I refer people who complain about redditors to this site and to this thread specifically.

@bhabbott

Quote:

bhabbott wrote: @Kronos The 20/20 hindsight of Amiga fans is amazing, especially since no two are the same. Imagine if they all hopped into a time machine, went back to 1984 and replaced Commodore's board of directors with themselves. Would this committee of geniuses each primed with 'proper planning and having a strategy' have been able to agree on anything?

Of course not. That’s exactly why you don’t put all of us in that time machine.

Thanks for pointing out and reaffirming that the proverbial grass was not all that greener elsewhere.
I’m no C= fanboy, I just can’t abide this idea of C= stupid and bad, and every other company smart and good.

Love your A2000 remarks, and pretty much everything in your response is well reasoned.

_________________
All the way, with 68k

bhabbott Quote:

As for purchasing the Amiga being a 'hail Mary' for Commodore, even before the split they realized the need to go 16 bit. Amiga corp were the ones looking for a 'hail Mary', and Commodore provided it - while getting the tech they needed to produce those 'next generation' machines they were going to make anyway, but with way less design effort.

Without the purchase of Amiga Corporation, I can't see the rudderless C= designing Amiga like hardware until the 1990s when it would have been too little too late. The Amiga was not a good fit for C= philosophy wise even though they were a good fit capability wise. C= remained focused on repackaging the C64 and trying to compete with IBM for business computers. They tried to make the Amiga into a C64 and a business computer. Putting a PC clone inside the Amiga gives a more expensive PC clone.

bhabbott Quote:

The 20/20 hindsight of Amiga fans is amazing, especially since no two are the same. Imagine if they all hopped into a time machine, went back to 1984 and replaced Commodore's board of directors with themselves. Would this committee of geniuses each primed with 'proper planning and having a strategy' have been able to agree on anything?

Director voting decides arguments with an odd number of directors. The big problem remains that Irving Gould can likely remove directors (and CEOs) he doesn't want with a stockholder vote.

bhabbott Quote:

The A2000 (or something like it) was the obvious and necessary design to counter the PC. It didn't need 'Ranger' chipsets (which didn't do much more than OCS), it needed 100% Amiga compatibility in a nice big case with slots like the PC - and that's what it got. If by 'with 68020', you mean on the motherboard, that would have been an expensive mistake. Instead they did the sensible thing and put a CPU card slot in the A2000, then a year later released a 68020 card and then a 68030 card. They also put the video slot in it which allowed use of a flicker fixer to provide VGA output with 100% Amiga compatibility. So you could buy an A2000 with single floppy drive for a 'low' price, then upgrade it as your budget permitted - and still be able to operate it like a stock A1000/A500 to play those games that wouldn't work on a 68020.

I doubt Jay Miner was opposed to the German team Amiga 2000 expansion design and case as they had made improvements over the Amiga Corporation Ranger design. I still think the Amiga 2000 should have included a 68020 and Ranger chipset though. The 68020 was introduced in 1984 so it was already 3 years old in 1987 when the Amiga 2000 was introduced. A 68020 would have been better supported if it came out earlier signaled and would have been a signal to developers that they should support upgraded 68k CPUs. Yes, there were incompatibilities but was it better for software to only support the 68000+OCS standard for many years instead? OCS was good enough to compete with PC clones but VRAM upgrades closed the gap and then faster busses allowed higher resolutions, better refresh rates and better chipset performance than DRAM allowed dropping the Amiga to mid-performance instead of high performance by the late 1980s. The best option would have been a Ranger chipset which could support VRAM or DRAM like was planned with AAA and then a high res or low res daughter board option could be chosen and changed out like on the C900. The Ranger chipset would have allowed the Amiga to be more competitive and compatible with VGA, including the use of cheaper monitors. Without VRAM for the integrated chipset, much more expensive graphics cards were required to provide competitive high performance for the Amiga. The low end Amiga computers could receive the 68020 and Ranger upgrade 1-3 years later when it was cheap enough and the high end could then move on to the 68030 and Ranger+ upgrade when ready. Apple incrementally upgraded their 68k CPUs and chipsets eventually surpassing the Amiga while the Amiga was still on the 68000+OCS/ECS standard. By the time the 68EC020+AGA standard came out, Apple was transitioning to 68040 CPUs and using VRAM in their graphics cards in 1991.

https://en.wikipedia.org/wiki/Macintosh_Quadra_700#Hardware Quote:

Video: Like the IIci, the 700 has integrated graphics built into the system board but, unlike the earlier model, it uses dedicated VRAM for its video memory. The onboard video came with 512 kilobytes of VRAM soldered to the motherboard, and supported resolutions up to 1152x870. The video memory was expandable to 2 megabytes via six 256-kilobyte 100ns VRAM SIMMs in each of the VRAM SIMM expansion slots on the motherboard. Expanding the video memory to 2 megabytes allowed for 24-bit (Millions) color at resolutions up to 832x624.

The Amiga was still using 150ns DRAM for chip memory when Apple was using 100ns dual ported VRAM with twice the memory bandwidth. With AGA that became 80ns DRAM. With 2MiB of VRAM, the 1991 Quadra could handle 832x624 in 24 bit (millions of colors) and 1152x870 in 8 bit (256 colors). There was no high end Amiga even with AGA without buying a $500+ graphics card with VRAM and this was much less compatible.

bhabbott Quote:

The A2000 was a great workhorse for use by hobbyists and professionals alike. That's why NASA chose it over the PC for their telemetry jobs, why Newtek chose it for the Video Toaster, and why so many 3rd party developers jumped at the chance to make cards for it. I made a card for my friend's A2000 so it could use a PC hard drive and controller in an AT slot (pity I didn't develop that idea further, could have been a big seller!).

