Was anyone else disappointed with the A1200?

[hammer]

Originally Posted by Bruce Abbott

No, it doesn't.

The proximate reason Macs are so slow running Doom is that they don't have a 320x200 video mode. Their video hardware is so crude that they can't optimize and have to write 4 times more bytes to get the same screen size. So much for chunky pixels!

20 Mhz NuBus 90 upgrade from Quadra 700's 10 Mhz NuBus is LOL.

VLB is linked to 486's baseline clock speed with 486DX-50 being the fastest VLB with Am486DXL-40 coming second. Mainstream 486s with SX2 or DX2 have a lower cost 25 or 33 Mhz VLB. VLB (VESA Local Bus) is closer to the idea with Amiga's CPU local bus with a direct graphics solution attached.

The lowest resolution mode for Quadra 800 and 840av is 512x384 (196,608 pixels).
https://everymac.com/systems/apple/m...uadra_800.html
https://everymac.com/systems/apple/m...dra_840av.html

512x384's 196,608 pixels resolution is 3X when compared to 320x200's 64,000 pixels. For 320x200 render, run MacDoom II in small mode and 512x384 desktop resolution.

Unlike Shapeshifter, the Fusion solution preserves the real MacOS's change resolution feature. Shapeshifter is missing a few real Mac features.

In Caffeine OS's P96 RTG via PiStorm-Emu68, Fusion (with Quadra ROM image) can change desktop resolution inside MacOS which includes 512x384 resolution. My A1200 with PiStorm-Emu68 doubles as my retro MacOS 7.5 and 8.x in my early university years.

Originally Posted by Bruce Abbott

BTW Doom runs silky smooth on my A600, even in 640x480. That's what a Vampire does for it.

The PC has a "standalone Vampire" and is it's called the latest PC.

Vampire is a "what if" when Amiga's development path continued like a PC's preserve legacy while moving forward instead of following Macs.

[Thomas Richter]

Originally Posted by Bruce Abbott

I thought it just moved data memory around to avoid fragmentation. I can't imagine swapping to disk being viable when you only had a 400k floppy drive.

It did move memory around, but it also moved resources (or handles, to be precise) out of memory that were not locked. Those resources could be brought back in memory once they were needed. What MacOs did not do is to write changed resources back to disk, but that was typically not required due to the organization of its binaries, where code was just one resource out of many. That's quite different from the Amiga where you just put a couple of structures or "dc.b" statements into your code. MacOs had a very rich structure in its binaries that consisted of a collection of resources.

Originally Posted by Bruce Abbott

Not only did it require programs 'following rules' (which Amiga OS also requires) it limited the instructions you could use and/or made the code messy and inefficient.

So inefficient that it could run more than a system without handles?

Originally Posted by Bruce Abbott

Even worse, you never knew when the OS might decide to compact memory, which could take quite a while. This was incompatible with an OS that was supposed to be capable of real-time animation.

Then don't allocate memory during the animation.

[AestheticDebris]

Originally Posted by Bruce Abbott

Not only did it require programs 'following rules' (which Amiga OS also requires) it limited the instructions you could use and/or made the code messy and inefficient. Even worse, you never knew when the OS might decide to compact memory, which could take quite a while. This was incompatible with an OS that was supposed to be capable of real-time animation.

I'm not sure "real time animation" was considered a big deal for the Mac, it's focus initially was really the DTP space. Unless you mean on the Amiga? As I said, you couldn't really bring that approach over to a pre-emptive system anyway, since it relies quite heavily on knowing that an application is in a safe state when the system purges resources from memory.

I'd still say AmigaOS was, on the whole, a better experience (having to manually allocate RAM to an application in MacOS felt insane) but I'd have taken a version with an MMU powered VM and swap file in a heartbeat.

[hammer]

Originally Posted by AestheticDebris

I'm not sure "real time animation" was considered a big deal for the Mac, it's focus initially was really the DTP space. Unless you mean on the Amiga? As I said, you couldn't really bring that approach over to a pre-emptive system anyway, since it relies quite heavily on knowing that an application is in a safe state when the system purges resources from memory.

