Was anyone else disappointed with the A1200?

[Thomas Richter]

Originally Posted by Foebane

Isn't RTG chunky by design? Can't Amiga games run on RTG without modification?

Actually, no. Most graphics cards of this age supported also planar modes, though only four bitplanes (16 colors) at most. They were not very popular because the whole way how to address them is rather clumbsy. You first had to write in one register the plane you wanted to paint into, then do the operations, then switch planes again. It's some kind of legacy with EGA, likely.

P96 supports this type of planar, for example.

[grond]

Originally Posted by roondar

I think I finally get the issue here. You've misunderstood what I mean when I say that one CPU grade didn't make that much difference. I was not talking about money, I was talking about performance.

Well, I stated many times that a CPU grade made a three-digit difference in the major currencies. The hypothetical Amiga-saviour computers I described would have come out instead of the A3000 and A500+, respectively, and would have had AA+ (as I now know it was codenamed) and CPU cards with fastmem slot and a variety of 020s and 030s to choose from. That was the time when (as you yourself pointed out) another MB of RAM made a lot of a difference pricewise. However, Commodore should have raised the base-CPU to the 020 already with the introduction of the successor to OCS (which should have been AA+ and not just ECS). They did implement a 32bit chipmem bus for the A3000, why oh why did they cut it down to 16 bit again for the A500+/A600? (Easy answer: because they used a 68000 that only had a 16 bit bus).

And then you point to the 50 MHz 030 and say "AA+ wouldn't have mattered, people could just have used a 50 MHz 030 instead" (or that's how it came across). A chunky display with 020/030 would have made flat-shaded and gouraud-shaded 3D space shooters of the X-Wing type feasible on the Amiga (another game that ran very well on my friend's 386SX25) and many other great games. And this Amiga would have been in the market waiting happily for somebody to write Doom. With Doom even the last empty fastmem slot would have got some precious fastmem, 020s would have been upgraded to fast 030s and so on.

Secondly, your prices are for 1990. We're talking about a game launched December 1993. They're not actually irrelevant for this discussion. What is relevant is the prices circa December 1993 when everyone wanted to upgrade to play doom.

No, not really. The suddenly required enormous step up killed the Amiga. If the base configuration had grown with time in a meaningful way, the required upgrade step wouldn't have been as big. People able to run Doom in a small window at a low framerate are more likely to buy upgrades than what the situation was really like: "there might be good new games if all of you upgraded". This wasn't going to happen (simple game theory (the field of mathematics)). A 25 MHz 030-A3000 with Z3-RTG should be able to play Doom well enough (I know, you disagree). But too few people actually had an Amiga like that. With AA+ and swappable CPU cards (and not wasted 020 on the PCB) it could have happened.

Furthermore, 386s had also been around for a long time before Doom came out. Such a groundbreaking game is more likely to get published if there already is an installed user base for two reasons: it makes more sense economically to only write games for which there is an audience and there are more programmers able to play around with the capabilities of that hardware. The AA+ Amiga would have had to already be around to get exciting games such as Doom. It wasn't enough to start putting something together that would have been capable of running Doom when Doom was already out.

Secondly, planar does not mean 'multiple blits'. I'm rather surprised you didn't know that. All you need to do is interleave the bitplanes in memory and then you need just a single blit per layer. It's what Amiga programmers have been doing for decades now.

Um, yes, but then your cookie cut mask will be as large as the graphics you are blitting and thus you will be as inefficient as ORing two chunky buffers together. Or not?

And, btw, I'm not sure about this, can interleaved bitmaps scroll into individual ways?

Thirdly, claiming that you can simply add additional hardware to fix the limits of chunky mode is really pushing credibility.

Where did I do this?

we could add a 'byte-mode' to planar to remove the c2p overhead, but keep the chip internally work in bitplanes and call that 'planar'.

I don't understand. Please explain.

Lastly, it's worth noting that dual layer scrolling using Bitplanes is more efficient than chunky if we level the playing field. Meaning, when we have zero hardware support for any form of scrolling on either. Chunky needs a second layer in hardware for any form of efficiency here, planar does not.

I don't understand. Assuming we could address single bytes in chipmem with screen DMA, a chunky layer could do pixel-wise scrolling without any additional hardware while a planar mode needs a barrel-shifter for each bitplane. Since we address chipmem word-aligned, the chunky mode would need a very simple FIFO to achieve the shifting (which would only be either "pass unshifted" or "pass shifted by one pixel"). There is nothing that makes scrolling in hardware easier in planar than in chunky. Actually it is the other way round. For a scrolling chunky display you need less temporary storage and less shifter units.

Quite a few, actually, if we look at 2D console sales in the early 90s. Millions more than bought Doom, that's for sure.