The Amiga 2000 was an expandable workhorse but that doesn't mean it couldn't have been better.
The inactive ISA slots did see use for both PC bridgeboards and the Toaster for powering cheaper PC ISA cards but Zorro cards cost significantly more because they had to extend across the ISA slots.

bhabbott Quote:

Perhaps you are forgetting that Commodore had already produced the Sidecar? I had one on my A1000 - a very awkward design. Putting it on a card that hid inside the machine was a brilliant idea. Yes, the original Bridgeboard was 'under powered', but it did provide the IBM compatibility many users needed, without costing more than an actual PC. Like it or not, the fact was that PC compatibility was essential to most businesses. One must also remember that in 1987 4.77MHz XTs were still very popular in the business world. In 1989 Commodore released the A2286AT bridgeboard with 8MHz 80286, 1MB RAM and 1.44MB floppy controller, which is similar to what most PCs had at the time. Pop an EGA or VGA card into a 16 bit slot and you had a pretty good PC hiding inside your Amiga!

Somewhere between a low clocked 68020 and high clocked 68030 is the ability to emulate a 8088@4.77MHz at full performance. The C= supplied Amiga Transformer software was horrible performance, buggy and didn't work on a 68020 but later emulators like PCTask showed it was possible. A 68020@14MHz should have provided at least half the emulation performance of a 8088@4.77MHz which was good enough for most business software. At that point, a hard drive would have added more convenience for business use but the Amiga 2000 didn't get that built in either. Amiga business computers didn't come standard with hard drives until many years later, much later than IBM business computers (hard drive optional on PC and standard on XT). Sorry, I would have rather had more general purpose standard Amiga features than specialized PC compatible Amiga 2000 features.

bhabbott Quote:

The CD32 launch was relatively successful in the places where it was launched. In the UK it sold a lot more than its rival, the Sega Mega CD. Unfortunately it missed in North America due to the XOR troll. One might say that not buckling in to that extortion was a mistake, but I bet most of us would have resisted too.

There was a big Amiga store in Canada that likely had NTSC CD32s available through mail order and the C= sales and distribution network in the U.S. was so bad that C= may have been better off staying out of the U.S. rather than pay the ridiculous XOR troll. That is said knowing the U.S. is normally the best console market in the world.

bhabbott Quote:

As for 'proper planning and a strategy', the real world has a tendency to screw up even the best-laid plans. Sometimes it's better to just create something, put it out there and see what happens, a strategy that worked for Commodore in the past. Problem was their financial position was severely weakened when Jack left, and never really recovered. Then in 1992 the bottom dropped out of the 'alternative' home computer market as PC sales picked up, thus the urgent need to pivot.

VRAM graphics cards and hard drives weren't enough alone to push PC sales but then came higher performance CPUs to push data with brute force. C= Amiga was selling 68000 CPUs when PC clones were using 386s and 68EC020 CPUs when PC clones were using 486s and Pentiums. C= had confidence the Amiga chipset could offload the CPU but they didn't upgrade the chipset enough either.

bhabbott Quote:

An Amiga console in eg. 1987 would have been a huge mistake. But in 1993 it was overdue. Commodore actually started designing the CD32 before the ink was dry on the A1200, but it was also too late. Why? Because Commodore had a plan and a strategy. The plan was to make the Amiga way more powerful to compete against PCs, while milking the A500 and C64 for all they were worth in the mean time. This strategy worked, but not for as long as expected. So you can ding Commodore for misreading the market, but so did many others. Commodore even hired an expert who specialized in forecasting trends in the PC market, and he failed to see what was coming.

Why would a follow up console a year after the Amiga 500 was released be worse than a follow up console a year after the Amiga 1200 was released? Was the financial expert C= hired Mehdi Ali? While PC clone sales "picked up", why was the alternative and unique Amiga market more resilient for C= in 1993 than their PC clone business?

@bhabbott

Quote:

bhabbott wrote: The A2000 (or something like it)

"something like it" sure, just not what C= Germany build, as that was not only a bad design but also BROKEN hence it had to be reengineered AFAP resulting in the still stupid 2000B

Quote:

was the obvious and necessary design to counter the PC. It didn't need 'Ranger' chipsets (which didn't do much more than OCS), it needed 100% Amiga compatibility

100% Amiga compatibility meant 0 process and an open invitation to directly hit the HW as all Amigas are always all the same...

Quote:

in a nice big case with slots like the PC - and that's what it got. If by 'with 68020', you mean on the motherboard, that would have been an expensive mistake. Instead they did the sensible thing and put a CPU card slot in the A2000, then a year later released a 68020 card and then a 68030 card. They also put the video slot in it which allowed use of a flicker fixer to provide VGA output with 100% Amiga compatibility.

yeah yeah CPU slot..... but thats not the point, as it did come with an 8 year old CPU just the same as the kiddy-Amiga only at much higher price and by virtue of the 8 year old CPU all those slots got limited to 16Bit and a 16MB addressrange (half of which wasn't assigned to the slots).


What you all are getting wrong is that C= created all the great and the dumm things they did out of pure chaos and not out of some grand plans failing or succeeding, hence a proper planned console launch never was in the cards.

As for the CD32 being a "good idea" given the limited resources at that time I'd say it would have been much smarter to throw them at an A1200 aka something with full 32Bit design, some FAST-RAM, a proper place to put a 3.5" HD (and having it ship as standard) and maybe some sort of basic scandoubler so one could use an 31kHz monitor.

Didn't happen, because for people at C= it was always "cooler" to go for a new direction that fixing issues with product that already ship.

