@bhabbott

Quote:

True. EGA was introduced with the PC-AT in 1984, but was still in common use in 1991, 7 years later. Not that different from the Amiga, where OCS was the 'standard' for 7 years

VGA crossed more than 50 percent of shipments in 1989.

According to Dataquest November 1989, VGA crossed more than 50 percent market share in 1989 i.e. 56%.
http://bitsavers.trailing-edge.com/components/dataquest/0005190_PC_Graphics_Chip_Sets--Product_Analysis_1989.pdf

Low-End PC Graphics Market Share by Standard Type
Estimated Worldwide History and Forecast


Total low-end PC graphic chipset shipment history and forecast
1987 = 9.2. million, VGA 16.4% market share i.e. 1.5088 million VGA.
1988 = 11.1 million, VGA 34.2% i.e. 3.79 million VGA.
1989 = 13.7 million, VGA 54.6% i.e. 7.67 million VGA.
1990 = 14.3 million, VGA 66.4% i.e. 9.50 million VGA.
1991 = 15.8 million, VGA 76.6% i.e. 12.10 million VGA.
1992 = 16.4 million, VGA 84.2% i.e. 13.81 million VGA.
1993 = 18.3 million, VGA 92.4% i.e. 16.9 million VGA.

In absolute numbers, VGA's 1.5 million sales in 1987 had no problems beating A500/A2000 yearly shipments. Your "most" argument hides VGA's absolute numbers.

A desktop PC clone can upgrade to a 256-color VGA/ SVGA clone at any time without replacing the entire machine. A500 and "big box" A1500/A2000/A2500/A3000 are dead end for 256-color AGA games.

The "big box" A2000 was supposed to be the equivalent to 16bit 286 with 386SX's 32bit programming model-based PC. The "big box" A2500/020 (14.3 Mhz) and A2500/030 (25 Mhz) are the 386DX-16 to 386DX-25 PC equivalents.

From https://landley.net/history/mirror/commodore/haynie.html on Dave Haynie's push for partitioned Amiga graphics i.e. Acutiator. Most of my argument is framed from Commodore engineers' statements.

"Only Amiga" caused a full 32bit 68020/68030 equipped desktop computer to be obsolete for 256 color gaming and hit the metal applications e.g. video apps with AGA support e.g. Scala MM400.

The other 68K desktop platforms like Atari ST also have dead-end graphics upgrade paths or a hard graphics capability separation between ST and TT.

This is why I still purchased the A1200 despite owning an A3000 with 25 MHz 68030. My real intention is AGA functions for A3000.

A pure RTG-equipped AmigaOS compatibles are sales failures i.e. PowerAmiga NG, Amithlon, DraCo and open-source AROS x86. The Amiga is not a Mac.

PC stakeholders figured out gaming PCs minimum with a single or tiny few fast slot configurations e.g. two VLB, single AGP(similar to PCI-X), and single PEG (PCIe X16) as the product separation from workstation PCs.

A500 and A1200 have single Zorro like expansion potential, but it wasn't common with big-box counterparts, hence Commodore couldn't expand the Zorro II/III market into millions install base. Economies of scale matter.




Quote:

While the vast majority of PCs could have a VGA card installed, it didn't make make much sense for XT class machines. I installed a VGA card into my Amstrad PC-20, but that didn't magically make it suitable for running VGA games. I now have an Amstrad PC2086 (8 MHz 8086) with built in VGA and it isn't much better - won't even play Wolfenstein 3D properly.

Are you claiming Amiga's 4 to 5 million install base beats 386DX or 486 unit sales?

From https://www.intel.fr/content/dam/doc/report/history-1994-annual-report.pdf
Intel reported the following
1. In 1994's fourth quarter, Pentium unit sales accounted for 23 percent of Intel's desktop processor volume.
2. Millions of Pentiums were shipped.
3. During Q4 1993 and 1994, a typical PC purchase was a computer featuring the Intel 486 chip.
4. Net 1994 revenue reached $11.5 billion.
5. Net 1993 revenue reached $8.7 billion.
6. Growing demand and production for Intel 486 resulted in a sharp decline in sales for Intel 386 from 1992 to 1993.
7. Sales of the Intel 486 family comprised the majority of Intel's revenue during 1992, 1993, and 1994.
8. Intel reached its 6 to 7 million Pentiums shipped goal during 1994. This is only 23 percent unit volume.

Your argument is absurd when the PC platform has a large customers base who can spend on the latest PC components.

Hint: https://www.pcgamer.com/steam-suggests-there-are-more-rtx-3090-cards-online-than-the-entire-rx-6000-series-put-together/
NVIDIA's RTX 3090 sales exceed the entire RX 6000-series put together. This does not include lower cost RTX 3080 Ti, 3080, 3070, and 3060.

This is repeated for the RTX 4000 series. NVIDIA is powerful for a reason.

Your argument is flawed. Your argument is framed from a weak currency and a tiny country like New Zealand. My argument is framed from the superpower United States of America.

Quote:

The A1200 wasn't intended to 'sync with' the 3DO, PS1 or Saturn. None of them existed when the A1200 was released. The 3D0 was out by 1993 but was hardly on the radar, and the CD32 was designed to be a CD-ROM console version of the A1200, not a radically new machine.

Your argument is framed from a retail POV, not from the 3rd party developer's POV.

3DO was released in Q4 1993, but dev kits were released earlier. Hint: Alone in the Dark (1992).

You didn't factor in 3rd party game developers meeting with Ali during CD32's development in 1992 and early 1993. Psygnosis is giving hints of Sony's Playstation to Ali.

Sony's PlayStation developer engagement is active in late 1992 and 1993.

Quote:

Sez you.

Who co-designed SuperFX for SNES? The game developer is from Amiga's core markets i.e. UK.
Who reverse-engineered Doom for SNES's port?
What are their development workstations? Hint: it was an Amiga!

The difference is Nintendo listened.

You didn't factor in 3rd party game developers meeting with Ali during CD32's development in 1992. Psygnosis is giving hints of Sony's Playstation to Ali.

Quote:

I say DSP would have unnecessarily raised the price and taken years to make full use of, and still not match the PS1. I don't remember any Amiga fans talking about DSP at the time, and it wasn't missed. It didn't do much for the Atari Falcon either.

Who co-designed SuperFX for SNES? The game developer is from the UK.
Who reverse-engineered Doom for SNES's port?
What are their development workstations? Hint: it was an Amiga!