Yes, and that's probably why Commodore did the CD32. That wasn't powerful enough either but it shows that they recognised that they couldn't keep up in the computer business but tried in the console business.

Funny thing that - they didn't do transparent fog of war in Warcraft 2.

Why should it have been transparent to make use of planar? Just set all palette colours which have the fog-of-war-plane set to black.

More seriously - there is absolutely no reason other than market size that such 2D games couldn't be done on the Amiga.

I think you are dramatically overestimating the capabilities of the blitter as we have it. I did mention that if Commodore had done things right, they would have updated the blitter to 32bit accesses. That would have been quite simple, too.

PC 2D render speed had exactly nothing to do with chunky modes.

Yes, that's right. There were many shortcoming in the AGA chipset that could have been sorted out with only very little extra effort. The slow chipmem bus is one of them. Z3 preceded AGA and had better bandwidth.

The Amiga has the elegant hardware 2D approach where almost no data needs to be pushed over the bus for scrolling. The PC had the stupid brute-force approach where for scrolling all the screen content had to be updated. However, there was a point in time when the stupid approach was just as fast as the elegant approach. Changing from Amiga to PC was perhaps made easier by the fact that the PC could not only do what the Amiga couldn't do but could now also do what the Amiga could do. No more "only Amiga makes it possible".

You've pointed out your credentials in regards to electronics and designing silicon many times. Do I seriously have to point out to you, of all people, that merely adding the same amount of additional bandwidth that PC graphics cards got at the time to the existing planar display of the Amiga would've allowed it do everything the PC did in 2D mode?

Are we trying a bit of that ad-hominem stuff here?

I've also never argued that the Amiga was going to do much better if only it had better hardware. That's what you seem to think. If only it had chunky...

Yes, that's what I think. But I didn't pin it all on the lacking chunky mode. I said that Commodore should have steadily invested into R&D and should have raised the base configuration in a way that would have kept the entry-level price more or less constant rather than selling the same old technology for always less money.

The omission of a chunky mode was probably the biggest single fault and was my example to show how quickly put together AGA really was. They didn't even spend the time to add this tiny modification that any sane engineer (I think those were my words) would have made if s/he had been allowed to.

Thanks for the info, 1000 transistors isn't at all bad. But I'm not entirely sure that includes the path in the chip that allows the switching between the various modes.

Well, I'm sure an unused bit in an already present register could have been found for that purpose. Then add say four transistors for the drivers that drive the input of the switches.

If it's that cheap, it also makes me wonder why they didn't just add it in during the original Amiga design. They kept HAM and that was something Jay Miner thought was pretty pointless in an RGB chip.

Because the chipmem was too slow for a then wasteful 8bit chunky mode. Remember: 8 bit chunky only would have come (almost) free the moment they made an 8 bit planar mode, too. The OCS chipmem bus is already very slow with just 5 or 6 bit planes active. No way they could have wasted the bandwidth equivalent to 8 bits planar in that design. The CPU wouldn't have got any chipmem cycles at all which makes it rather pointless.

As for the other pixel modes. I have a genuine question: if they're complicated to guestimate (as you seem to say), then why did you say earlier they would be extremely cheap to add?

There is a difference between a rough estimate (which you didn't find sufficient) and a more thoroughly made calculation. I'm not willing to think how I would have implemented some graphics modes that went into that paper, propably would have been cut by management anyway and that I personally consider superfluous.

And crucially, the two years before that it sold 23 and 20 million units respectively. Which fits with my point: PS2 sales were down 20% for two years straight and only then did Sony introduce the PS3.

In other words: the PS2 was declining in sales when the PS3 was launched - it was not gaining. The Amiga 500 meanwhile, was not declining in sales in 1991, it was still gaining.

The XBox 360 was released in May 2005. This probably cut into the sales of the PS2. It kept going pretty strong, though. There was healthy competition and both companies had to improve their products. They understood what Commodore did not understand.

Nonetheless the important point remains: big companies have actually put the next generation of a product out and continued to sell the preceding generation for many years with good success. Commodore had to sell the A600 under cost because nobody wanted the same old stuff. Commodore could have offered an AA+ Amiga in 1990/91 and still continued to sell the A500 with good success. This disproves your earlier point that no company would have done that.

[grond]

BTW, RESHOOT-R is technically a great game and those overlays would not have been possible with PC graphics cards of the 90s or with just an 8 bit chunky mode. It would definitely have done well within the Amiga market but I can't see a technically more advanced 2D shooter compete with the sort of games that became popular then.

[swinkamor12]

Apple was worth less than Commodore in 1992 and now is richest company int thw World.
Even Atari live long enough to get to next level - console with 3D support.