_________________
- We don't need good ideas, we haven't run out on bad ones yet
- blame Canada

In the end, the hanging on to 68000 paid off for us users, as that is still _today_ the most available 68k on ASIC, and the 68k with most and best reimplementations both when it comes to FPGA softcores and software emulation.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

Quote:

100% Amiga compatibility

How is this measured? Compaible with what exactly? Which Amiga?

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@Kronos

Quote:

As for the CD32 being a "good idea" given the limited resources at that time I'd say it would have been much smarter to throw them at an A1200 aka something with full 32Bit design, some FAST-RAM, a proper place to put a 3.5" HD (and having it ship as standard) and maybe some sort of basic scandoubler so one could use an 31kHz monitor.

It would have been wiser to double down on the CD32 with a SX32 Pro type SKU from the get go. A CD-Rom and hard drive based/multibutton joypad supporting Amiga was better birthed out of the CD32 than a A1300 in my opinion!

_________________
"Art challenges technology. Technology inspires the art."
John Lasseter, Co-Founder of Pixar Animation Studios

Kronos Quote:

100% Amiga compatibility meant 0 process and an open invitation to directly hit the HW as all Amigas are always all the same...

The C64 didn't change so the C64 replacement, the Amiga, didn't need to progress either, according to C= thinking. Jay Miner designed the 68k Amiga hardware to be upgradeable and expandable though. There was clearly a philosophy conflict here.

Kronos Quote:

yeah yeah CPU slot..... but thats not the point, as it did come with an 8 year old CPU just the same as the kiddy-Amiga only at much higher price and by virtue of the 8 year old CPU all those slots got limited to 16Bit and a 16MB addressrange (half of which wasn't assigned to the slots).

That would be limited to a 24 bit address range which the 68000 and 68EC020 could address (68000: 16 bit data bus, 24 bit address bus; 68020: 32 bit data bus, 32 bit address bus). It doesn't look like the 68EC020 was available until 1991 so they couldn't have used the downgraded embedded version of CPUs like C= was known to do in their supposedly "high end" big box Amigas. The Apple Mac II was introduced in 1987 using a 68020 and the IIx was introduced in 1988 with a 68030. The Amiga 2500/20 with 68020 shipped in 1988 and the Amiga 2500/30 with 68030 in 1989. C= could have introduced the Amiga 2000 with a 68020 on the CPU card and no motherboard CPU like they did with the cost reduced Amiga 4000. Full 32 bit addressing would have made the Amiga 2000 expansion much more powerful like the Amiga 3000 and Amiga 4000. Graphics cards are handicapped on the Amiga 2000 Zorro II bus.

Kronos Quote:

What you all are getting wrong is that C= created all the great and the dumm things they did out of pure chaos and not out of some grand plans failing or succeeding, hence a proper planned console launch never was in the cards.

C= had a plan. The low end was the as cheap as possible C64 replacement Amiga 500 and the high end was the as cheap as possible C64 replacement with a PC stuck inside it Amiga 2000.

Kronos Quote:

As for the CD32 being a "good idea" given the limited resources at that time I'd say it would have been much smarter to throw them at an A1200 aka something with full 32Bit design, some FAST-RAM, a proper place to put a 3.5" HD (and having it ship as standard) and maybe some sort of basic scandoubler so one could use an 31kHz monitor. Didn't happen, because for people at C= it was always "cooler" to go for a new direction that fixing issues with product that already ship.

If C= had upgraded the AGA or AA+ chipset to run at 28MHz with a 68030@28MHz like it looks like was planned, they wouldn't have needed a scandoubler. VRAM wouldn't have needed a scan doubler either. If C= had planned better, it may have been easier to support a CD-ROM drive as well as a 3.5" hard drive. Yea, C= needed better overall planning which should have resulted in fewer models to boost mass production. I think the CD32 was a good idea and would have been that much better with the 68030 and chipset running at 28MHz as planned. The Amiga 1200 would have received the same upgrade and been more competitive as well. There is no excuse for using NMOS Amiga chipset chips on a mid 1970s chip process that had not been upgraded by 1993.

Last edited by matthey on 28-Sep-2023 at 05:37 PM.
Last edited by matthey on 28-Sep-2023 at 01:37 AM.
Last edited by matthey on 28-Sep-2023 at 01:29 AM.

BigD Quote:

It would have been wiser to double down on the CD32 with a SX32 Pro type SKU from the get go. A CD-Rom and hard drive based/multibutton joypad supporting Amiga was better birthed out of the CD32 than a A1300 in my opinion!

A CD32 was simpler than an Amiga 1200 and the production cost likely would have dropped below the cost to produce an Amiga 1200 before long. It would save the trouble of producing a variety of keyboards with modifications to support different languages. A little nicer CD32 case design may have been in order as it was not as sexy as the Amiga 1200. I doubt low profile CD-ROMs were available back then but the following prototype is nice.



For a modern Amiga, the internal floppy drive could be eliminated (an optional retro I/O board and/or GPIO could add an external floppy port). We'll have to see what Retro Games Limited comes up with for an Amiga Maxi but expect a nice facade to hide emulation on weak ARM hardware inside. With the CD32 controller, It would be nice if they could add support for CD32 .ISO images at least. I'm not interested in buying weak emulation hardware regardless but it would boost demand for CD32 compatible .ISO images and that was the best real Amiga hardware (CD32 is upgradeable to a 68060).

Last edited by matthey on 29-Sep-2023 at 06:42 PM.
Last edited by matthey on 28-Sep-2023 at 01:27 AM.

@Kronos

Quote:

Kronos wrote: @bhabbott What you all are getting wrong is that C= created all the great and the dumm things they did out of pure chaos and not out of some grand plans failing or succeeding, hence a proper planned console launch never was in the cards.