The difference is Nintendo listened.

You didn't factor in 3rd party game developers meeting with Ali during CD32's development in 1992. Psygnosis is giving hints of Sony's Playstation to Ali.

Quote:

Some RAM boards did map into that area to get up to 5.5MB with PCMCIA. But if you are talking about putting 1MB on the motherboard, this was a non-starter due to the cost. The A1200 was a low-end machine like the A500, designed to keep the cost of entry as low as possible while still being expandable to mid or high end performance.

A1200's 140 ns read/write cycle 2 MB Chip RAM's BOM cost is about $52.

There was a "healthy profit margin" for A1200's 399 UKP/$599 USD and Ali pushed for lower profit margin CD32 with 299 UKP/$399 USD.

Lew's plan for AA+ A1200 replacement has DSP to boost its math power. You're wrong.

Last edited by Hammer on 07-Jul-2024 at 02:12 AM.
Last edited by Hammer on 07-Jul-2024 at 02:05 AM.
Last edited by Hammer on 07-Jul-2024 at 01:57 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 7950X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB

@Karlos

Quote:

Karlos wrote: @bhabbott Cue wall of spam from dataquest archive.

bhabbott is in dreamland.

_________________
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 7950X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB

@Karlos

Unlike A1200's CPU, Falcon's 030 CPU couldn't be recycled for the full 32-bit memory expansion since Falcon's 030 CPU is wired for a 16-bit data bus.

Falcon's 32 Mhz DSP56 has 24-bit data bus connection to the 32-bit memory bus.

Atari had a Falcon prototype with a 32-bit data bus 030.

Atari had plans for ST and TT segmentation for the Falcons.

Last edited by Hammer on 07-Jul-2024 at 02:23 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 7950X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB

@Hammer

Quote:

Hammer wrote: VGA crossed more than 50 percent of shipments in 1989.

But not the number of machines in use. Which is why games continued to include EGA graphics for several more years.

Quote:

In absolute numbers, VGA's 1.5 million sales in 1987 had no problems beating A500/A2000 yearly shipments. Your "most" argument hides VGA's absolute numbers.

In absolute numbers, PCs beat everything. But most of them were used in businesses, not homes.

Quote:

A desktop PC clone can upgrade to a 256-color VGA/ SVGA clone at any time without replacing the entire machine. A500 and "big box" A1500/A2000/A2500/A3000 are dead end for 256-color AGA games.

After AGA came out, some 'big box' Amiga owners added a graphics card, and in later years some games supported them. But most people didn't because the Amiga already had excellent graphics. OCS/ECS games were still coming out in 1994, and some were so good you would swear they were AGA!

The videos below show some OCS/ECS games alongside their AGA versions. In many cases it's hard to tell the difference, and all the OCS/ECS versions look pretty good! IMO some of the AGA versions are actually too busy, with added parallax backgrounds that are just a distraction.

Amiga games : AGA vs ECS OCS Part 1

Amiga games : AGA vs ECS OCS Part 2

There are plenty of other excellent Amiga games that didn't get an AGA version because they didn't need it. With up to 64 colors bitmap colors and many more with copper effects, 256 colors just wasn't needed for anything but PC ports.

In the PC world it was a different story. EGA was pretty sucky, and CGA was awful. So VGA was a huge step up for them. Nevertheless some games actually looked better in EGA than VGA, eg. those Sierra adventure games that were digitized (poorly) from images drawn on paper. The older EGA versions were drawn by computer artists who knew how to get the best out of the format.

Quote:

The "big box" A2000 was supposed to be the equivalent to 16bit 286 with 386SX's 32bit programming model-based PC. The "big box" A2500/020 (14.3 Mhz) and A2500/030 (25 Mhz) are the 386DX-16 to 386DX-25 PC equivalents.

The A2000 wasn't supposed to be the 'equivalent' of any PC, it was an Amiga. What PC had genlocking, and 4 channel PCM sound, and draggable screens, and an efficient preemptive multitasking GUI OS, and a mouse that worked properly? What 386 PC could have 112MB of RAM, and have the CPU upgraded to Pentium class or beyond?

Quote:

From https://landley.net/history/mirror/commodore/haynie.html on Dave Haynie's push for partitioned Amiga graphics i.e. Acutiator. Most of my argument is framed from Commodore engineers' statements.

Commodore engineers had some funny ideas.

Quote:

This is why I still purchased the A1200 despite owning an A3000 with 25 MHz 68030. My real intention is AGA functions for A3000.

I did the same. In my case it was specifically so I could develop AGA titles for the CD32.

Quote:

A pure RTG-equipped AmigaOS compatibles are sales failures i.e. PowerAmiga NG, Amithlon, DraCo and open-source AROS x86. The Amiga is not a Mac.

A friend of mine had a Draco. He purchased and used it for one thing - making commercial videos (or video commercials :) ). It did it very well.

As for the rest, they all appeared after Commodore was gone and the Amiga was really a retro computer. People created that stuff because they wanted it. If others like me didn't they didn't care. This is a similar attitude to what Commodore's engineers had.

Quote:

A500 and A1200 have single Zorro like expansion potential, but it wasn't common with big-box counterparts, hence Commodore couldn't expand the Zorro II/III market into millions install base. Economies of scale matter.

What matters to me is that Commodore produced the A1200, and the CD32, CDTV, A500, A500+, A600, A1000, A2000, A3000 and A4000 - all great machines. I wish I could own them all but I won't because there isn't enough time left in my life to appreciate them. I also have a C64 and Plus 4 and they aren't getting any love at all. Shall I tell you about all the other retro computers I have that are sitting around doing nothing? My point is, Commodore didn't need to make millions of Zorro-III machines. We got way more than we need!

@Hammer

Quote:

Hammer wrote: bhabbott is in dreamland.

Dreaming of the Amstrad CPC664 selling on TradeMe right now, that I won't buy because I wouldn't use it. Nostalgia can be a bummer, I really regret throwing mine away back in 2015 just because the keyboard membrane was faulty (then I found out you can get replacement membranes!).