Commodore bankrupt because base model Amiga 1200 has too slow graphics.
It is obvious for everyone outside Amiga asylum where some fanatics still does not accept reality.

chunky pixels and SIMM slots where in 1992 not rocket science and was even in cheapest pc.
It should be in Amiga 1200. If it will be in Amiga 1200 Commodore will not bankrupt.

Commodore should not force Amiga users to buy expensive upgrade.
chunky pixels and SIMM slots should be in Amiga 1200.

They can easily add chunky pixels by just bypasing planar to chunky conversion in lisa chip,
but they Commodore screw it.

[donnie]

Commodore only success was budget computers sold to kids. Their only successful models where c64 and amiga 500.

That is the market commodore had.

[Foebane]

Originally Posted by swinkamor12

It is obvious for everyone outside Amiga asylum where some fanatics still does not accept reality.

No need to be rude.

[Dunny]

Originally Posted by swinkamor12

Commodore bankrupt because base model Amiga 1200 has too slow graphics.
It is obvious for everyone outside Amiga asylum where some fanatics still does not accept reality.

Commodore died because they mismanaged their company into the ground. It was nothing to do with whether or not the Amiga was at all useful or suitable.

[Retro1234]

Previous Commodore models hadn't been compatible so they could of produced a new Non Amiga computer.

[Foebane]

Originally Posted by Dunny

Commodore died because they mismanaged their company into the ground. It was nothing to do with whether or not the Amiga was at all useful or suitable.

I heard that the head management, especially the reprehensible Mehdi Ali, couldn't care less about their products, which they saw as hopeless for profit, and spent Commodore's remaining assets on fast cars, expensive mansions and luxury holidays. What bastards. No wonder the Amiga ex-staff, freshly laid off from the defunct Commodore, were happy to burn effigies of them at barbeques.

Originally Posted by Retro1234

Previous Commodore models hadn't been compatible so they could of produced a new Non Amiga computer.

The Amiga was the best thing Commodore ever had, and they neglected it by not sufficiently upgrading it nor marketing it properly.

[Galahad/FLT]

Originally Posted by swinkamor12

Apple was worth less than Commodore in 1992 and now is richest company int thw World.
Even Atari live long enough to get to next level - console with 3D support.

You jester. Atari might had cobbled 3D support into the Jaguar, but it helps if you have software support, which quite clearly, the Jaguar NEVER had, so lets not pretend that Atari made better decisions!

Originally Posted by swinkamor12

Commodore bankrupt because base model Amiga 1200 has too slow graphics.

Wrong. Commodore went bankrupt because it was overseen by morons and a patent dispute that they couldn't afford to settle.

The hardware level in the A1200 had ZERO to do with Commodores problems.

Originally Posted by swinkamor12

It is obvious for everyone outside Amiga asylum where some fanatics still does not accept reality.

No, I think the problem is mainly yours. We are all aware of some of the Amigas shortcomings, but your revisionist bollocks won't go unchallenged.

Originally Posted by swinkamor12

chunky pixels and SIMM slots where in 1992 not rocket science and was even in cheapest pc.

You're a moron. You're comparing DESKTOP/TOWER PC machines with the A1200 that fit its entire machine inside an enlarged keyboard case.

Of course the PC came with SIMM slots, because it had lots of room inside its DESKTOP case or TOWER case ffs!

Everyone agrees the A1200 should have had a chunky mode, but dependent on how it was implemented would have determined whether it could compete with PC still. Chucking in a chunky mode says nothing about how much effor the programmer will need to get that mode operating at a decent enough FPS. The PC had a faster processor and was getting more powerful graphics cards.

Originally Posted by swinkamor12

It should be in Amiga 1200. If it will be in Amiga 1200 Commodore will not bankrupt.

As I said above, rubbish. a chunky mode on its own won't achieve anything if theres no kind of acceleration to use it properly.

Originally Posted by swinkamor12

Commodore should not force Amiga users to buy expensive upgrade.
chunky pixels and SIMM slots should be in Amiga 1200.

Repetition.

Originally Posted by swinkamor12

They can easily add chunky pixels by just bypasing planar to chunky conversion in lisa chip,
but they Commodore screw it.

Every post from you slags off the Amiga, why the FUCK are you here if the Amiga bugs you so much?

[roondar]

Originally Posted by grond

Well, I stated many times that a CPU grade made a three-digit difference in the major currencies. The hypothetical Amiga-saviour computers I described would have come out instead of the A3000 and A500+, respectively, and would have had AA+ (as I now know it was codenamed) and CPU cards with fastmem slot and a variety of 020s and 030s to choose from. That was the time when (as you yourself pointed out) another MB of RAM made a lot of a difference pricewise. However, Commodore should have raised the base-CPU to the 020 already with the introduction of the successor to OCS (which should have been AA+ and not just ECS). They did implement a 32bit chipmem bus for the A3000, why oh why did they cut it down to 16 bit again for the A500+/A600? (Easy answer: because they used a 68000 that only had a 16 bit bus).