And what you refuse to understand is that most other companies didn't have grand plans and still found success.

And you refuse to define "success", or what a "proper planned console launch" would be in 1993/94, which we could use to contrast it with other contemporary successes and launches.

You'd just rather stick to your narrative of how C= had the shitty version of a Midas Touch. So much chaos at this company that they'd be better off being managed by a Magic 8 Ball.

Last edited by agami on 28-Sep-2023 at 02:45 AM.

_________________
All the way, with 68k

@matthey

“ARM emulation?” You obviously mean 68k emulation.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@Thread

The more we talk about this the more I simply detest Commodore. The brought Dave Haynie and George Robbins onboard to work on the Amiga tech but sidelined the original team other than Carl Sassenrath with the CDTV abomination! I would never wear a C= logo with pride unless the Amiga trademark was front and centre!

Here is to the CD32!

:pint

...as good a last hurrah as Commodore were ever going to get! :

_________________
"Art challenges technology. Technology inspires the art."
John Lasseter, Co-Founder of Pixar Animation Studios

BigD Quote:

The more we talk about this the more I simply detest Commodore. The brought Dave Haynie and George Robbins onboard to work on the Amiga tech but sidelined the original team other than Carl Sassenrath with the CDTV abomination! I would never wear a C= logo with pride unless the Amiga trademark was front and centre! Here is to the CD32! :pint ...as good a last hurrah as Commodore were ever going to get! :

Amiga owners are not as proud of the chicken lips as the 8 bit C= owners but then the Amiga happened in the post Jack era. It looks like Jay Miner remained a technical advisor to C= which shows his calm demeanor and peaceful nature even after being marginalized.

https://timenote.info/en/Jay-Miner Quote:

Jay worked at Commodore-Amiga for several years, in Los Gatos, California. They made good progress at the beginning, but as Commodore management changed, they became marginalised and the original Amiga staff was fired or left out on a one-by-one basis, until the entire Los Gatos office was closed. Miner later worked as a consultant for Commodore until it went bankrupt.

Hot heads that left the Easter egg, "We made Amiga, they fucked it up" in Workbench 1.2 just made the rift worse. Jay wasn't satisfied with the Amiga treatment either but there are worse scenarios like Atari getting the Amiga where Jack said he would fire the developers. The best scenario may have been if Nolan Bushnell had not sold out Atari to Warner Communications and had been more open to personal computers and not just consoles and arcade games. Not only did Jay Miner and the Amiga slip away from Atari but Jobs offered Nolan a 1/3 stake in Apple for $50,000 which he turned down. Ray Kassar pushed Nolan, the co-founder of Atari, out replacing him as CEO. He was toxic causing a mass employee exodus at Atari but sales went from $75 million in 1977 to over $2.2 billion just three years later partially due to his marketing and making him untouchable. It's somewhat similar to Jack being pushed out of C= except Nolan didn't sabotage his business before he was pushed out. The video game crash of 1983 would have caused uncertainty at Atari if they were developing the Amiga.

Small businesses tend to be more creative and flexible than large businesses but what Amiga Corporation was trying to do as a fabless semiconductor business was expensive back then.

https://semiwiki.com/semiconductor-manufacturers/1535-a-brief-history-of-the-fabless-industry/ Quote:

Dr. Morris Chang worked for one of the aforementioned IDMs (Texas Instruments) and went on to found Taiwan Semiconductor Manufacturing Corporation (TSMC). At TI, Morris Chang worked on a four transistor project where manufacturing was done by IBM. This was one of the early foundry relationships. At the same time Morris pioneered the then controversial idea of pricing semiconductors ahead on the cost curve, sacrificing early profits to gain market share to achieve manufacturing yields that would result in greater long-term profits. Morris also noticed in the early 80’s at TI that top engineers were leaving and forming their own semiconductor companies. Unfortunately the heavy capital requirement of semiconductor manufacturing was a gating factor. The cost back then was $5-10M to start a semiconductor company without manufacturing and $50-100M to start a semiconductor company with manufacturing. Some of these engineers went to the IDMs to get wafers from excess capacity but this was not a customer friendly process and sometimes they were getting wafers from a competitor.

Amiga Corporation didn't just want to be a fabless chip developer but also a personal computer and console manufacturer which makes it tough to stay independent. Interest rates and business loan rates were much higher in the early to mid 1980s than they are currently even though they are considerably off their recent lows. In a better economic environment, an IPO may have been an option for Amiga and Nolan Bushnell considered it for Atari before selling to Warner. A Nolan Atari Amiga with an IPO would have been interesting. The early to mid 1990s were a tough business environment even for large businesses.

We may not have covered all the biggest console flops in the early to mid 1990s.

1990 C= C64GS https://en.wikipedia.org/wiki/Commodore_64_Games_System

1990 CD-i https://en.wikipedia.org/wiki/CD-i
Co-developed by Phillips and Sony. Nintendo licensed games.

1991 C= CDTV https://en.wikipedia.org/wiki/Commodore_CDTV

1993 3DO https://en.wikipedia.org/wiki/3DO_Interactive_Multiplayer
Supported by Electronic Arts, Panasonic, Sanyo, GoldStar and Creative Technology

1993 Atari Jaguar https://en.wikipedia.org/wiki/Atari_Jaguar

1994 Sega 32X https://en.wikipedia.org/wiki/32X

1994 SNK Neo Geo CD https://en.wikipedia.org/wiki/Neo_Geo_CD

1995 Nintendo Virtual Boy https://en.wikipedia.org/wiki/Virtual_Boy

1996 Apple Pippin https://en.wikipedia.org/wiki/Apple_Pippin
Bandai the large toy company produced them.

There were some big businesses backing some of these failures.

