Was anyone else disappointed with the A1200?

[grond]

Originally Posted by swinkamor12

Just chunky pixel, slots for fast ram, simple mmu for 020.

I don't see any need for the fast ram slots. As I already said, Commodore just shouldn't have put a CPU on the main PCB but offered a 020 and an 030 CPU card with fastmem to choose from. CPU cards just like all those accelerators we all have and that still could have provided more sophisticated alternatives to Commodore's hypothetical offerings. In the end we all paid for the 020 we never or only hardly used...

[Retro1234]

It's like a car crash - I can't look away.

[mcgeezer]

Originally Posted by roondar

Yeah, right

They were nowhere near that cheap and you know it. FYI, I did check prices and even the 286 was still more expensive than that at the time. The 386SX started closer to 600 pounds. That's without a harddisk, with 1/2 the memory of the A1200 and with a mono monitor (which is just 'brilliant' for games).
Claiming UU runs well when there's a video linked in this thread that shows it actually doesn't isn't very clever. UU is also not a wolf clone, which is why it doesn't run so well on a 386. Also: affordable SVGA cards were not fast and this was in fact a known problem. I already pointed this out several times.
And you apparently just can't stop with the thinly veiled personal attacks and posts filled with sheer nonsense.

Roondar.... I honestly don't know why you give this guy your quality time in typing out proper responses, he's clearly a jelly head.

You should just write ....
The lady doth protest too much, methink

[Retro1234]

Poor Damien every other mod has done a runner, get your pacifier ready.

[mcgeezer]

Originally Posted by Retro1234

Poor Damien every other mod has done a runner, get your pacifier ready.

Damien isn't a mod.

[grond]

Originally Posted by grond

the important point is that you can blit chunky modes with a planar blitter without problems because actually there is no additional functionality required (not considering line drawing, of course). Only the line drawing mode would have required a hardware update of the blitter in order to make it deal with chunky graphics.

I'm probably too tired to think clearly but it just occured to me that it would probably only have taken a very small fix to make the blitter able to deal with chunky modes: doesn't the blitter allow patterns for the lines it draws? In a chunky mode this pattern would automatically become the line colour and all it would take would be to make the blitter's bresenham use only multiples of 8 for its line coordinates. Other than that the blitter would also draw as if in planar mode with diffX just being x8 as compared to the planar mode.

[roondar]

Originally Posted by grond

We can't know but that is what I think. I think the added features would have made it clear to the audience that the higher price was indeed giving them more than the A500 did and thus would have created a stronger desire to upgrade. There never was much reason to buy an A500+ or A600 when you already had an A500. Commodore could sell these models only to customers who had not already bought an Amiga from them. If they had upgraded the base configuration of the Amiga earlier, they could have sold more Amigas to the very same customers that already liked their computers.

Well, I think differently - but I can see what you're getting at. I guess we'll never know.

When you already had an originally more expensive A500 bought in 1987 the fact that the A1200 was a cheaper machine and thus only a little more capable than the A500 (because technology had evolved for five years which mostly went into making the A1200 cheaper and not so much into making it more powerful) also made it a technically too small a step to upgrade and not switch to something completely different.

I honestly don't get this entire argument - yes, it would've been nice to have gotten more and there is a good argument to be made that AGA was underwhelming compared to what we were all hoping for, but the A1200 really was quite a bit more powerful than the A500.

Let's compare them:

• CPU: 2x as fast (4x with fast memory)
• RAM: 4x as much, 2x the speed for the CPU/4x for display (meaning that 640x512x256 is quite a bit faster than 640x512x16 ever was)
• Graphics: 4x the colours (technically - but only if you count EHB as a 'usuable' mode for games, otherwise it's 8x). Also: DPF and Sprites were much better and for people who didn't have ECS there were now flicker free hires modes when using the correct monitor
• Blitter: speed went up by a factor of around 1,5 due to lower bus congestion

IMHO, that is not 'a little more'. The A1200 ran rings around the A500 speed wise and could do all sorts of things the A500 simply couldn't.

I'm not saying it's the best ever computer or that you can't have wanted more and I understand why some are disappointed with it. Nor am I claiming that Commodore couldn't have done a better job with the time they had (as they certainly could've!). But the notion that the A1200 is effectively the same machine as the A500 with some tiny changes is just not true in a practical sense.

If the A500 sold for 500UKP in 1987, what would an A1200 in 1992 have been like that had justified a starting price of 500 UKP (or 600UKP accounting for devaluation of the Pound and inflation in general)? Certainly more exciting than the A1200.

Quite possibly, but then again - there was the 600GPB Atari Falcon and no one bought that. Even though it has many of the features you're talking about here and could do 3D chunky games fairly well.

It's machines like the Falcon and the later Archimedes models (which were more expensive, but also (much) more powerful than the A1200) failing as well that has me so convinced it wouldn't have changed anything.

