better part of the 90ies, as long as it's cheap enough. But you cannot build a future on that, sooner or later cheap PCs are coming and than you're over. The only future Amiga could build on in the 90s is mobile. Desktop wars were over and no alternative OS survived, bar MacOS(barely, more by chance than design) or opensource/free software.

_________________

@WolfToTheMoon: Amiga was never mobile-ready. There was no such thing as power/sleep modes to draw less power, for example.

@WolfToTheMoon

Quote:

In eastern europe, you could probably continue selling an A500 variant for the better part of the 90ies, as long as it's cheap enough. But you cannot build a future on that, sooner or later cheap PCs are coming and than you're over.

I dont know about that, we have seen the rise of the Pi platform and its community, yes in the scale of the computer world its not going to set the world on fire, but it's had a moderate success. This platform did'nt arrive until 2012.

Why was it success, because the CEO, who grew up learning his computer craft on an Amiga A600, couldn't afford a PC and thought, why isnt there a low cost computer to learn simple computer stuff and that is also hackable like this out their!

Now I agree again, the current classic Amiga range, would of had to made the jump to some kind of platform by the end of 2000, if not before. But I still think the Amiga would of had a market, maybe not to the scale of Windows or Mac, but I think we could of been bigger than Linux.

_________________
AmigaNG, YouTube, LeaveReality Studio

@amigang

Quote:

I dont know about that, we have seen the rise of the Pi platform and its community, yes in the scale of the computer world its not going to set the world on fire, but it's had a moderate success. This platform did'nt arrive until 2012. Why was it success, because the CEO, who grew up learning his computer craft on an Amiga A600, couldn't afford a PC and thought, why isnt there a low cost computer to learn simple computer stuff and that is also hackable like this out their! Now I agree again, the current classic Amiga range, would of had to made the jump to some kind of platform by the end of 2000, if not before. But I still think the Amiga would of had a market, maybe not to the scale of Windows or Mac, but I think we could of been bigger than Linux.

raspberry pi was, what, 30ish $ on launch?
You're not getting a new computer/MB for that in the mid 90s(especially adapted for inflation).
By the second part of the 90s, you'll be competing with the used 386/486 PCs at the low end, which will be faster and have much more software available. Even advanced OSes like BeOS(in theory at least, far superior architecture to Win 95/98), which was, in Gasse's own admission, a more advanced version of the AmigaOs, failed to gain a foothold in the market in the late 90s(BeOS was available for x86). It's juts not happening.

Like I've said, you can sell a lot of very cheap Amigas based on the A500 in Eastern Europe(we never did really get the A1200 since it was too expensive at launch, so AGA is not needed) where even used PCs stayed expensive in the 90s, and software was mostly pirated because nobody had money to buy legitimate software anyway. But that market will dry up by late 90s, and than you're stuck with nowhere to go. If you port to another CPU, you also need developers to port their software, but since most software was pirated on the Amiga, they'll not be interested.

_________________

Quote:

amigang wrote: Escom Walker, with its 68030, 2mb chip, 4mb fast ram, cd rom drive, I actually think was a ok spec.

The motherboard was technically a stripped down A4000/30 system, no Buster no Zorro slots, instead they built in a super I/O chip for RS232 (that was important for modems) and a fast parallel port. Amiga Tech. asked a company that didn't know the Amiga to do make this changes, that costed them time and money. I would have never been as cheap as the A1200 even without CD-Rom.

It might had some impact if it came 2-years before C= bankruptcy and if the price was OK.

Quote:

amigang wrote: I remember reading also in the final years of escom they where going to put together a A1200 surfer pack, that was to a low cost all in one internet package for £600

It was a 'naked' A1200 and didn't had enough RAM for a browser, it was a nice idea but not a usable product. There where people buying it using the included internet access for their PC and selling the A1200.

@WolfToTheMoon

Quote:

WolfToTheMoon wrote: ... PDAs, small netbooks and similar... that would have been a great place for AmigaOS to carve a niche for itself.

It would have needed new hardware and a heavy changes in the GUI, you would hardly recognized it as an Amiga.

WolfToTheMoon Quote:

... PDAs, small netbooks and similar... that would have been a great place for AmigaOS to carve a niche for itself.

OneTimer1 Quote:

It would have needed new hardware and a heavy changes in the GUI, you would hardly recognized it as an Amiga.

Power management, LCD driver, CMOS chipset with integration (preferably SoC), low power CPU (3.3V 68060 or 68040V), reskinned GUI like CDTV and CD32, etc. Yea, well beyond the competence level of CBM management.



https://www.commodore.ca/commodore-brochures/



https://vintage-laptops.com/en/commodore-lcd/

Only $600 USD in 1985 but an Amiga version would have been much more expensive that early. Perhaps a more general purpose Amiga LCD would have been possible by 1995 with integration and planning but still expensive. Today, a 68060 Amiga would likely give all day batter life and a 68060+AA+ SoC ASIC would likely cost less than a meal and maybe even a drink to mass produce.

Last edited by matthey on 27-Aug-2024 at 02:33 AM.