@Hypex

Quote:

Hypex wrote: @bhabbott Quote: I ran the Doom timedemo on all standard settings with no sound and the result was... 1.34 fps! That's shocking. I expected better from 386 compared to an A1200. An A1200 at 14Mhz and 16MB I expect would get that FPS if you are lucky. With a 68030, which compares better with a 386 (both 32 bit) an Amiga can perform better, but it usually is clocked faster. Did you shrink the screen? If your 386 felt better from it. I don't think John Carmack was thinking of the 386 people, did he think of what impact Doom would have on the majority of 386DX with TVGA8900C VGA card users? I don't think so!

386SX is only 16-bit bus and the CPU doesn't have any internal cache (and most 386SX motherboards didn't have external cache either). Still the frame rate is lower than I expected.

If you shrink the window to postage stamp size it becomes almost playable.

John Carmack designed Doom for 486, with 66MHz recommended. He wasn't thinking about 386SX at all - the days of coding for the low end were over!

@bhabbott

He didn't design it for 486 at all. He designed it for any sensible 32-bit CPU and he built it all on a 68040 NeXT box. All the optimisations at this stage were algorithm level. No great surprise that it runs on just about everything these days.

Porting it to DOS for the release was a separate stage of development and focused more on low level optimisations. My first experience of it was on a 486SX 25MHz and it was perfectly playable at fullscreen, with the occasional bit of lag.

_________________
Doing stupid things for fun...

@Karlos

Quote:

Karlos wrote: @bhabbott He didn't design it for 486 at all. He designed it for any sensible 32-bit CPU and he built it all on a 68040 NeXT box. All the optimisations at this stage were algorithm level. No great surprise that it runs on just about everything these days. Porting it to DOS for the release was a separate stage of development and focused more on low level optimisations. My first experience of it was on a 486SX 25MHz and it was perfectly playable at fullscreen, with the occasional bit of lag.

The efforts Amiga fans will go to to 'prove' that black is white are amazing. The NeXTstation had a 'sensible 32 bit CPU' but only a 2 bit greyscale display. This was the host machine for cross-development, not the target.

Working this way was common in those days - you might use a workstation or PC to compile the code, an Amiga to do the graphics, and the target could be anything from an 8 bit home computer to a games console (often hooked up through a high speed parallel interface or ROM emulator to quickly download the code for testing).

Yes, Carmack developed Doom on a 33MHz 040 NeXTstation, but it was always intended to target the PC. Performance on the NeXTstation was poor but that didn't matter. He optimized the 'low level' code for the 486 and then marketed it for the 486. He knew it would be slow on a 386DX and appallingly slow on a 386SX, but didn't care because he wasn't aiming at that market. That's why he told George Sanderson that on the Amiga Doom would need 'the full speed of a 68040 to play the game properly even if you have a chunky pixel mode".

But why would he design a game that only runs properly on the latest most powerful PCs? Before then he was developing games under contract for SoftDisk, but they insisted that all their games run in CGA because many of their customers had older less capable machines. So when he developed a smooth scrolling routine that only worked in EGA they rejected it. At that point Carmack and Romero decided to develop their own games, and id Software was born. Thus started the trend of coding for machines powerful enough to do what you want and to hell with the 'base'.

Doom 'runs' on just about anything these days? That's a lie. Many of those so-called Doom ports take huge liberties to the point where there is little if anything left of the original code and data. Unlike the original, the game is shoehorned into the hardware no matter how much it must be compromised, rather than designing it for hardware that is powerful enough to achieve the desired results.

@bhabbott

Carmack used a NextStation Color (sic). Do your homework. The exact model has been demoed running the original builds on YouTube even.

Last edited by Karlos on 08-Jul-2024 at 08:57 PM.

_________________
Doing stupid things for fun...

@bhabbott

There are many, many ports of the full game to 32 (and 64 bit) platforms. Just because pregnancy-test doom may be some comically cut down version does not detract one bit (no pun intended) from the fact that Doom was designed, first and foremost, for 32-bit processors in general. From the strict use of 16:16 fixed point arithmetic to the lack of endian specific tricks and hacks in the *vanilla* targets. Platform specific code is abstracted out cleanly into replaceable modules that can be as coupled to their respective targets as you want.

Carmack wasn't interested in that specific type of optimisation. On his own words, he has functional low level optimisation skills but he's always been much more focused on the algorithm side. This is why Abrash was brought in to optimise Quake, for example, doing all the clever overlapping integer and floating point work.

But anyway, I don't know why we fixate so much on Doom. It's a great gane franchise but I'm not remotely butthurt that it was never released officially for the Amiga back in the day. Could the Amiga have ran a properly optimised port in the early days? Sure. That's been proven pretty much. You could've had a playable full version of the game on an A1200 with at least a ram expansion, if you took some 2x2 pixelmode approach. In 1993 only a minority of users would have had the hardware necessary to do any kind of fullscreen port any justice whatsoever.

_________________
Doing stupid things for fun...

@Karlos, @bhabbott

Whether or not Carmack specifically targeted the 486 at 66MHz or come the end of 1993 found that spec to be the sweet spot in the "Goldilocks Zone" of the DOS PC landscape, is neither here or there because the game was adjustable to play within or, as you point out, even outside (reduced experience) of the said Goldilocks Zone.

As it happens, I first played it on a friend's 486DX2 66 who bought this first PC after selling his Amiga setup. The game was interesting, but I was obviously not within its target demographic. Give me a good strategy game, a puzzler, or even an adventure game. I liked playing things at a more thoughtful pace.

It wasn't until some time after October 1995, after I bought my first PC (Win 95 P90), that I had a modem-linked "multi-player" session of Doom II with that same friend. The game still didn't pull me in for its FPS game play, but rather for the unpredictable game dynamic of having another person in the game shooting back at you.

What finally got me into FPS games was GLQuake running on a 3Dfx Voodoo 1 card. The fidelity jump was remarkable, and it turns out that's what I needed.

_________________
All the way, with 68k

@bhabbott

Quote:

bhabbott wrote: That's shocking. I expected better from 386 compared to an A1200. An A1200 at 14Mhz and 16MB I expect would get that FPS if you are lucky. With a 68030, which compares better with a 386 (both 32 bit) an Amiga can perform better, but it usually is clocked faster.

386 has 32 entry TLB cache. Socketed 386 has the option for faster 486 clones.

Am386DX-40 arrived in 1991 during many SVGA clone chipsets.

Until Q4 1992 in very limited AGA unit numbers, Amiga's 68030 offerings weren't coupled with arbitrary 256-color display capability.