Last edited by matthey on 30-Sep-2023 at 04:18 AM.

@Kronos

Quote:

Kronos wrote: [quote]"something like it" sure, just not what C= Germany build, as that was not only a bad design but also BROKEN hence it had to be reengineered AFAP resulting in the still stupid 2000B

The A2000 wasn't 'broken'. It was based on the A1000 chipset because that's what they had at the time. The A500 got a new chipset, which was then used in the 'B2000-CR' model. This is how products are typically developed. For example the IBM PC was followed by the PC-XT, which was the exact same design except for more onboard RAM and expansion slots increased from 5 to 8 (and removing the cassette port).

Quote:

100% Amiga compatibility meant 0 process and an open invitation to directly hit the HW as all Amigas are always all the same...

...which was typical of home computers of the time. The PC was also very sensitive to this issue, as the low performance CPU made the BIOS too slow for acceptable video speed. Games typically hit the hardware directly for best performance, so graphics cards had to be 100% compatible with the original IBM VGA (and CGA and EGA) including the over 300 registers in the video gate array if they wanted games to run on them. This requirement even extended to PCI and beyond.

Quote:

yeah yeah CPU slot..... but thats not the point, as it did come with an 8 year old CPU just the same as the kiddy-Amiga only at much higher price and by virtue of the 8 year old CPU all those slots got limited to 16Bit and a 16MB addressrange (half of which wasn't assigned to the slots).

Same as the A1000 you mean. Yes, the 68000 was 8 years old, but the 8086 was even older (and slower). Most PCs at that time had an 8088. The IBM PC-AT had an 80286 running at 6MHz with 1 wait state, making it about the same speed as the Amiga's 7.2MHz 68000. In 1987 IBM introduced the PS/2 series, including the model 25 and model 30 which had an 8MHz 8086 CPU and 720 KB floppy drives. Another popular line at that time in the UK was the Amstrad PC1512 and 1640, both of which had an 8MHz 8086. In 1988 Amstrad introduced the PC2086 with 8MHz 8086 and onboard VGA. I have one, it's a dog!

In 1987 8MB was a huge amount of RAM. The Compaq Deskpro 386 (first 32 bit PC) was introduced in late 1986 with 1MB RAM for US$6,499 (equivalent to over $17,000 today). It could take a maximum of 2MB on the provided daughterboard, and 10MB with another 8MB board plugged into the daughterboard. In comparison Commodore's A2630 card (introduced in 1989) could take 112MB of 32 bit RAM.

Quote:

What you all are getting wrong is that C= created all the great and the dumm things they did out of pure chaos...

No, it was lack of CAOS that made the Amiga what it was. :)

Quote:

...and not out of some grand plans failing or succeeding, hence a proper planned console launch never was in the cards.

In 1992 Commodore had more of plan than Sony did.

Sega's plan started with the SC-3000 and a cut down console model called the SG-1000. The SC-3000 was moderately successful (where they bothered to sell it) but the SG-1000 was a flop. The Sega Master System - which was successful - was a later development of it.

Atari had a plan with the XEGS. It sold a miserable 100,000 units in its 5 year lifespan (1987-1992). Atari had a plan with the Jaguar too (released in the same year and month as the CD32). It didn't do much better.

Quote:

As for the CD32 being a "good idea" given the limited resources at that time I'd say it would have been much smarter to throw them at an A1200 aka something with full 32Bit design, some FAST-RAM, a proper place to put a 3.5" HD (and having it ship as standard) and maybe some sort of basic scandoubler so one could use an 31kHz monitor.

Wonderful idea. Let's produce a model that costs a lot more than the A1200 so everyone will buy a PC instead.

Well it turns out they did make that model - and called it the A4000-030. Sold like hot cakes, right?

Quote:

Didn't happen, because for people at C= it was always "cooler" to go for a new direction that fixing issues with product that already ship.

Except it did happen (see above).

If you are suggesting they should have 'fixed issues'' with the A1200 instead of producing the CD32, I'm calling BS. The A1200 didn't have any issues to speak of.

Last edited by bhabbott on 02-Oct-2023 at 02:25 AM.
Last edited by bhabbott on 02-Oct-2023 at 02:22 AM.

@matthey

Quote:

matthey wrote: I doubt Jay Miner was opposed to the German team Amiga 2000 expansion design and case as they had made improvements over the Amiga Corporation Ranger design. I still think the Amiga 2000 should have included a 68020 and Ranger chipset though. The 68020 was introduced in 1984 so it was already 3 years old in 1987 when the Amiga 2000 was introduced.

Do you even know what was in the Ranger chipset? The actual Ranger prototype shows the reality of its progress - an awkward design with no new custom chips in sight. Commodore went with the German-designed A2000 instead, and looking at that prototype I can see why.

Quote:

A 68020 would have been better supported if it came out earlier...

That would have been difficult, and perhaps a little foolish. The 68030 was announced in 1986, and by 1987 was about the same price as the 68020 - but faster and with built-in MMU. Better to put a cheap 68000 on the motherboard and provide a slot for future accelerator cards.

Quote:

...and would have been a signal to developers that they should support upgraded 68k CPUs.

Developers should already have been aware of the requirements for compatibility, and even if they weren't it wouldn't take much to fix any issues provided they had followed Commodore's guidelines. One of the biggest offenders here was Microsoft. They ported their Macintosh BASIC to the Amiga complete with code that wasn't 32 bit complaint (using upper 8 bits of addresses for some other purpose). Amiga BASIC also had a bug that caused it to crash on the 68020 due to a malformed instruction that just happened to still work on the 68000.