I'd still say AmigaOS was, on the whole, a better experience (having to manually allocate RAM to an application in MacOS felt insane) but I'd have taken a version with an MMU powered VM and swap file in a heartbeat.

Some 68K Macs are real-time animation capable.

For Quadra 660AV and 840AV at 4:13
[ Show youtube player ]

Adobe Premiere NLE 68K for MacOS works on PiStorm-Emu68 Amiga.

[hammer]

Originally Posted by Bruce Abbott

How exciting for you.

I believe EMU68 can allocate up to 2GB of RAM to the Amiga. Should be enough...

Atm, Warp3D hardware acceleration for PiStorm-Emu68 only works for Video Core IV IGP equipped RPi e.g. RPi 3A+, hence no 2GB RAM.

RPi 3A+ has 512 MB of memory.

RPi 4B has Video Core VI IGP.

[AestheticDebris]

Originally Posted by hammer

Some 68K Macs are real-time animation capable.

For Quadra 660AV and 840AV at 4:13
[ Show youtube player ]

Adobe Premiere NLE 68K for MacOS works on PiStorm-Emu68 Amiga.

Sure, but by then they were MMU equipped machines with a harddrive and System 7 which supported more comprehensive memory management.

[Bruce Abbott]

Originally Posted by hammer

Vampire is a "what if" when Amiga's development path continued like a PC's preserve legacy while moving forward instead of following Macs.

Not for me it isn't. My Vampire is a replacement for the A3000 with 060 and RTG that I used to have. It's pretty much equivalent to the best 68k Amiga you could have back in the 90's. Efforts to go beyond that don't interest me much, which is why I am sticking with the V2 Vampire in my A600.

[CCCP alert]

The 486 was really the first powerful AND affordable Intel CPU. Commodore had no clue where to to get back the 68000 vs 8086 advantage they had in the 80s.

040 was OK but 68060 and PPC was not really value for money vs AMD x86 chips or faster for th sam CPU price. You needed a PPC A1200 to play Wipeout vs $299 PS1, PS1 killed the 'cheap home computer for gaming". Commodore had nowhere to go, nor did Acorn or Atari and the 25mhz 040 £1100-1200 Mac LC475 was the last value for money Mac.

[TEG]

Originally Posted by CCCP alert

040 was OK but 68060 and PPC was not really value for money vs AMD x86 chips or faster for th sam CPU price. You needed a PPC A1200 to play Wipeout vs $299 PS1, PS1 killed the 'cheap home computer for gaming". Commodore had nowhere to go, nor did Acorn or Atari and the 25mhz 040 £1100-1200 Mac LC475 was the last value for money Mac.

Commodore should have had the custom chips to compensate the weak 68040, it was all their strength, but they did not managed to make it.

[hammer]

Originally Posted by Bruce Abbott

Not for me it isn't. My Vampire is a replacement for the A3000 with 060 and RTG that I used to have. It's pretty much equivalent to the best 68k Amiga you could have back in the 90's. Efforts to go beyond that don't interest me much, which is why I am sticking with the V2 Vampire in my A600.

Your non-gaming use case on the Amiga platform is a minority. I used my A3000 for games for most of the time.

Vampire V2 has a "turtle" mode feature. AC68080 V2 clone didn't exist in the 1990s and it has a performance difference when compared to 68060 Rev 6 @ 100 Mhz OC. The best-equipped wedge Amiga in the late 1990s would be Phase 5's BlizzardPPC/68060 with BVision (mini PCI connected) Warp3D RTG.

Instead of Vampire, I have my PiStorm-Emu68 equipped A500 ECS and A1200. Without RTG active, the A500 ECS with PiStorm-Emu68 is just a super fast A3000.

Gaming entry vector leads to other non-gaming use cases. Learn how NVIDIA dominated the gaming PC market and attacked workstation graphics vendors like SGI and 3D Labs. NVIDIA's superior economics of scales and excellent game developer relationships beats SGI and 3D Labs.