And then you point to the 50 MHz 030 and say "AA+ wouldn't have mattered, people could just have used a 50 MHz 030 instead" (or that's how it came across). A chunky display with 020/030 would have made flat-shaded and gouraud-shaded 3D space shooters of the X-Wing type feasible on the Amiga (another game that ran very well on my friend's 386SX25) and many other great games. And this Amiga would have been in the market waiting happily for somebody to write Doom. With Doom even the last empty fastmem slot would have got some precious fastmem, 020s would have been upgraded to fast 030s and so on.

I've pointed out several times that my comments regarding the 68030 were not about a hypothetical machine, but rather what we have now. If you took that to mean that a chunky AA+ wouldn't have sped up Doom, I am sorry - that is not what I meant. I specifically mentioned in quite a few of my posts that chunky modes made Doom run better.

A hypothetical AA+ machine with more bandwidth and chunky will run Doom better. I have no doubt about that. However, I'm also not interested in it as we don't have such a machine. I've tried to point out that I was more interested in machines around the release of Doom and the actual performance we did end up getting. Perhaps I've failed there, but in my mind we were not discussing the AA+ machine for the last five or six posts on the subject of Doom.

No, not really. The suddenly required enormous step up killed the Amiga. If the base configuration had grown with time in a meaningful way, the required upgrade step wouldn't have been as big. People able to run Doom in a small window at a low framerate are more likely to buy upgrades than what the situation was really like: "there might be good new games if all of you upgraded". This wasn't going to happen (simple game theory (the field of mathematics)). A 25 MHz 030-A3000 with Z3-RTG should be able to play Doom well enough (I know, you disagree). But too few people actually had an Amiga like that. With AA+ and swappable CPU cards (and not wasted 020 on the PCB) it could have happened.

It's warm and I don't feel like repeating the same arguments - I think we both pointed out how we feel and though I could point out I still disagree, I don't think it'll change anything.

Perhaps we should just end the whole Doom and A1200 upgrade cost conversation. It's ok to disagree.

Furthermore, 386s had also been around for a long time before Doom came out. Such a groundbreaking game is more likely to get published if there already is an installed user base for two reasons: it makes more sense economically to only write games for which there is an audience and there are more programmers able to play around with the capabilities of that hardware. The AA+ Amiga would have had to already be around to get exciting games such as Doom. It wasn't enough to start putting something together that would have been capable of running Doom when Doom was already out.

I seem to recall experiments on much earlier hardware, including Amiga games. But my memory is fuzzy and I may be getting the dates wrong. Anyway, there were plenty of really nice 3D games on the Amiga by the time of Doom's release. Many of which ran like a dog without upgrades. Yet no one upgraded. On the flip side there were official ports to platforms that were accustomed to better hardware and upgrading and as far as I know, this did not cause an increase in people upgrading there.

I'll grant you this: if Commodore had brought out user-changeable CPU/RAM (and perhaps even graphics) from the moment of the A500's release it's possible more people would've been aware the Amiga could be upgraded and perhaps they would've. However, I'm not at all convinced that the same people who bought A500's would've bought such a machine and as a result am much more pessimistic about how successful such a machine would be (given how poorly the A1000 sold), but I seem to recall pointing that out before.

Um, yes, but then your cookie cut mask will be as large as the graphics you are blitting and thus you will be as inefficient as ORing two chunky buffers together. Or not?

I'm not sure I understand. You don't need to cookie cut anything for dual layer graphics. Perhaps you're referring to blitting stuff onto the buffers? That doesn't change bandwidth costs - the Blitter always fetches the mask for every word it cookie-cuts. With or without interleaving.

And, btw, I'm not sure about this, can interleaved bitmaps scroll into individual ways?

Yup.

Where did I do this?

That was where you added special hardware to display and deal with two layers to the chunky system.

I don't understand. Please explain.

What I mean is this: we were comparing chunky to planar as it existed on the Amiga and PC. The solution you've come up with for half-chunky dual layer mode is not a standard option on any PC graphics card. So, I suggested that if changing the hardware into something that is not really chunky (as normally defined) is ok to counter drawbacks, we could simply change the planar hardware to something that counters it's drawbacks and update our definition of what planar is.

I don't understand. Assuming we could address single bytes in chipmem with screen DMA, a chunky layer could do pixel-wise scrolling without any additional hardware while a planar mode needs a barrel-shifter for each bitplane. Since we address chipmem word-aligned, the chunky mode would need a very simple FIFO to achieve the shifting (which would only be either "pass unshifted" or "pass shifted by one pixel"). There is nothing that makes scrolling in hardware easier in planar than in chunky. Actually it is the other way round. For a scrolling chunky display you need less temporary storage and less shifter units.