Quote:

Yes, there were incompatibilities but was it better for software to only support the 68000+OCS standard for many years instead?

In 1993 new games were still coming out for the A500 that were so good you had to look closely to tell they weren't AGA. The OCS chipset's 7 year lifespan was similar to other gaming platforms such as the Sega Megadrive and Sony PlayStation. Apogee were still insisting that their developers produce games for EGA in 1992, because they didn't want to alienate users with older machines. EGA was introduced in 1984 with the PC-AT, 8 years earlier.

Some Amiga software doesn't support 68020, but it is mostly games that failed for other reasons too, such as expecting ChipRAM in certain places, being incompatible with later OS's etc. The huge number of games we can now run under WHDload show that many of the issues were not that hard to fix. The real problem was lack of support from publishers, exacerbated by rampant piracy which gave them less incentive to provide it. Also many developers moved on and source code was lost.

The sad fact of the matter is that yes, compatibility with existing software was vitally import - even for software that broke the rules. It was the same on the PC. Right from the start, a PC clone's success was dependent on it being '100% IBM compatible'. Which of course was also the reason the Amiga was bound to fail - since it wasn't and couldn't be.

bhabbott Quote:

Do you even know what was in the Ranger chipset? The actual Ranger prototype shows the reality of its progress - an awkward design with no new custom chips in sight. Commodore went with the German-designed A2000 instead, and looking at that prototype I can see why.

Jay Miner didn't have the go ahead to make prototype Ranger custom chips. MOS/CSG would have a much shorter turnaround time as they weren't outdated circa 1986. See the 6 layers of breadboards in the prototype? The bottom layers are filled up when possible and only another layer is added when needed. That is why the top layer doesn't have much. There aren't as many wire wrapped breadboards as original Amiga prototypes but some of the custom chips may not have changed and more integrated support electronic parts and chips were available. Even if it not a fully working prototype, it was likely enough to test some of the logic. The most important point which you seem to miss is the spec. There is little doubt that Jay Miner could have made the Ranger chipset close to what was planned. I believe the Ranger chipset would have been a game changer, especially if coupled with a 68020. It could have displayed 1024x1024 in 16 colors so maybe the Amiga would have gained the desktop publishing market instead of the Mac. It would have been more competitive with (S)VGA resulting in the Amiga receiving more and better PC ports, nearly identical if 128 color plus EHB mode were added to give 256 colors. (S)VGA monitors would be usable saving cost and eliminating the unprofessional interlace flicker. Not only high end professional users would have bought high end Amigas but existing Amiga users would have an upgrade path to higher performance Amigas which had better margins for C=. DRAM gfx memory was for the low end and VRAM was for the high end so C= didn't have a high end. A bridgeboard attached to a C64 doesn't make a professional high end or business computer anymore than the C64 replacement attached to a bridgeboard. C= upper management was brain damaged and Jay Miner was right. Hind sight should be 20/20.

bhabbott Quote:

That would have been difficult, and perhaps a little foolish. The 68030 was announced in 1986, and by 1987 was about the same price as the 68020 - but faster and with built-in MMU. Better to put a cheap 68000 on the motherboard and provide a slot for future accelerator cards.

The 68020 and 68030 compatibility means it wouldn't take much to upgrade to the 68030 which should have been after the stock of 68020s ran out. The CPU could have been on the CPU card but there is a small increased chance of problems with the contacts (corrosion or inserting/removing). The AmigaOS was not using the MMU so that was not a major factor.

bhabbott Quote:

Developers should already have been aware of the requirements for compatibility, and even if they weren't it wouldn't take much to fix any issues provided they had followed Commodore's guidelines. One of the biggest offenders here was Microsoft. They ported their Macintosh BASIC to the Amiga complete with code that wasn't 32 bit complaint (using upper 8 bits of addresses for some other purpose). Amiga BASIC also had a bug that caused it to crash on the 68020 due to a malformed instruction that just happened to still work on the 68000.

C= was better at providing programming guidelines later. They didn't even provide adequate programming documentation early which resulted in less Amiga software and slower Amiga adoption.

bhabbott Quote:

In 1993 new games were still coming out for the A500 that were so good you had to look closely to tell they weren't AGA. The OCS chipset's 7 year lifespan was similar to other gaming platforms such as the Sega Megadrive and Sony PlayStation. Apogee were still insisting that their developers produce games for EGA in 1992, because they didn't want to alienate users with older machines. EGA was introduced in 1984 with the PC-AT, 8 years earlier.

Sure, upgrading was as important in 1985 but then Moore's Law kicked in and kicked C='s behind. Being able to upgrade is an advantage including for a console which we found out much later because C= went bankrupt before they could upgrade the CD32 as planned.

bhabbott Quote:

Some Amiga software doesn't support 68020, but it is mostly games that failed for other reasons too, such as expecting ChipRAM in certain places, being incompatible with later OS's etc. The huge number of games we can now run under WHDload show that many of the issues were not that hard to fix. The real problem was lack of support from publishers, exacerbated by rampant piracy which gave them less incentive to provide it. Also many developers moved on and source code was lost.

Self modifying code was legal on the 68000 and many programmers had no familiarity with a 68020 to know better. The "move.w SR,EA" instruction that became supervisor only on the 68020 is also a problem but that is partially Motorola's fault. I agree that the majority of upgrade problems were not due to the CPU. Many problems were due to poor programming but upgraded 68020+Ranger hardware would have been useful for testing.

bhabbott Quote:

The sad fact of the matter is that yes, compatibility with existing software was vitally import - even for software that broke the rules. It was the same on the PC. Right from the start, a PC clone's success was dependent on it being '100% IBM compatible'. Which of course was also the reason the Amiga was bound to fail - since it wasn't and couldn't be.