@WolfToTheMoon

Quote:

raspberry pi was, what, 30ish $ on launch? You're not getting a new computer/MB for that in the mid 90s(especially adapted for inflation). By the second part of the 90s, you'll be competing with the used 386/486 PCs at the low end, which will be faster and have much more software available. Even advanced OSes like BeOS(in theory at least, far superior architecture to Win 95/98), which was, in Gasse's own admission, a more advanced version of the AmigaOs, failed to gain a foothold in the market in the late 90s(BeOS was available for x86)

I think your forgetting if commodore had survived and went this direction and focus was a low cost computer, then you would do what you can to bring down costs. So like reduce the aga chipset to one chip.

I would have also sold just a Amiga motherboard.
I mean look at these costs listed in these docs
https://archive.org/details/amiga-computer-cost-estimates-1991-04-16/mode/2up
A500 cost $177
A3000+ cost $1028
You can see one the big cost was the cpu, floppy disk drive, etc so they could of made a board maybe without a cpu for very cheap and users / business could add cpu card.

I did a quick calculation based on these docs they could of made an A3000+ motherboard without CPU & disk drive keyboard etc, just the board at cost for around $516. Sell it at $750, again I think it would of done well.

Again A500 just the board you could get down to $109.

Also Steven jones had a cool idea with his siamese project that was to be a plug in pci card so x86 platform could access Amiga and mac software in late 90s.

All this expands the market imo.

Last edited by amigang on 27-Aug-2024 at 10:41 AM.
Last edited by amigang on 27-Aug-2024 at 10:39 AM.

_________________
AmigaNG, YouTube, LeaveReality Studio

@matthey

Quote:

matthey wrote: Only $600 USD in 1985 but an Amiga version would have been much more expensive that early. Perhaps a more general purpose Amiga LCD would have been possible by 1995 with integration and planning but still expensive. Today, a 68060 Amiga would likely give all day batter life and a 68060+AA+ SoC ASIC would likely cost less than a meal and maybe even a drink to mass produce.

"But is it IBM compatible?"

The Commodore LCD was yet another example of throwing something at the wall to see if it sticks. Where was the market? "Build it, and they will come?". I watched a video of the prototype in operation. The LCD display was just as dire as I thought it would be. Nobody would subject themselves to that unless they had to.

In the 1980's I was supplied with an NEC PC8201a in my job as a telephone exchange technician. We used them to program systems that communicated via a serial port, as a portable replacement for a teleprinter. I designed a parallel port interface that automated the process of reloading configuration data on Mitel PABX systems (which took hours when done manually). The LCD screen was hard to read and the BASIC interpreter wasn't fast, but it did the job. One great feature was the file system which stored programs and data in its 32k of battery-backed RAM. Without that it would have been far less useful.

The PC8201 was a fairly rugged tool that was made for the kind of jobs we used it for. Its extremely low power draw gave it an operating battery life of around 5 days on 4 AA dry cells, and a standby time of over a year. This meant we didn't have to plug it into the mains or worry about rechargeable batteries dying. And it was lightweight too. Would the Commodore LCD have done the same job? Perhaps, but that fold-out screen looks flimsy.

@bhabbott: IBM compatibility doesn't matter.

Other systems survived despite the PCs.

@matthey

Quote:

matthey wrote: Power management, LCD driver, CMOS chipset with integration (preferably SoC), low power CPU (3.3V 68060 or 68040V), reskinned GUI like CDTV and CD32, etc. Yea, well beyond the competence level of CBM management.

A PDA should have been something like a Palm Pilot, monochrome B/W , 200 x 320 BW (that would have been nearly a tablet in that time) but it would not had the Amiga chipset (no color, no OCS/ECS/AGA), 128kB RAM so most games wouldn't had worked and the GUI would had been totally different to be pen compatible

That's what I meant with: "You wouldn't had recognized it as an Amiga"

What you posted are pictures of C64 or other PETs from Commodore, that might have worked because you could have 320x200 monochrome ( maybe 4 shades of gray) displays and a connector for a monitor at the back. The price would have been much higher than for a standard C64 but maybe they would have got enogh people buying it.

An Amiga on the other hand, with such a slow and sluggish low res LCD would have made little sense.
The LCD as accessory could have been a product but not a battery driven laptop, that would have been a A1200 for a price of an A4000.

Therew where some official AtariST Laptops:
https://en.wikipedia.org/wiki/Atari_STacy
https://en.wikipedia.org/wiki/ST_Book

It was easier for Atari, because it was less focused on games and GEM could easily support 3rd party GFX, but the Notebooks where expensive.

Last edited by OneTimer1 on 27-Aug-2024 at 10:10 PM.

amigang Quote:

I think your forgetting if commodore had survived and went this direction and focus was a low cost computer, then you would do what you can to bring down costs. So like reduce the aga chipset to one chip.