There's only a tiny few thousand of A3000/A4000 production scale and it wouldn't sustain Commodore. A tiny few thousand of A3000/A4000 production scale would also limit the Zorro IIII addon market's economics of scale. Commodore didn't build a mass-produced Amiga model with a single Zorro III which is akin to a gaming PC's single active PCIe 16X lanes or AGP or tiny few VLB slot idea. Workstation PC has many fast slots vs gaming PC's few fast slots.

[hammer]

Originally Posted by AestheticDebris

Sure, but by then they were MMU equipped machines with a harddrive and System 7 which supported more comprehensive memory management.

In 1993, Apple was able to mass-produce 68LC040-25 based Macs for about $1000 USD far more than Commodore's tiny few thousand of A3000/A4000s. Mac 660AV and 840AV models have multimedia acceleration via DSP3210 55 Mhz to 66 Mhz.

Apple's Macintosh Display Card 8·24 GC for its Macintosh IIfx has Quickdraw acceleration via AMD's 29K RISC CPU@ 30Mhz (with 64K static cache, 2 MB VRAM, and 2 MB optional video DRAM).

[hammer]

Originally Posted by TEG

Commodore should have had the custom chips to compensate the weak 68040, it was all their strength, but they did not managed to make it.

68LC040 @ 28 Mhz with a 3D custom chip @ 50Mhz would effectively be a desktop PS1. Without mass production on the A500 scale, 3D custom chips would be largely unused.

The problem is Commodore sells A4000/040 in a similar price range as Pentium 60 PC clones.

Commodore is unwilling to provide mass-produced 68LC040-25 Amigas in the $1000 range. The price gap between A1200 and A4000/030 @ $1599 is largely covered by Commodore's 486 PCs.

[Bruce Abbott]

Originally Posted by hammer

Your non-gaming use case on the Amiga platform is a minority. I used my A3000 for games for most of the time.

I doubt that. The combination of KS2 and 25MHz 68030 meant that many games wouldn't work on the A3000, and it was 4 times the price of an A500 system. Only a rich fool would buy an A3000 to play games.

Vampire V2 has a "turtle" mode feature. AC68080 V2 clone didn't exist in the 1990s and it has a performance difference when compared to 68060 Rev 6 @ 100 Mhz OC. The best-equipped wedge Amiga in the late 1990s would be Phase 5's BlizzardPPC/68060 with BVision (mini PCI connected) Warp3D RTG.

Unfortunately I bought my Cyberstorm 060 a little too early, so it only had a rev 5 chip. At 66MHz it was a bit flaky so I settled for 60MHz. If I could have overclocked to 100MHz I would have! Vampire V2 is approximately equivalent to an 80MHz 060.

I already have a Blizzard 1230-IV in my A1200 since the 90's so I didn't want to change that. I used to have an A600 with 33MHz 030 but stupidly threw it away in 2015 because the keys got yellowed! (dark times when I thought the Amiga scene was dead). So that was another reason to choose the A600 (got one off eBay for NZ$300).

Instead of Vampire, I have my PiStorm-Emu68 equipped A500 ECS and A1200. Without RTG active, the A500 ECS with PiStorm-Emu68 is just a super fast A3000.

That's nice. Unfortunately PiStorm didn't exist when I bought the Vampire.

Gaming entry vector leads to other non-gaming use cases. Learn how NVIDIA dominated the gaming PC market and attacked workstation graphics vendors like SGI and 3D Labs. NVIDIA's superior economics of scales and excellent game developer relationships beats SGI and 3D Labs.

More irrelevant PC noise.

There's only a tiny few thousand of A3000/A4000 production scale and it wouldn't sustain Commodore. A tiny few thousand of A3000/A4000 production scale would also limit the Zorro IIII addon market's economics of scale.

This is true.

However the A4000 did have a use case with the Video Toaster, and might have been attractive to those rich people who would previously have bought an A2000 (quite a few around here). Commodore knew this and put a higher margin on the A4000 to compensate for the lower sales volume. Some Amiga fans complained about it being too expensive, but they didn't understand that for professional use it was cheap enough (A4000 with Toaster was much cheaper than alternatives). Commodore could have dropped the price by a few hundred dollars but that still wouldn't make it cheap enough for the average Amiga fan, so there was no point.