The PC world requires way more kludges than the Amiga. Motorola and Jay Miner deserve credit for creating and choosing the 32 bit ISA 68k for the Amiga. In a sad twist of fate, the ugly and kludgy PC lives on while the beautiful and good 68k is dead. All because IBM chose the vastly inferior 8088 instead of the futuristic 68000 or we would be using the 68k today.

@matthey

I don't disagree with you that the availability of the Ranger Chipset would've been better than it not being available. I totally get Jay Miner's desire to create a less gaming focused SKU, following the release of the A1000 and seeing the things that needed correcting.

Correcting things quickly is the agile engineering thing to do, but finding a market for what has already been released is the business thing to do. But in the grand scheme of things C= could've done better/sooner, Ranger for me scores lower than several others.

If I had to rank the C= faux pas, from top to bottom for me it would be:

1. CD32 (better)
2. AGA (sooner)
3. ECS (better and sooner)
4. CDTV (better)
5. Ranger (actually released)

I suppose an argument could be made that if Ranger was made available in 1987, then ECS would not have been a thing, or maybe given the production schedule some hybrid of the goals of Ranger and ECS ended up being available in 1988.

Obviously the further back in the Amiga timeline that things would've been done differently, the more the timeline of subsequent things would've shifted more dramatically. To the point where certain products as we know them may never have been released.

Would we then have gotten the AGA on it's original planned timeline, or would they go in a different direction? Would we have only got a minor update in a Ranger+ in the early '90s? And what would that mean for when Commodore would implode?

Because without a major win, like the C64 sales numbers, or major cash injection like the one from Microsoft to save Apple Computers, the company would've ended sooner or later.

_________________
All the way, with 68k

@agami

Quote:

Obviously the further back in the Amiga timeline that things would've been done differently

you know if multiverse exists, they can have saved lot of money, buy using same people in different timelines.

_________________
http://lifeofliveforit.blogspot.no/
Facebook::LiveForIt Software for AmigaOS

agami Quote:

I don't disagree with you that the availability of the Ranger Chipset would've been better than it not being available. I totally get Jay Miner's desire to create a less gaming focused SKU, following the release of the A1000 and seeing the things that needed correcting.

I don't really think of the Ranger chipset as being less gaming focused although it was in the C= mindset of gaming using the cheapest possible mass produced C64 like hardware. The gaming reality is more nuanced and value plays a big part. More expensive and higher performance x86 PC hardware became a gaming platform while the Amiga upgrade path fell behind. A 68020 and Ranger chipset would have paved the way up with it being high end at first but then could become the base Amiga standard later when it was cheap enough. The C= strategy was to freeze the Amiga technology until it became cheap enough for mass production. This did work to some extent with the Amiga 500 but it was late not being introduced at the beginning of the 68000+OCS technology curve. The followup Amiga 500+, Amiga 300/600 and CDTV with 68000+ECS were not enough of an upgrade and C= had failed to at least integrate the chipset and convert it to CMOS so it could be cost reduced more. Customers wanted the Amiga 1200 and CD32 with AGA but it was delayed, rushed late and the complete chipset had not been upgraded to run at 28MHz with a 68030@28MHz. Jay Miner understood the importance of integration, keeping technology ahead of the curve and value which C= upper management did not. The 1992 AUI Jay Miner interview highlights Ranger advantages, exudes his vision and shows his subdued frustration with C=.

http://www.bambi-amiga.co.uk/amigahistory/jayminerinterview.html

agami Quote:

Correcting things quickly is the agile engineering thing to do, but finding a market for what has already been released is the business thing to do. But in the grand scheme of things C= could've done better/sooner, Ranger for me scores lower than several others. If I had to rank the C= faux pas, from top to bottom for me it would be: 1. CD32 (better) 2. AGA (sooner) 3. ECS (better and sooner) 4. CDTV (better) 5. Ranger (actually released) I suppose an argument could be made that if Ranger was made available in 1987, then ECS would not have been a thing, or maybe given the production schedule some hybrid of the goals of Ranger and ECS ended up being available in 1988. Obviously the further back in the Amiga timeline that things would've been done differently, the more the timeline of subsequent things would've shifted more dramatically. To the point where certain products as we know them may never have been released. Would we then have gotten the AGA on it's original planned timeline, or would they go in a different direction? Would we have only got a minor update in a Ranger+ in the early '90s? And what would that mean for when Commodore would implode? Because without a major win, like the C64 sales numbers, or major cash injection like the one from Microsoft to save Apple Computers, the company would've ended sooner or later.

I consider Ranger to be a minor upgrade like ECS but better and sooner. It would have likely been out in 1987 with the Amiga 2000 while the Amiga 500 continued to be OCS. Both the Ranger and ECS chipsets allowed up to 2MiB of chip memory while the Amiga 2000 was stuck at 1MiB, had only 64 colors max with EHB mode in low resolution and had only the original chip memory bandwidth.