A standard byte-per-pixel chunky screen without a hardware supported second layer can't efficiently create such a second layer. A standard planar mode screen without a second layer can. As is shown by several Amiga games that add extra layers on top of the two that were supported by default.

Yes, and that's probably why Commodore did the CD32. That wasn't powerful enough either but it shows that they recognised that they couldn't keep up in the computer business but tried in the console business.

We were not talking about why Commodore launched the CD32 or whether or not the Amiga CD32 could compete. We were talking about the number of people that still only played 2D games. My whole point here was that chunky 3D, while impressive, did not actually conquer the market. 3D only got seriously big after hardware accelerated 3D started to arrive in droves (primarily in the form of 3DFX cards and the PSX/Saturn/N64).

Why should it have been transparent to make use of planar? Just set all palette colours which have the fog-of-war-plane set to black.

I was responding to your joke with one of mine. It was not meant to be serious (hence the 'More seriously' directly following it).

I think you are dramatically overestimating the capabilities of the blitter as we have it. I did mention that if Commodore had done things right, they would have updated the blitter to 32bit accesses. That would have been quite simple, too.

We're talking about a game released in 1996. It should have been obvious that I was not talking about a 2MB chip memory only A1200 here. Perhaps it wasn't - it's been warm and I'm trying (and failing) to be slightly more brief in my posts.

Yes, that's right. There were many shortcoming in the AGA chipset that could have been sorted out with only very little extra effort. The slow chipmem bus is one of them. Z3 preceded AGA and had better bandwidth.

The Amiga has the elegant hardware 2D approach where almost no data needs to be pushed over the bus for scrolling. The PC had the stupid brute-force approach where for scrolling all the screen content had to be updated. However, there was a point in time when the stupid approach was just as fast as the elegant approach. Changing from Amiga to PC was perhaps made easier by the fact that the PC could not only do what the Amiga couldn't do but could now also do what the Amiga could do. No more "only Amiga makes it possible".

Of course. By 1996, the Amiga was long dead. I didn't want to admit that at the time, but in retrospect it's very clear. But to be clear here - as late as 1994 the PC was not quite there yet on the 2D front. Even in 1996 a PC was still a pretty poor option for people who liked those smooth scrolling games.

Are we trying a bit of that ad-hominem stuff here?

Nope. I was specifically targeting the strange twists your arguments sometimes seem to take. To be precise, I was trying to point out that you know that chunky vs planar had nothing to do with these 2D games performing better on PC, but still kind of tried to present it as such in the very segment I was replying to ('if only they had planar').

It has nothing to do with you as a person.

Yes, that's what I think. But I didn't pin it all on the lacking chunky mode. I said that Commodore should have steadily invested into R&D and should have raised the base configuration in a way that would have kept the entry-level price more or less constant rather than selling the same old technology for always less money.

The omission of a chunky mode was probably the biggest single fault and was my example to show how quickly put together AGA really was. They didn't even spend the time to add this tiny modification that any sane engineer (I think those were my words) would have made if s/he had been allowed to.

I know and I disagree. Perhaps this is another subject to just agree to disagree on. You know that my view is that in the real world the PC conquered the market regardless of specs of the competition and thus that it wouldn't have helped. Repeating that for ten more posts is not necessarily a good idea considering the weather

Well, I'm sure an unused bit in an already present register could have been found for that purpose. Then add say four transistors for the drivers that drive the input of the switches.

Wow, that little?

I'm learning all kinds of interesting things the past few posts.

Because the chipmem was too slow for a then wasteful 8bit chunky mode. Remember: 8 bit chunky only would have come (almost) free the moment they made an 8 bit planar mode, too. The OCS chipmem bus is already very slow with just 5 or 6 bit planes active. No way they could have wasted the bandwidth equivalent to 8 bits planar in that design. The CPU wouldn't have got any chipmem cycles at all which makes it rather pointless.

Well sure, but if you really wanted to you could design a 'byte per pixel' approach that doesn't use all bits. I mean, this wasn't done in practice, but it should've been possible. Kind of like how the C64 doesn't use all bits in it's colour ram but is still byte-accessible.

Or is that harder than I think it is?

There is a difference between a rough estimate (which you didn't find sufficient) and a more thoroughly made calculation. I'm not willing to think how I would have implemented some graphics modes that went into that paper, propably would have been cut by management anyway and that I personally consider superfluous.

No, I did actually find the rough estimate enough. I immediately accepted your figure and only asked for some clarification on the register use as this wasn't clear to me. The question I asked here was about something else.