Amiga 1200 chipset (CBM cost of $11 near bankruptcy)
Alice - DMA, blitter, copper
Lisa - display, sprites, pallette
Paula - audio, I/O
CIAx2 - timing, I/O
Gayle - IDE
Budgie - PCMCIA

Amiga CD32 chipset (CBM cost of $6 near bankruptcy)
Alice - DMA, blitter, copper
Lisa - display, sprites, palette
Paula - audio, I/O
Akiko - CIAx2, CD-ROM, I/O, c2p

AA+ (planned for 1994)
Alice+Lisa replacements integrated?
Paula replacement? (maybe Paula and Akiko+ chip could be integrated)
Akiko+? (May be possible to integrate Paula+ and Akiko+ but analog Paula adds a challenge)
IDE and PCMCIA functionality integrated? (maybe could be integrated into Akiko+ but pin count high)

68k SoC (planned for early 1995)
68k CPU licensed from Motorola with Amiga chipset (single chip)

Considering CBM failed to integrate any of the core 3 chips of the chipset and only Lisa had been upgraded to use a semi-modern CMOS process, the early 1995 date in the Commodore post bankruptcy docs seems ambitious for CBM. CBM knew what to do to cost reduce, shrink/integrate and lower the power of the Amiga at least.

amigang Quote:

I would have also sold just a Amiga motherboard. I mean look at these costs listed in these docs https://archive.org/details/amiga-computer-cost-estimates-1991-04-16/mode/2up A500 cost $177 A3000+ cost $1028 You can see one the big cost was the cpu, floppy disk drive, etc so they could of made a board maybe without a cpu for very cheap and users / business could add cpu card.

CBM no doubt sold boards and other separate parts but to businesses rather than directly to customers. Some high end Amiga 4000(T) models did not have a CPU on the motherboard as the 68040 was expensive but it was cheap to include a 68EC030 on the motherboard by this time. CBM cost on the 68EC020 was $8 while the 68EC030 would not be much more than this and was useful for diagnosis purposes as a motherboard CPU is more reliable than a plug-in accelerator card.

The Commodore post bankruptcy docs give AGA motherboard costs but no date.

parts | A4000/040/6MiB | A1200 | CD32
chips $20 $11 $6
DRAM $186 $50 $50
PCB $33 $19 $12
CPU $170 $8 $8
other $207 $50 $67
---
PCB+ $616 $138 $143

amigang Quote:

I did a quick calculation based on these docs they could of made an A3000+ motherboard without CPU & disk drive keyboard etc, just the board at cost for around $516. Sell it at $750, again I think it would of done well. Again A500 just the board you could get down to $109.

The A3000+ motherboards likely would have sold as upgrades for Amiga 3000 users which would have enlarged the AGA install base, especially if CBM limited their profit margin to something reasonable. One problem though, is that the Amiga 3000 case did not include a 5.25" bay for a 68EC020+AGA+CD-ROM standard but poor planning meant the Amiga 1200 didn't get the Akiko chip either. There was no official Amiga 4000 CD-ROM upgrade for the standard either.

The Commodore post bankruptcy docs "Pricing Trend" for 12/93 has the A500 down to $122, A600 down to $191, A1200 down to $292 and A4000/040 down to $1659. The A500 may have sold in Eastern Europe but it was too late to further cost reduce or upgrade so late in life. The A600 should have been cancelled and the A1200 and CD32 cost reduced and upgraded. With better planning, AGA would have been introduced with 68EC030@28MHz, AA+ and Akiko features and integration for about the same cost as the A1200 and CD32 were introduced at. Chipset development had upfront costs but then provided the best cost reduction of fewer chip transistors and pins used by integrating chips, smaller cheaper PCBs, fewer components, cheaper power supplies, cheaper cooling (fans and heat sinks) and cheaper cases. This is why SoCs are so popular today. RPi SBCs are available directly from the producer and they are cheap enough to be tossed rather than wasting time on diagnosing problems. CBM was good at cost reducing hardware but their modus operandi was to introduce a product and later replace it, failing to recognize the importance of compatibility that could be maintained while upgrading better designed and newer technology like the 68k Amiga. They realized but too late to enhance and cost reduce the Amiga into RPi like hardware. The same RPi like (50+ million unit) niche was open back then as x86(-64) hardware did/does not scale as low as 68k hardware. ARM Thumb hardware may scale a little lower than 68k hardware but is weaker performance while AArch64 hardware does not scale as low as 68k hardware.

@bhabbott & OneTimer1
CBM was often late to the game and their products often lacked quality which is why the quality 68k Amiga was a bad fit for the low cost hardware producer CBM. Their LCD prototype looks cool for the time and shows they had some manufacturing skills but there were already similar and better products on the market that were not 6502 handicapped. The same was true of the C900 with Z8000 Unix systems being available 4-5 years before them and during the transition to a large flat 32-bit address space which was also possible 5-6 years earlier with the 68000. The CBM LCD and C900 are often considered their two best unreleased products and both were obviously close to release with full working prototypes. CBM lacked leadership and vision even before the Amiga. Some people today think the 6502 family CPUs and using old chipsets to reduce costs was the answer but the opposite is true. Jay Miner had the right vision which is to plan wisely, enhance, integrate and conquer.

Last edited by matthey on 28-Aug-2024 at 01:36 AM.
Last edited by matthey on 27-Aug-2024 at 10:55 PM.