OCS/ECS Amigas with "full 32-bit" 68020 and 68030 CPUs didn't participate in Amiga's "full 32-bit" 68K with AGA gaming. Amiga AGA started from ground zero at Q4 1992.

A1200 wasn't in Australia's Xmas Q4 1992 market until sometime in 1993.

Many A500 owners pulled the PC purchases in Xmas Q4 1992 when the A1200 was largely MIA (missing in action)!!

44,000 units for the UK's Xmas Q4 1992 sale period is LOL while gaming PCs and SNES were shipping in large economies of scale.

Commodore's "more than six months" wasted on ECS A1000Jr and ECS A600 comes back to bite Commodore in the ass. The market knows the ECS is a dead end for 256-color gaming.

Commodore failed by their own actions.

Quote:

Did you shrink the screen? If your 386 felt better from it. I don't think John Carmack was thinking of the 386 people, did he think of what impact Doom would have on the majority of 386DX with TVGA8900C VGA card users?

Doom has a "low detail" mode for fast 386DX PCs.

https://youtu.be/oJgQJaqmPx4?t=795

Doom (low details) on 386DX-40 with 128K cache
Tseng ET4000 ISA = 26.751 fps
Trident 8900CL ISA = 23.0088 fps
WD90C32 = 26.838 fps via Diamond Speedstar 24X.

I jumpered my i386DX-33 into 40 Mhz overclock via the Am386DX-40 jumpered settings.

i386DX-33 uses the same process node as i486DX-50.

ISA cards in the early 1990s were designed to handle 10 Mhz MCA, hence 10 Mhz ISA overclock from AT standard 8.3 Mhz is not a major problem.


https://dosdays.co.uk/topics/Manufacturers/tseng_labs.php
By 1991, according to IDC, Tseng Labs held a 25% market share in the total VGA market.

In absolute numbers, Tseng Labs has no problems shipping SVGA units that exceed Amiga's best 1 million annual sales.

Hint: Tseng Labs ET4000 sales murdered Commodore's AGA unit sales!

The USA, Canada, and Australia have stronger dollar currencies and larger markets when compared to tiny New Zealand. New Zealand should have joined the Australian Federation.

European Union matters for large economies of scale and strong Euro currency.

Last edited by Hammer on 10-Jul-2024 at 02:37 AM.
Last edited by Hammer on 10-Jul-2024 at 02:33 AM.
Last edited by Hammer on 10-Jul-2024 at 02:20 AM.
Last edited by Hammer on 10-Jul-2024 at 02:19 AM.
Last edited by Hammer on 10-Jul-2024 at 02:16 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 7950X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB

@bhabbott

Quote:

But not the number of machines in use. Which is why games continued to include EGA graphics for several more years.

It wouldn't matter when the gaming PC minority is still larger than the Amiga's entire install base.

When you have a Macintosh customer base who can spend 1.2 million PowerMacs units during 1994 and Amiga's best effort is 1 million units mostly from A500 models, the user base demographics between Mac and Amiga are very different.

The PC market is even larger than "Mr 10 percent" of Mac sales.

The Amiga is like bringing a peashooter to a gunfight.

Intel can sell Pentiums in the 1993-1994 time period with 6 to 7 million units magnitude. The main reason, Intel's X86 home market is attached to the world's economic superpower i.e. the United States of America.

Your market intelligence's argument position is seriously flawed.

Quote:

After AGA came out, some 'big box' Amiga owners added a graphics card, and in later years some games supported them. But most people didn't because the Amiga already had excellent graphics

"Big box" Amigas are only numbered in a few thousand, NOT in the mentioned 6 to 7 million Pentium 60/66.

Quote:

. OCS/ECS games were still coming out in 1994, and some were so good you would swear they were AGA!

That's subjective and your bar is low. For strong 2D example is SNES's examples.

SNES has a 60 million install base.

Quote:

The videos below show some OCS/ECS games alongside their AGA versions. In many cases it's hard to tell the difference, and all the OCS/ECS versions look pretty good! IMO some of the AGA versions are actually too busy, with added parallax backgrounds that are just a distraction.

SNES has a 60 million install base.

Quote:

Amiga games : AGA vs ECS OCS Part 1 Amiga games : AGA vs ECS OCS Part 2

That's a flawed argument.

AGA has games or tech demos such as Lion King or Reshoot R or Reshoot Proxima 3 or the arcade-quality Final Fight AGA tech demos.

Your YouTube examples didn't show Super Star Dust AGA's tunnel scenes.

Certain game designs that target AGA will not be cross-compatible with OCS.

1. Bitmap Bros' baseline artwork and game design is based on Atari ST. This is a lower-bar game design when compared to something like Amiga OCS exclusive Elf Maina, but this graphics quality is not consistent because Commodore SDK is not action game-centric.

Super Cars 2's game design is old 16-bit. The 32-bit gaming PC has texture-mapped 3D IndyCar Racing in 1993 and Amiga AGA with 68030 minimum CPU equivalent is Virtual GP in 1999.

2. AGA's dual playfield is 15+16 colors which is a small step up from OCS's 7+8 colors.

The programmer needs to use the ganged 64-bit wide sprites to preserve a single playfield's higher color count mode method. This method is not cross-compatible with Amiga OCS's 16-bit wide sprites.

For 2D, Gaming PC VGA has games like Tyrian, arcade-quality Mortal Kombat 1, arcade-quality Mortal Kombat 2 and 'etc'. 32-bit gaming PC gaming experience is in texture-mapped 3D and Doom is not the only title.

A low-effort AGA version for Mortal Kombat 2 would have a higher color version, and 64-bit wide sprites would have handled foreground parallax objects. AGA has the potential and OCS acted like the "Atari ST".

Quote:

There are plenty of other excellent Amiga games that didn't get an AGA version because they didn't need it. With up to 64 colors bitmap colors and many more with copper effects, 256 colors just wasn't needed for anything but PC ports. In the PC world it was a different story. EGA was pretty sucky, and CGA was awful. So VGA was a huge step up for them. Nevertheless some games actually looked better in EGA than VGA, eg. those Sierra adventure games that were digitized (poorly) from images drawn on paper. The older EGA versions were drawn by computer artists who knew how to get the best out of the format.

That's outside of this topic's CD32's time scale.