See, you've said it would've been little effort to add these extra pixel formats, which seems to be at odds with what you're saying here.

Just so you understand: I'd be perfectly happy with 'it's not going to be much more than the chunky thing'. I don't need full figures.

The XBox 360 was released in May 2005. This probably cut into the sales of the PS2. It kept going pretty strong, though. There was healthy competition and both companies had to improve their products. They understood what Commodore did not understand.

Nonetheless the important point remains: big companies have actually put the next generation of a product out and continued to sell the preceding generation for many years with good success. Commodore had to sell the A600 under cost because nobody wanted the same old stuff. Commodore could have offered an AA+ Amiga in 1990/91 and still continued to sell the A500 with good success. This disproves your earlier point that no company would have done that.

No it doesn't. My point was not that no company would release two product at the same time - though I did wonder out loud about Commodore's ability to do this.

My point was that companies wait to release such new products until after the old product(s) have started to decline (for whatever reason). The figures you've posted back this point up. They do not disprove it at all.

Originally Posted by swinkamor12

Apple was worth less than Commodore in 1992 and now is richest company int thw World.

Apple in 1992: net sales of about 7 billion, profit of 530 million
Commodore in 1992: net sales of about 900 million, profit of 27 million


[ Show youtube player ]

Sources: https://dfarq.homeip.net/commodore-f...ory-1978-1994/
https://en.wikipedia.org/wiki/Histor...ancial_history

Even Atari live long enough to get to next level - console with 3D support.

Released in 1993, lasted until 1996 and sold fewer units than the A1200 did in 1992-1994 when Commodore was still around. Great success story there.

By the way, if you actually read up on Atari history - you'd have known that a big part of the reason Atari lasted so 'long' was that they had won a few lawsuits that awarded them lots of money. It had little to do with market success in the mid 1990s.

It is obvious for everyone outside Amiga asylum where some fanatics still does not accept reality.

Still going with the personal attacks eh?

[swinkamor12]

To sum it up.

Commodore bankrupt because Amiga 1200 was underpowered overpriced shit.

Amiga 1200 was crap because it has not:

- chunky pixels
- slots for fast ram
- simple mmu for protection of first memory page

Everything from list in 1992 was not rocket science and was even in cheapest pc.
Everything from list in 1992 should be in base model of Amiga - Amiga 1200 for price Commodore ask for Amiga 1200 in 1992.
Not in very expensive big Amigas.
Users of Amiga 1200 should not be forced to buy expensive hardware upgrade to get things from this list.
It was easy to make Amiga 1200 worth of use and Commodore has enough money to do it but they screw it.

[roondar]

Originally Posted by swinkamor12

To sum it up.

Commodore bankrupt because Amiga 1200 was underpowered overpriced shit.

Amiga 1200 was crap because it has not:

- chunky pixels
- slots for fast ram
- simple mmu for protection of first memory page

Everything from list in 1992 was not rocket science and was even in cheapest pc.
Everything from list in 1992 should be in base model of Amiga - Amiga 1200 for price Commodore ask for Amiga 1200 in 1992.
Not in very expensive big Amigas.
Users of Amiga 1200 should not be forced to buy expensive hardware upgrade to get things from this list.
It was easy to make Amiga 1200 worth of use and Commodore has enough money to do it but they screw it.

To sum it up: everything in the post above is wrong or misleading. All your points have been talked about before and I’m not going to bother showing why it’s all wrong yet again.

[Daedalus]

Originally Posted by roondar

To sum it up: everything in the post above is wrong or misleading. All your points have been talked about before and I’m not going to bother showing why it’s all wrong yet again.

Everything in most of his posts is wrong or misleading. Which is why a lot of them are simply ignored. Just seems to be either deluded, or a bored troll with nothing better to do.

[Glen M]

Having glanced through this thread its funny how a fair amount of it seems to revolve around doom. You don't see too many PC forums complaining that their 386 PC couldn't run (insert any amount of 2D Amiga games here) as good as our machines.

I might write a write a letter to Hyundai and complain that my 2lt coupe that cost £20k new isn't as fast as a £100k Porsche.

If you want to know why commodore failed its because it was ran by muppets then bought by donkeys and eventually sold off to arseholes.

Or then again I might just get over it and use what I have for what it is and what it can do.

[grond]

Originally Posted by roondar

A hypothetical AA+ machine with more bandwidth and chunky will run Doom better. I have no doubt about that. However, I'm also not interested in it as we don't have such a machine. I've tried to point out that I was more interested in machines around the release of Doom and the actual performance we did end up getting. Perhaps I've failed there, but in my mind we were not discussing the AA+ machine for the last five or six posts on the subject of Doom.

From my point of view all our discussion was about how in detail Commodore presented too little of an upgrade with the A1200. Looking at the details required looking at what they could have and should have done instead and what technical and ecnomical effects this would have had.