1987 68020+Ranger (Amiga 2000 only until VRAM prices dropped)
2MiB chip VRAM, x2 chip mem bandwidth, 99% CPU performance without fast mem, 7MHz clock
128 colors max (256 cheaply if they added EHB mode and another bit plane)
320x256x8 (LORES 256 colors), 640x256x8 (HIRES 256 colors), 1280x256x4 (SUPER-HIRES 16 colors)
640x480x4 (SVGA 16 colors), 1024x1024x4 (16 colors)

1990 68000+ECS (all Amigas)
2MiB chip DRAM, 7MHz chipset clock
64 colors max (with EHB mode)
320x256x6 (LORES 64 colors), 640x256x4 (HIRES 16 colors), 1280x256x2 (SUPER-HIRES 4 colors)
640x480x2 (SVGA 4 colors), 1024x1024x2 (4 colors)

1992 68020+AGA (all Amigas)
2MiB chip DRAM, x4 chip mem bandwidth, 14MHz chipset clock
256 colors max (plus HAM8)
320x256x8 (LORES 256 colors), 640x256x8 (HIRES 256 colors), 1280x256x5 (SUPER-HIRES 32 colors)
640x480x5 (SVGA 32 colors), 1024x1024x5 (32 colors)

I'm assuming a 7MHz clock for the Ranger chipset but the Agnus and Denise would have been updated and should have been converted to CMOS and moved to newer chip processes that allowed 14MHz clocking. Paula didn't change for AGA yet it was able to be clocked at 14MHz, perhaps with revisions to allow higher clock speeds, but this could have been done for the Ranger chipset as well allowing for AGA x4 chip mem bandwidth. C= documentation suggests that AGA should have allowed a 28MHz chipset clock speed and 68030@28MHz giving x8 chip mem bandwidth and AA+ was planning to allow 57MHz with x16 chip mem bandwidth and 68030@57MHz (likely as a single chip SoC Amiga only). All The custom chips required moving to a newer CMOS process than the mid 1970s 5000nm NMOS process though. It makes sense to integrate the chips together once as it saves development cost and a single custom chip SoC was likely possible when AGA came out but C= didn't want to invest in properly upgrading all the custom chips and the more expensive chipset process even though it was a cost reduction in the long term. Dave Needle and RJ Mical did similar integration in about 2 years for the 3DO custom chips. There were some important upgrades in ECS but some of them would have been included in the Ranger chipset while others may have come later with AA(+). The 68000+ECS standard in 1990 with only the original chip memory bandwidth is a travesty and C= upper management was brain damaged to think this was good enough for new hardware like the Amiga 500+, Amiga 300/600 and CDTV. AGA was late and behind the tech curve when introduced while 28MHz chipset and 68030@28MHz would have retained more value but AA+ should have been available instead of AGA. With Jay Miner in charge, the Amiga gets Ranger in 1987 and perhaps AA+ in 1990 instead of ECS. C= botched the Amiga and Jay Miner told them so.

Jeri Ellsworth posted on April 11, 2023 a picture of her cabinet and posted the following, "My cabinet of failed products just got two new denizens."

https://twitter.com/jeriellsworth/status/1646002897166151680

There are several consoles on the shelves that we have talked about in this thread. I wonder which console has had the most game releases since it failed. Perhaps the same console has more game releases today than AmigaNOne hardware?

More than 10% of her Twitter posts are C= related.

https://twitter.com/jeriellsworth

She has an old picture she found where she is showing Bil Herd the C64 DTV toy she created. So many hardware innovators are still around but not around here where the Amiga is AmigaNOne and hardware cost and value don't matter. It's too bad the 68k Amiga can't have low cost hobby and retro toys like THEA500 Mini and Raspberry Pi.

Last edited by matthey on 04-Oct-2023 at 11:16 PM.

@matthey

Quote:

matthey wrote: 1987 68020+Ranger (Amiga 2000 only until VRAM prices dropped) 2MiB chip VRAM, x2 chip mem bandwidth, 99% CPU performance without fast mem, 7MHz clock 128 colors max (256 cheaply if they added EHB mode and another bit plane) 320x256x8 (LORES 256 colors), 640x256x8 (HIRES 256 colors), 1280x256x4 (SUPER-HIRES 16 colors) 640x480x4 (SVGA 16 colors), 1024x1024x4 (16 colors) 1990 68000+ECS (all Amigas) 2MiB chip DRAM, 7MHz chipset clock 64 colors max (with EHB mode) 320x256x6 (LORES 64 colors), 640x256x4 (HIRES 16 colors), 1280x256x2 (SUPER-HIRES 4 colors) 640x480x2 (SVGA 4 colors), 1024x1024x2 (4 colors) 1992 68020+AGA (all Amigas) 2MiB chip DRAM, x4 chip mem bandwidth, 14MHz chipset clock 256 colors max (plus HAM8) 320x256x8 (LORES 256 colors), 640x256x8 (HIRES 256 colors), 1280x256x5 (SUPER-HIRES 32 colors) 640x480x5 (SVGA 32 colors), 1024x1024x5 (32 colors)

So I'm assuming that in 1987 you'd still have an A500 (as we had it) accompanying the Amiga Ranger big box machine?
Following it up with an A600 (as we had it) in 1990?
If at that time they could've had it at $599 then that would've been a seller. It'd be beating the C64 launch price (when adjusted for inflation). There might've even been more sales of the A600 in 1990/91 than A500s in 1987/88. Creating a new gaming baseline, which then would've seen more A500 owners upgrading.

An A300 at $299 would've matched the 1983 C64 $250 price, but from 1990 and onward it was no longer a low-price market. That price bracket was the domain of gaming consoles, but also people wanted to do more with their personal/home computers, so $499+ is where consumers felt they're getting a device that can help with work, and also entertain.

In the above release schedule, do you have an AGA A3000 coming out in 1992, and therefore the very expandable Amiga Ranger covering from 1987 to 1992?
Skip the slim desktop episode and make sure the AGA A3000 can accommodate Video Toaster cards?

In this schedule, Amiga is better challenging Apple: Having a Desktop that is comparable to the Mac II, and then an A3000 better challenging the IIvi/vx/Performa 600.
And eating some of Quadra's lunch if/when equipped with an 040.

_________________
All the way, with 68k