Last edited by Hammer on 10-Jul-2024 at 03:39 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 7950X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB

@Karlos

Quote:

Karlos wrote: @bhabbott Carmack used a NextStation Color (sic). Do your homework. The exact model has been demoed running the original builds on YouTube even.

Thanks for that. You seem to be right about him having a color NeXTstation. However according to this video The Tools that Built Doom he used the NeXTstation to cross-develop with the target being a PC. In his own words:-
Quote:

"I'm in Next step and I create the level real quick and I bsp it, so I just go over to my DOS machine, I copy the level from the Z drive where the map was, and I just copy into the Doom directory and I just run Doom."

The video also shows Doom running in a window on a color NeXTstation, and it's very slow (perhaps 2-3 fps).

According to Wikipedia,
Quote:

id Software was using a NeXTcube for its graphic-engine development. This version is sluggish on anything below an 040 NeXTstation/cube (though it runs smoother with a higher amount of memory), and is missing sound, which was added on the PC side. With NeXT-Step based on i486 architecture, it ran smoothly...

Carmack did later port Doom to the Sega 32x and Atari Jaguar, but had to seriously cut down the game to get it working on them. About the Jaguar, he said:-
Quote:

"The Jaguar was definitely significantly hampered by its technical flaws, which kept me from ever being too big of a Jaguar booster... "The little risc engined were decent processors. I was surprised that they didn't use off the shelf designs, but they basically worked ok. They had some design hazards (write after write) that didn't get fixed, but the only thing truly wrong with them was that they had scratchpad memory instead of caches, and couldn't execute code from main memory. I had to chunk the DOOM renderer into nine sequentially loaded overlays to get it working..."

Let's face it, the only viable platform to release Doom on in 1993 was the PC, and it needed the power of a 486 to run at a reasonable speed. At that time id was not looking at releasing it on any other platform. They even refused to port it to Microsoft Windows.

@Hammer

Quote:

Hammer wrote: @bhabbott Quote: bhabbott wrote: That's shocking. I expected better from 386 compared to an A1200. An A1200 at 14Mhz and 16MB I expect would get that FPS if you are lucky. With a 68030, which compares better with a 386 (both 32 bit) an Amiga can perform better, but it usually is clocked faster. 386 has 32 entry TLB cache. Socketed 386 has the option for faster 486 clones.

I didn't write that. The TLB cache is only used to avoid slowdowns on virtual memory paging. Doom doesn't use virtual memory so...

Quote:

Until Q4 1992 in very limited AGA unit numbers, Amiga's 68030 offerings weren't coupled with arbitrary 256-color display capability.

This thread is about Doom on the CD32. By the time it was released in late 1993, only AGA machines were being made.

Quote:

44,000 units for the UK's Xmas Q4 1992 sale period is LOL while gaming PCs and SNES were shipping in large economies of scale.

Again, totally irrelevant to this thread.

Quote:

Doom has a "low detail" mode for fast 386DX PCs.

But not 386SX.

Here's what is relevant. DoomAttack won't run on a stock CD32 because it needs 4 MB of FastRAM. When that is added, Doom runs much faster on the CD32 with Akiko C2p than it does on a 16 MHz 386SX. In fact as these benchmarks show, it is even faster than a 40 MHz 386SX, and only 26% slower than an i386DX-33 (5.8 fps vs 6.89 fps on standard settings).

Last edited by bhabbott on 10-Jul-2024 at 05:14 PM.

@bhabbott

Quote:

bhabbott wrote: This thread is about Doom on the CD32. By the time it was released in late 1993, only AGA machines were being made.

By the time CD32 could run Doom (read RAM expansion being available) C= was dead and the numbers sold made such a combo irrelevant anyways.

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

@bhabbott

Quote:

This thread is about Doom on the CD32. By the time it was released in late 1993, only AGA machines were being made.

CD32 and A1200 are shared platform with a single Akiko C2P difference.

Archos drive uses the PCMCIA slot, emulates Akiko through software. Only a small effort is required to enable A1200 to fake a CD32.

Wing Commmander CD32's Akiko C2P was patched to work for A1200.

Quote:

Again, totally irrelevant to this thread.

Again, CD32 and A1200 are shared platform with a single Akiko C2P difference.

Archos drive uses the PCMCIA slot, emulates Akiko through software. Only a small effort is required to enable A1200 to fake a CD32.

Akiko C2P design is less than a day job and should been included with A1200's Budgie.

Quote:

But not 386SX.

Refer to FastDoom. Like the Amiga, PC also has "optimized" Doom.

Budgie's and Akiko's Ramsey memory controller functions can support 32-bit Fast RAM.

There's potential, but it was half-baked.

Commodore rather spend on FMV VCD module instead of Amiga CPU accelerator bundled with game bundles.


CD32's FMV module has the following:

1. 24-bit DAC (STM's STV8438CV) for 16.7 million colors display.

2. MPEG-1 decoder from C-Cube CL450, 352 x 240 pixels @ 30hz, 352 x 288 pixels at 25 Hz, pixel interpolation and frame duplication to produce output formats of 704 x 240 pixels at 60 Hz or 704 x 288 pixels at 50 Hz.

https://websrv.cecs.uci.edu/~papers/mpr/MPR/ARTICLES/060803.PDF

CL450 has about 398K transistors with up to 40 MHz. CL450 includes a licensed MIPS-X RISC processor with semi-custom extensions. In quantities of 100K or more per year, the price is less than $50 in 1992.
CL450's MIPS-X RISC processor still has the usual RISC instruction set.

3. LSI l64111qc (Digital Audio Decoder, 16-bit DAC),

4. 512 KB local RAM, NEC 423260 DRAM 4Mbit (512 KB) with 80 ns.

5. Lattice ispLSI 1024-60LJ CPLD.

Commodore is willing to spend on this non-core business by following the failed CDI.

Commodore says NO to MIPS RISC processor @ 40 Mhz for Amiga's general-purpose games.

Commodore says NO to 512 KB Fast RAM for Amiga's general-purpose games.

Commodore says NO to 24-bit color 704 x 288p for Amiga's general-purpose games.

Commodore says NO to 16-bit stereo audio for Amiga's general-purpose games.

There's more than enough total BOM cost from FMV for Amiga's 68EC030 accelerator with Fast RAM.

Commodore is avoiding mid-range accelerated A1200/CD32 out-of-the-box configurations from competing against Commodore Germany's mid-price range PCs.