Anyway, there were plenty of really nice 3D games on the Amiga by the time of Doom's release. Many of which ran like a dog without upgrades. Yet no one upgraded. On the flip side there were official ports to platforms that were accustomed to better hardware and upgrading and as far as I know, this did not cause an increase in people upgrading there.

I'd be interested in examples just out of historical interest. The only 3D games on the Amiga I ever liked were Alien Breed 3D (the first part, the second was unplayable) and Ambermoon.

I'm not sure I understand. You don't need to cookie cut anything for dual layer graphics. Perhaps you're referring to blitting stuff onto the buffers? That doesn't change bandwidth costs - the Blitter always fetches the mask for every word it cookie-cuts. With or without interleaving.

Unless you are merely copying tiles together and only using sprites for moving objects, you will have to use quite a few cookie-cut BOBs for a game such as Warcraft. With interleaved bitmaps the mask needs to have the same size as all pixel data, with non-interleaved bitmaps you can reuse the mask for all bitplanes. Bandwidth remains the same, but memory impact is less for non-interleaved bitmaps. BTW, such cookie-cut BOBs also require reading the entire background which means a bandwidth penalty when compared to chunky. Chunky graphics can save memory bandwidth here because you can just skip writing to the graphics buffer if you have read a transparent (i.e. zero value) pixel and thus save bandwidth while writing and all bandwidth for reading the background image.

That was where you added special hardware to display and deal with two layers to the chunky system.

All chunky is hypothetical when talking about Amigas, isn't it? Whenever we were discussing the pros and cons of chunky vs. planar it is only relevant for my point that AGA clearly is a job half done and that a typical engineering team would not have been satisfied with the result. They would have wanted to add chunky the moment they stepped up to 8 bit palette depth. It's a natural choice.

My whole point here was that chunky 3D, while impressive, did not actually conquer the market.

And I disagree. All the Amiga demo and coding scene was about from 1993 to 1996 was to implement Doom-type engines. Perhaps it was a mass psychosis but the effect Doom had on young people at that time was immense.

You know that my view is that in the real world the PC conquered the market regardless of specs of the competition and thus that it wouldn't have helped.

Yes, you are probably right. Looking at all the technical shortcomings of the PC it becomes clear that the technology itself wasn't the decisive factor. At least not until the PC technology was so powerful that it could do great games. Until then it was probably just about the software that a PC could run and that somehow had become standard in offices around the world.

However, in my time as a great hindsight-manager I would also have had Commodore invest in software, porting and developer support. And thus Commodore also would have been economically more powerful than Apple...

Well sure, but if you really wanted to you could design a 'byte per pixel' approach that doesn't use all bits.

I'm not sure I understand. You can always leave bits unused in each byte but you would have to transmit the meaningless bits anyway during DMA.

See, you've said it would've been little effort to add these extra pixel formats, which seems to be at odds with what you're saying here.

Well, rearranging pixel data is surely a small effort. Of course, the meaning of "small" just depends on what you are comparing it to. If support for those pixel formats could be designed in the time it takes to write a forum comment, then they would have done it. If you compare to the effort for a hardware polygon rasterizer, the effort becomes invisible.

My point was that companies wait to release such new products until after the old product(s) have started to decline (for whatever reason). The figures you've posted back this point up. They do not disprove it at all.

I'm not sure about this. There were two peak years and then a long plateau phase at a very high level. The end of the peak seems to be rather related to the Xbox 360 eating away a chunk of the market. The next generation was due because Sony did not disregard the competition as Commodore did but not so much because they saw their current product become a shelf warmer starting next year.

BTW, I'm pretty sure that both Sony and Microsoft start developing their next generation of consoles the moment they have the recent one out of the door. If they didn't, they would have to lay off half of their engineering department for a few years (the other half would make cost-reduced slim versions of the current model). The Commodore R&D team was also always busy but unfortunately with a lot of braindead products (264, 128 and 65 as most prominent examples) when they should have concentrated on the Amiga.

[activist]

Originally Posted by Glen M

Having glanced through this thread its funny how a fair amount of it seems to revolve around doom. You don't see too many PC forums complaining that their 386 PC couldn't run (insert any amount of 2D Amiga games here) as good as our machines.

watershed game at the time which altered the landscape. No surprise it get's mentioned a lot

Originally Posted by Glen M

I might write a write a letter to Hyundai and complain that my 2lt coupe that cost £20k new isn't as fast as a £100k Porsche.

yes write that letter. Complain that if they had only added slightly better exhaust header and camshaft you'd have much more of a chance with the Porsche when the ecu software was improved down the road

Originally Posted by Glen M

If you want to know why commodore failed its because it was ran by muppets then bought by donkeys and eventually sold off to arseholes.

yep, that's clear, understood

Originally Posted by Glen M

Or then again I might just get over it and use what I have for what it is and what it can do.