Which makes the A1200 vitally important for Commodore's survival. They needed to get as many AGA machines out as possible to attract developers. Unfortunately by this time Commodore was running out of steam so they couldn't produce enough to meet demand. I regret not buying as many A1200s as I could get my hands on to sell in the shop.

Commodore didn't build a mass-produced Amiga model with a single Zorro III which is akin to a gaming PC's single active PCIe 16X lanes or AGP or tiny few VLB slot idea. Workstation PC has many fast slots vs gaming PC's few fast slots.

Gamers today don't buy single slot workstation PCs, they buy big box PCs with fast CPUs that they can put the best possible graphics card in. Hardcore gamers put water cooling systems in them to enable max overclocking.

The problem with the single Zorro III slot idea is that it becomes much more expensive for little gain. The engineer who was designing the A1000+/jr wanted to just put an edge connector on it like the original A1000 and A500 - only 32 bit like the A1200. Unfortunately this design was delayed by efforts to make the case cheaper, in yet another example of engineers not appreciating the urgency of getting products out ASAP.

But even if it did get out the door, with one or two slots this mid-range machine would be encroaching on the price of a 386SX PC. More Amiga fans would buy it than the A4000, but not as many as would buy the A1200 - which had equivalent functionality for a much lower price.

The A1200 set a base level with sufficient sales potential that developers could justify supporting it, while still permitting virtually unlimited expansion to keep it up to date. The all-in-one design also made it easier to sell and use, particularly for those who had enough funds for more than a console but not enough for a gaming PC.

[Bruce Abbott]

Originally Posted by hammer

In 1993, Apple was able to mass-produce 68LC040-25 based Macs for about $1000 USD

The retail market for such a machine was limited, as shown by the Quadra 605's short lifespan of only 1 year. You should also realize that they were sold bundled with a keyboard/mouse and monitor, which raised the price dramatically. Apple's efforts to court the low end didn't turn out well - by 1995 sales were tanking and they were headed for bankruptcy.

Despite having a powerful CPU the Quadra 605 was not a competitor to the A1200. Gaming performance was weak. It didn't have the CD-ROM drive required for the latest hot Mac games, and external SCSI CD-ROM drives were expensive. You couldn't connect it to your TV so you had to buy the expensive Mac monitor (plus a keyboard and mouse of course), making the real price 4 times more than an A1200. The A1200 could run many awesome Amiga games which were ahem available at very low cost. The Amiga also had a more vibrant gaming community which the A1200 tapped into.

[oscar_ates]

Originally Posted by Bruce Abbott

The retail market for such a machine was limited, as shown by the Quadra 605's short lifespan of only 1 year. You should also realize that they were sold bundled with a keyboard/mouse and monitor, which raised the price dramatically. Apple's efforts to court the low end didn't turn out well - by 1995 sales were tanking and they were headed for bankruptcy.

Despite having a powerful CPU the Quadra 605 was not a competitor to the A1200. Gaming performance was weak. It didn't have the CD-ROM drive required for the latest hot Mac games, and external SCSI CD-ROM drives were expensive. You couldn't connect it to your TV so you had to buy the expensive Mac monitor (plus a keyboard and mouse of course), making the real price 4 times more than an A1200. The A1200 could run many awesome Amiga games which were ahem available at very low cost. The Amiga also had a more vibrant gaming community which the A1200 tapped into.

If the A1200 was so perfectly engineered machine, why it failed to save commodore going bust?

[Thomas Richter]

Originally Posted by oscar_ates

If the A1200 was so perfectly engineered machine, why it failed to save commodore going bust?

Wrong audience. It was seen and sold as a toy. The Mac was seen and sold as a tool.

[Bruce Abbott]

Originally Posted by CCCP alert

The 486 was really the first powerful AND affordable Intel CPU. Commodore had no clue where to to get back the 68000 vs 8086 advantage they had in the 80s.