For my mid-range budget in 1993, Commodore is pushing me to purchase their DT486 class PCs instead of the Amiga.

Apple is 100% loyal for their Mac platform while Commodore is NOT 100% loyal for their Amiga platform. The rot is within Commodore.


Quote:

Here's what is relevant. DoomAttack won't run on a stock CD32 because it needs 4 MB of FastRAM.

I already know this. I have A1200 with selectable 4 MB, 5.5 MB and 8 MB Fast RAM card from https://www.amigakit.com/a1200-flash-fast-memory-expansion-40mhz-option-p-12702.html


Quote:

When that is added, Doom runs much faster on the CD32 with Akiko C2p than it does on a 16 MHz 386SX.

Reminder, 32-bit Fast RAM equipped CD32's 68EC020 @ 14Mhz is equivalent to a "full 32bit" 386DX-16.

Quote:

In fact as these benchmarks show, it is even faster than a 40 MHz 386SX, and only 26% slower than an i386DX-33 (5.8 fps vs 6.89 fps on standard settings).

You have forgotten my support for Commodore UK MD David Pleasance's and Psygnosis's minimal cost upgrade for CD32 vs Mehdi Ali's rejection.

https://www.youtube.com/watch?v=1B1jKjrRUmk
A little "side by side" video of Doom running on accelerated Amiga 1200 with a Terrible Fire 1230 (030 @ 50 Mhz) verse AMD 386DX 40 MHZ with a Tseng Labs ET4000 ISA video Card. Result: very similar performance


https://youtu.be/B4fzMaadcHk?t=488
CD32's Doom Akiko and SX32 Pro with 030 @ 50 Mhz. SX32 Pro was released in 1996 which is too late for 1993-to-1994 time frame.

PS1 and Pentium era in Xmas Q4 1995. I have a Pentium 166 MHz based PC in Q3 1996.

Last edited by Hammer on 11-Jul-2024 at 04:27 AM.
Last edited by Hammer on 11-Jul-2024 at 04:22 AM.
Last edited by Hammer on 11-Jul-2024 at 04:18 AM.
Last edited by Hammer on 11-Jul-2024 at 04:14 AM.
Last edited by Hammer on 11-Jul-2024 at 04:11 AM.
Last edited by Hammer on 11-Jul-2024 at 04:07 AM.
Last edited by Hammer on 11-Jul-2024 at 03:59 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 7950X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB

bhabbott Quote:

Thanks for that. You seem to be right about him having a color NeXTstation. However according to this video The Tools that Built Doom he used the NeXTstation to cross-develop with the target being a PC. In his own words:- Quote: "I'm in Next step and I create the level real quick and I bsp it, so I just go over to my DOS machine, I copy the level from the Z drive where the map was, and I just copy into the Doom directory and I just run Doom." The video also shows Doom running in a window on a color NeXTstation, and it's very slow (perhaps 2-3 fps).

The Mac and PC DOS were not as efficient for productivity because they lacked preemptive multitasking.

https://www.forbes.com/sites/quora/2016/09/01/why-john-carmack-chose-next-for-developing-doom-and-other-favorites/ Quote:

I have no regrets at all about developing Doom on a NeXT! I bought our first NeXT (a ColorStation) out of personal interest. Jason Blochowiak had talked to me about the advantages of Unix-based systems from his time in college and I was interested in seeing what Steve Jobs' next big thing was. It is funny to look back; I can remember honestly wondering what the advantages of a real multi-process development environment would be over the DOS and older Apple environments that we were using. Actually, using the NeXT was an eye-opener, and it was quickly clear to me that it had a lot of tangible advantages for us, so we moved everything but pixel art (which was still done in Deluxe Paint on DOS) over. Using Interface Builder for our game editors was a NeXT unique advantage, but most Unix systems would have provided similar general purpose software development advantages (the debugger wasn’t nearly as good as Turbo Debugger 386, though!) Kevin Cloud even did our game manuals, starting with Wolfenstein 3D, in Framemaker on a NeXT. This was all in the context of DOS or Windows 3.x; it was revolutionary to have a computer system that didn’t crash all the time. By the time Quake 2 came around, Windows NT was in a similar didn’t-crash-all-the-time state; it had hardware accelerated OpenGL, and Visual Studio was getting really good, so I didn’t feel too bad about moving over to it. At that transition point I did evaluate most of the other Unix workstations and didn’t find a strong enough reason not to go with Microsoft for our desktop systems. Over the entire course of Doom and Quake 1’s development we probably spent $100,000 on NeXT computers, which isn’t much at all in the larger scheme of development. We later spent more than that on Unix SMP server systems (first a quad Alpha, then an eventually 16-way SGI system) to run the time consuming lighting and visibility calculations for the Quake series. I remember one year looking at the Top 500 supercomputer list and thinking that if we had expanded our SGI to 32 processors, we would have just snuck in at the bottom.

Why not use a cheaper Amiga that had the best version of Deluxe Paint as well? The developers were Jobs/Mac fans and the Amiga was a threat.

http://rome.ro/2006/12/apple-next-merger-birthday.html Quote:

Apple-NeXT Merger Birthday! It's been 10 years since Apple merged with NeXT and then later brought us the awesome OS-X. I still remember seeing OS-X in action for the first time and how the Finder displayed directories in NeXTSTEP format and the rotating CD for the new Wait cursor.....YES! I was totally convinced at that point that some version of Mach was at the core of OS-X. Jobs had brought The Power back home. At that point NeXTSTEP had morphed into Cocoa - the development environment that Mac coders live in. Check out the celebration at NeXT Computers.org and browse the forums to find out more about this legendary company, their hardware and their unsurpassed software. Why do I care so much about NeXT computers? Because we at id Software developed the groundbreaking titles DOOM and Quake on the NeXTSTEP 3.3 OS running on a variety of hardware for about 4 years. I still remember the wonderful time I had coding DoomEd and QuakeEd in Objective-C; there was nothing like it before and there still is no environment quite like it even today. When id Software was stationed in Madison, Wisconsin during the winter of 1991, most of us were gone for the Christmas holiday - except John Carmack. John's present, which he bought with $11,000 of his own money, procured by walking through the snow and ice to remove from the bank, arrived during the holiday and he spent the whole time learning as much as he could about the computer and started working on vector quantization algorithms for compressing graphics. His test graphic was a 256-color screen from King's Quest 5. After his research was done it was agreed that the entire company needed to develop our next game on NeXTSTEP. id's first NeXT hardware was all black - both Cubes and Stations. We upgraded through the years to the Turbo model then to other hardware like the HP Gecko and then Intel hardware at the end. We were building fat binaries of the tools for all 3 processors in the office - one .app file that had code for all 3 processors in it and executed the right code depending on which machine you ran it on. All our data was stored on a Novell 3.11 server and we constantly used the NeXTSTEP Novell gateway object to transparently copy our files to and from the server as if it was a local NTFS drive. This was back in 1993! In fact, with the superpower of NeXTSTEP, one of the earliest incarnations of DoomEd had Carmack in his office, me in my office, DoomEd running on both our computers and both of us editing one map together at the same time. I could see John moving entities around on my screen as I drew new walls. Shared memory spaces and distributed objects. Pure magic. We wrote all of DOOM and Quake's code on NeXTSTEP. We debugged the code in NeXTSTEP with DOOM and Quake's 320x200 VGA screen drawing in a little Interceptor window while the rest of the screen was used for debugging code. When all the code ran without bugs we cross-compiled it for the Intel processor on NeXTSTEP then turned over to our Intel DOS computers, copied the EXE and just ran the game. The DOS4GW DOS-Extender loaded up and the game ran. It was that easy. One funny and strange note: I'm left-handed but while using NeXTSTEP I used a right-hand mouse. Then when I ran the game on my DOS computer I switched hands and played left-handed. Windows is not yet hardcore enough for me to switch to right-handed mice. I'll bet you didn't know that DOOM, DOOM II and Quake weren't the only games developed on NeXTSTEP. When I got Raven Software to agree to develop Heretic for us I had them buy several Epson NeXT computers (Intel based) and I flew up to Madison, WI to get them all set up and teach them how to develop the game with our tools and engine. It was a great time I'll never forget - seeing their team get excited about the power of the new environment and that they got the game developed and released in under a year. They signed on for another title and developed Hexen on NeXTSTEP as well. Back at home base we were busy with the beginnings of Quake - developed on NeXTSTEP of course but we also had another game that we were having developed using our tools and tech by Rogue Entertainment - the super fun, one of a kind, action RPG that was Strife. Unfortunately most people hadn't heard of the game because the publisher we sold the game to went out of business upon its launch. If you can find this game you should really play it - it's quite a ride. As I was leaving id Software in August 1996 the move to the Windows 32 platform was underway. John Carmack was porting our QuakeEd editor over to Win32 and preparing for a NeXT-less future. Several short months later NeXT made their fateful move over to Apple and a new era was begun as Steve Jobs set about changing the future. Again. Up to that point I had spent 15 years of my life working on computers that Steve Jobs was involved in bringing to the world. First the Apple II+, then the IIe, the IIgs and finally NeXT. Maybe someday I'll get one of those kickass iMacs.

I could have just posted the last paragraph but this article gives an idea of the NeXT software usefulness as well as other games the low spec Amiga would have struggled with. The 2-3 fps of Doom on the NeXT 68040 color workstation was likely because of VGA simulation and lack of financial incentive to optimize or even release a version for NeXT, a problem for the small high performance 68k Amiga market then and today.

bhabbott Quote:

Let's face it, the only viable platform to release Doom on in 1993 was the PC, and it needed the power of a 486 to run at a reasonable speed. At that time id was not looking at releasing it on any other platform. They even refused to port it to Microsoft Windows.

The Mac 68040 market may have been viable. I expect there were a few million 68040 Macs considering even FPU-less 68040s were sufficient. They were Mac fans but maybe they were waiting for PPC and NeXT/Mac OS X on the Mac?




The only problem is that Macs market share took a nose dive with PPC starting in about 1994.



The curse of PPC lived on even though Jobs tried to kill it several times.

bhabbott Quote:

Here's what is relevant. DoomAttack won't run on a stock CD32 because it needs 4 MB of FastRAM. When that is added, Doom runs much faster on the CD32 with Akiko C2p than it does on a 16 MHz 386SX. In fact as these benchmarks show, it is even faster than a 40 MHz 386SX, and only 26% slower than an i386DX-33 (5.8 fps vs 6.89 fps on standard settings).

A stock CD32 has no chance of running Doom unless it is severely cut down like the SNES version. Even the CD32+ successor with 68EC030@28MHz originally due in fall 1994 still has little chance due to the plan to stay at 2MiB chip. The 3DO, Saturn and soon PS1 competition used 3MiB of memory and the CD32+ could have been upgraded with 1MiB of fast to answer, possibly allowing a few tricks to get Doom running. The 68k has about 20% better code density than x86 which should save at least 100kiB from the original Doom executable alone. Better data compression and the game may have run without being cut down much if any. With fast memory and 3x the AGA bandwidth, it likely would have been a relatively cheap Doom playing console. Would CBM have been smart enough to upgrade the CD32 and Amiga 1200 to 4MiB of memory just to be cheap hardware to play Doom? Not likely as even 3MiB wasn't planned even though adding fast memory was a relatively cheap way to increase performance and software porting attractiveness. CBM went out of business for good reason.

@matthey

Quote:

The Mac and PC DOS were not as efficient for productivity because they lacked preemptive multitasking.

PC had Mac ported MS Excel with Windows 2.x runtime version in October 1987. Mac ported MS Excel 2.0 for Windows 2.x in 1988. MS Excel 1.x was only available for the Mac.

Microsoft released Excel 2.1 for Windows 2.x in 1988.

A500/A2000 didn't have VGA's stable 640x480p 16 colors.

MS Excel for Mac's release includes a MacOS task switcher improvement.

Mac has stable "business" high resolution.

https://www.youtube.com/watch?v=PI23HAEN63c
The Rise of Microsoft Excel: Part 1

OS's "multitasking" is meaningless without the "killer app" for the target market.

Quote:

The curse of PPC lived on even though Jobs tried to kill it several times.

Market share hides absolute numbers e.g. 1.2 million PowerMac sales in 1994.

Quote:

A stock CD32 has no chance of running Doom unless it is severely cut down like the SNES version. Even the CD32+ successor with 68EC030@28MHz originally due in fall 1994 still has little chance due to the plan to stay at 2MiB chip.