The base A1200 was/is only ok at 2D. 3D no.
But the point that with just a few cheap tweaks the 1200 would have been 3D capable from the off is a point well worth making. This thread is convincing that the tweaks to the 1200 would have been both cheap and easily done. A better version of doom would have arrived much earlier on much more basic A1200. To use your car analogy chunky pixel mode, full speed bus would have paved a gravel road for your soon arriving Porsche and Lambo. Lot's of people bought that fast 486 DX 66 JUST to play Doom and nothing else. Who's to say they wouldn't have bought that 1200 instead

[Glen M]

Originally Posted by activist

The base A1200 was/is only ok at 2D. 3D no.
But the point that with just a few cheap tweaks the 1200 would have been 3D capable from the off is a point well worth making. This thread has been very convincing (primarily the enjoyabe grond/roondar debate & banter) that the tweaks to the 1200 would have been both cheap and easily done. A better version of doom would have arrived much earlier on much more basic A1200. To use your car analogy chunky pixel mode, full speed bus would have paved a gravel road for your soon arriving Porsche and Lambo. Lot's of people bought that fast 486 DX 66 JUST to play Doom and nothing else. Who's to say they wouldn't have bought that 1200 instead

I'll say they wouldn't because Doom is a PC game, was developed for PC from a PC software house building on their previous game.

It's easy to say it should of had this, that or the other but the A1200 is what it is so why not let sleeping dogs lie.

Yeah it would of been cool if it had all the features you talk about but would that of hurt backwards compatibility? I don't know as I'm not that technically minded. All I can see is a wish list that may seem cheap but do bear in mind on a production run it's down to pennies and since the muppets where only worried about that then that was the driving factor (car pun??)

Now if the donkeys had of released an A1200+ and made such upgrades that would of been a good move but instead the blew the budget on a new sticker.

I have enjoyed reading this thread so don't let me get in the way of the discussion and banter. Just throwing my oar in as you do but part of what I'm saying is why worry about something so long gone, just get on with what you have and enjoy it. Maybe I'm getting old.

Edit: can I just ask when you say the base 1200 is only OK at 2D are you basing that on hardware limitations or just from looking at the software?

I ask because some AGA games are beautiful to look at.

[AmigaHope]

I remember the big watershed games at the time being Wolf3D+Wing Commander, and then Doom+Tie Fighter. Chunky pixels would have helped a LOT with an A1200 version of Wolf3D and Wing Commander as writing the pixels actually made up a significant amount of overhead for those otherwise low-CPU games. Doom and Tie Fighter would have required a lot more CPU power to go along with the chunky pixel mode.

On later games chunky pixels aren't even really a big deal as the overall CPU overhead in the game logic makes the C2P process a relatively minor portion of the work involved, the slow chip ram bus actually becoming a bigger bottleneck. i.e. if you took a game that needed a Pentium/133 or a 68060 to run properly, it could render the planar graphics just fine but suffer when you tried to transfer it to slow-ass AGA bus.

That's why later games really needed a VESA Local Bus/MCA/EISA VGA card to run properly and chugged along on a slow ISA card (and why some low-end motherboards supported double-clock-rate ISA even though it broke the standard)

[malko]

From a player's point of view concerning some "textured" 3D games of the time, I have trouble understanding all this hay around Doom. It seems really overrated to me...

For my part, if Doom has had such an impact on people's minds, it's mainly because of the possibility of playing it networked (it was fun, that's clear!) but in any case not because of its game engine. Of course I do not deny that the Amiga could not race with such type of game, however, saying that this type of games (and identical) are what killed the Amiga is very, very exaggerated.

I remember (for having played a lot) that more than a year and a half before Doom, "Ultima Underworld" already allowed you to move in the same way (and the game was much more interesting : dramatisation, character presentation, introduction, riddles, etc.).

Bref...

Some textured 3D games (add those you think about in this list) :
----------------------
(First Person)
 03.1992 : Ultima Underworld (
				[ Show youtube player ])
 05.1992 : Wolfenstein 3D (
				[ Show youtube player ])
 12.1993 : Doom (
				[ Show youtube player ])
 01.1996 : Duke Nukem 3D (
				[ Show youtube player ])
 10.1996 : Tomb Raider (
				[ Show youtube player ])
 etc.
 
(In space) 
 09.1990 : Wing Commander (
				[ Show youtube player ])
           1992 : A500 version (
				[ Show youtube player ])
           1994 : CD32 version (
				[ Show youtube player ])
 03.1995 : Descent (
				[ Show youtube player ])
 etc.