Depends on what you call 'affordable'. The A1000 debuted at US$1,285 for the base unit. By that time more powerful 8MHz 286 PCs were selling for $3000. Granted the graphics and sound capabilities of those machines was not nearly as impressive, but that wasn't due to the 68000. By 1987 when the A500 debuted at $699, 286's running at 12MHz were coming with a hard drive and EGA monitor for $3000.

The real advantage of the Amiga wasn't the CPU, it was the custom chips that allowed it to do more with less. That made it more affordable, but the price ratio between a decked out PC and base model Amiga was about 4:1, same as it was in 1992 when the A1200 was released.

040 was OK but 68060 and PPC was not really value for money vs AMD x86 chips or faster for th sam CPU price.

Value for money is the wrong metric. By this time all the hot games and apps were only coming out on the PC, not the Amiga. So it didn't matter how much value for money the Amiga might have had, it was still useless.

You needed a PPC A1200 to play Wipeout vs $299 PS1, PS1 killed the 'cheap home computer for gaming".

Consoles were always cheaper than computers. Most fans bought an A1200 because they wanted to do more than just play games on it.

The PlayStation was released in Europe on 29 September 1995. By this time Commodore was already stone cold dead over a year ago. So clearly the PS1 had nothing to do with their demise.

Wipeout was a boring game that could have more effectively been implemented without texture-mapped 3D. What was the point - to copy the PlayStation? (most of its games were also boring with poor 3D effects). Compare it to Doom, released on the Amiga a year earlier and only requiring a 50MHz 030 (which many A1200 owners had by then).

Commodore had nowhere to go, nor did Acorn or Atari and the 25mhz 040 £1100-1200 Mac LC475 was the last value for money Mac.

Commodore could have gone quite a bit further if it concentrated on delivering 'more for less' and managed to attract software developers to keep the software flowing. People use computers to run software, not drool over hardware specs.

In 1988 I bought a PlayStation not because of the hardware specs, but because I wanted to play Tomb Raider and it was a lot cheaper than a Pentium PC. I also had an A3000 with 060 running at 60MHz and RTG which was equivalent to a Pentium PC, but no Tomb Raider for it!

[Bruce Abbott]

Originally Posted by Thomas Richter

Wrong audience. It was seen and sold as a toy. The Mac was seen and sold as a tool.

A tool for the elite who didn't care about price. It did well in education and had a niche in desktop publishing, but apart from that...

Regular people bought a PC of course, as it was the industry standard available everywhere.

The Amiga was indeed seen as a toy, but that applied to many other things too including expensive cars, yachts etc., and of course game consoles. Later on, when the PC became good enough to run good games, it was also seen as a toy. The 'gaming' PC is still a toy today!

[Thomas Richter]

Originally Posted by Bruce Abbott

A tool for the elite who didn't care about price. It did well in education and had a niche in desktop publishing, but apart from that...

..and by that, Apple survived (or at least was seen important enough to be saved by Microsoft at a point). Suprise, suprise!


So what does that tell you? That CBM hat possibly the wrong strategy when "selling to kids for computer games"?


Mac users had money. CBM users did not.

Originally Posted by Bruce Abbott

The Amiga was indeed seen as a toy, but that applied to many other things too including expensive cars, yachts etc., and of course game consoles.

And, did it surive?

Originally Posted by Bruce Abbott

Later on, when the PC became good enough to run good games, it was also seen as a toy. The 'gaming' PC is still a toy today!

The PC was perceived as tool, and thus taken serious, no matter what people used it for. The Amiga wasn't.


What matters are the people that buy the premium prices. For Apple, and for PC.

[AestheticDebris]

Originally Posted by Thomas Richter

So what does that tell you? That CBM hat possibly the wrong strategy when "selling to kids for computer games"?

Not really, the SNES and MegaDrive were doing perfectly well in that space despite games costing a fortune, so it's not like there wasn't a market or any money to be made in that space.

The problem with the Amiga was that CBM was still trying to keep a foot in both markets and being master of none. Failing to pick a lane and stick to it (and delivering appropriate hardware/software for that choice) was their biggest mistake.