68EC020-25 and 68EC030-25 were available in 1992.

https://archive.computerhistory.org/resources/access/text/2013/04/102723262-05-01-acc.pdf
Page 119 of 981

For 1992
68000-12 = $5.5
68EC020-16 PQFP = $16.06,
68EC020-25 PQFP = $19.99,

68EC030-25 PQFP = $35.94
68030-25 CQFP = $108.75, follows Intel's 386DX price guide as #metoo.

68040-25 = $418.52
68EC040-25 = $112.50, braindead for DMA-equipped desktop 68K computers.
---
Competition

AM386-40 = $102.50, the status quo breaker.
386DX-25 PQFP = $103.00,

486SX-20 PQFP = $157.75,
486DX-33 = $376.75,
486DX2-50 = $502.75,

Discounts are not included.

The IPC difference between 68EC030 and 68EC025 is minor.


http://www.bambi-amiga.co.uk/amigahistory/leweggebrecht.html
Lew Eggebrecht's compute boost for baseline Amiga's AA+ models comes from AT&T's DSP.

Last edited by Hammer on 13-Jul-2024 at 02:09 PM.
Last edited by Hammer on 13-Jul-2024 at 01:56 PM.
Last edited by Hammer on 13-Jul-2024 at 01:53 PM.
Last edited by Hammer on 13-Jul-2024 at 01:52 PM.

_________________
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 7950X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB

@Hammer

Quote:

Hammer wrote: OS's "multitasking" is meaningless without the "killer app" for the target market.

If you had ever run a business you would (or should) know that the 'killer app' is an accounting package. Word processors and spreadsheets have their place, but are not the right tools for handling sales, purchases, inventory, and tax. Every business needs that stuff. OTOH, being able to run that accounting package and do word processing, spreadsheets, web browsing and email etc. at the same time is a distinct advantage.

Take the shop I worked in for 13 years for example. Over half of our sales were mail order, by phone or internet. Typically a customer would ring up and tell us what they wanted. We would then create an invoice, find the products in the inventory and add them to the bill. The customer might then give us credit card info to do the transaction on the eftpos machine, or pay by direct debit and so needed the total. This had to be done in real time.

So what if I was working on eg. a spreadsheet or text document at the time? Without multitasking I would have to save my work, close down the other program, and load up the accounting package (which took a while) as the customer waited. Then afterwards I would have to close down the accounting package and reload whatever I was working on. The delays, loss of productivity and frustration at having to work this way would quickly convince me to look for a computer with multitasking (if such a thing was available).

There were many accounting packages, word processors and spreadsheet programs available. Even the Amiga (which wasn't designed primarily for business use) had several suitable apps in each of those categories. But the one thing DOS PCs didn't have was multitasking. The real 'killer' was the OS, not a particular app. If it wasn't for Windows 98/2000 (and an accounting package that worked in it) we would have used Amigas in our shop, like my friend and I did in our previous businesses.

Quote:

A500/A2000 didn't have VGA's stable 640x480p 16 colors.

Didn't need it. If you were using an A500 in your business then you already valued low price over a fancy hires display. The A2000 was effectively just an A500 with slots and a bigger case and power supply, but could have had a graphics card if there was enough demand for it. There wasn't, because it wasn't needed. People who needed flicker-free hires just installed a flicker fixer card.

In business the goal is to make money, not admire aesthetics. 99% of PCs in the 80's were run in text mode because unless you were doing CAD or desktop publishing it was more productive. So long as you could read the text the resolution was irrelevant. The only problem with this usage was that DOS wasn't multitasking.

But any talk about the effect of hardware or software on a machine's marketability is irrelevant when that machine isn't a PC or Mac. The business computer industry settled on the PC in 1981, and weren't interested in any other platform after that. The Mac managed to sneak into second place in the education and fledgling desktop publishing markets, but its impact on the industry as a whole was minimal. By the time the Amiga arrived it was too late to take a significant share of the business market because IBM compatibility was the standard. The PC even got just as many games as the Amiga - despite not being as suited to that market - simply because it was already there.

Even if the Amiga had the most amazing hardware and software you could imagine for a bargain price, it still wouldn't get anywhere in the PC market - unless it was a PC. In the real world you can't break the laws of physics, so such a combination was impossible anyway - and expecting it is delusional.

Furthermore, all but the smallest businesses were not so concerned about price as they were compatibility. So long as the computer made more money than it cost the price was largely irrelevant (it would be written off to nothing in the end). But compatibility was vitally important because support for other platforms was bound to be less, which would be a hassle. Businesses had enough to worry about without that, thus "Nobody ever got fired for buying IBM".

Quote:

For 1992 68EC020-16 PQFP = $16.06, ... Competition AM386-40 = $102.50, the status quo breaker. ... 486DX-33 = $376.75,

These CPUs weren't competition. Once you had built a machine around them the price difference was even greater, putting them in a totally different market.

The CD32 was designed to be:-
1. Compatible with the A1200 so that it would be part of the same 'family' and increase the user base.
2. As cheap as possible while still achieving 1.

IMO that was a rational choice. Only problem is it was implemented 2 years too late. One could say that was Commodore's 'fault', but they had a long history of taking years to get new machines out so we shouldn't have been surprised. This is all water under the bridge and not worth going over again. The fact is that Commodore did get the A1200 and CD32 out, which allowed us to have them 30 years later. So we should drop these silly discussions about Amiga vs PC etc. and just enjoy using them.

The challenge now is to refute the claim that "a stock CD32 has no chance of running Doom unless it is severely cut down like the SNES version". I don't think that's true. The only way it needs to be cut down is lowering the resolution to 160x100, like Dread. This is no different from running a smaller window on a low spec PC, except you don't need a big monitor to get a viewable window size. With lower resolution the textures can also be reduced to 1/4 the size, saving memory. If some levels are too large they can be split into smaller sections. The CD will be used for music so you don't need any memory for that.

Apart from reducing resolution, the frame rate can probably be improved by making the code more efficient. I bet that profiling will reveal places where it can be sped up, especially if more of it is converted to asm. Will it be worth it? Perhaps not, it all depends on how far we want to take it. Personally I would be happy seeing Doom running on a stock CD32 at 3-5 fps, just to show that it can be done.

On a related theme, Wolfenstein 3D could also run on a stock CD32 without compromising the game design. I have run Wolfenstein 3D for Amiga on a stock A1200 with 2MB, and while slower than with the 'required' FastRAM it's still very playable.