The cheap Amiga models canibalised the expensive ones. Apple was doing pretty well with just expensive Macintosh models. But they had the software and good developer support. Commodore didn't spend a penny on that.

Quite, but that's essentially part of the point I was trying to make. Commodore always mostly made money targeting the low end. I just don't see it. There were a few competitors in the home computer market that did have more powerful, yet more expensive alternatives and they didn't sell either.

Apple is... Interesting. I'm not sure I'd call what they did in the 1990's a sign of them doing well though. They only just made it.

Anyway, I think you and I have a different opinion here and that's fine. I can see your side of the argument, hopefully you can see mine.

Yes, probably, but a speed increase by a factor of 3 doesn't look too bad when comparing to a console's hardware that does a factor of 4.

I'll grant you that, but I'm sure it would still have been disappointing to people who bought one to see their new Amiga struggle to keep up in number of objects and colours on screen (and that's not counting all the nifty SNES effects).

AmigaOS does a lot of line drawing with the blitter which means the blitter has to draw a line in eight bitplanes per vertical screen line where it would only have to do it in one byte per vertical screen line (OK, this is only true for 45°...0° with respect to a vertical line while horizontal lines would almost be equal to chunky modes in efficiency).

It also does a lot of block based blitting, which is what I was getting at.

And of course - if you want to, you can remove most of the line drawing as windows and window decoration could be done as little blitter objects instead (which is how Windows draws them IIRC). As nice as Amiga OS is, I've never been convinced it was truly optimised for drawing speed. Then again, neither was Windows.

As an example: I remember having a 'toy program' that let you move windows across the screen fully drawn on the Amiga and that ran surprisingly well compared to the normal OS window drawing and updating.

Of course, one could argue that a blitter would have been superfluous with a chunky mode because the CPU was much more suitable for drawing in chunky modes than in planar modes. But the important point is that you can blit chunky modes with a planar blitter without problems because actually there is no additional functionality required (not considering line drawing, of course). Only the line drawing mode would have required a hardware update of the blitter in order to make it deal with chunky graphics.

Well, some more upmarket SVGA cards came with Blitters and these apparently made Windows run quite a bit faster so I guess they can still be useful for a chunky mode system.

Originally Posted by grond

I'm probably too tired to think clearly but it just occured to me that it would probably only have taken a very small fix to make the blitter able to deal with chunky modes: doesn't the blitter allow patterns for the lines it draws? In a chunky mode this pattern would automatically become the line colour and all it would take would be to make the blitter's bresenham use only multiples of 8 for its line coordinates. Other than that the blitter would also draw as if in planar mode with diffX just being x8 as compared to the planar mode.

Yes, the Blitter does allow for a 16 pixel pattern of your choosing when drawing lines (or filling or clearing for that matter).

Originally Posted by mcgeezer

Roondar.... I honestly don't know why you give this guy your quality time in typing out proper responses, he's clearly a jelly head.

You should just write ....
The lady doth protest too much, methink

You're right, I really shouldn't reply. But... here's why I still do reply to things like this (though probably far too frequently):

It's not about the person or even their post. It's not even about the tone of their message. It's about how forums work. I know only too well that people tend to only read and remember the last few pages of a long thread. This is why longer threads often end up repeating the same arguments over and over and it's also why pushing back to falsehoods is needed (regardless of why these falsehoods persist - I'm not saying people in this thread are doing this on purpose. I'm assuming people just misremember).

If no one pushes back when someone writes things that are not true, new readers might read these such posts and get the idea that they make sense. There's already so much misinformation out there in retro land (heck, despite trying very hard, I've certainly made mistakes and wrote things that ended up being false so I'm not excepting myself here).

If I can keep at least some accuracy in threads like this, I'll consider that a success.

Worst part about this is that all this forum activity comes straight out of my Amiga coding time. So perhaps I should spend less time on the forum after all.

[vulture]

Without a better cpu and some fast ram, I doubt a chunky mode would have any real importance at having a Doom port. I mean, if a 68030/50mhz+AGA runs it at about the same fps as a 386DX40mhz+VGA, there's not much a chunky mode by itself could've done.

[sandruzzo]

If A1200 was in all part 4x A500, who cares about Chunky mode!

[Foebane]

Originally Posted by sandruzzo

If A1200 was in all part 4x A500, who cares about Chunky mode!

Because Chunky mode would be faster, as it is on PC. Have a look at this article which describes why Doom just wouldn't work on a vanilla A1200.

[grond]

Originally Posted by roondar

I honestly don't get this entire argument - yes, it would've been nice to have gotten more

Not only get more but also to be allowed to give them more of our money for a better product.

but the A1200 really was quite a bit more powerful than the A500.



[...]



IMHO, that is not 'a little more'. The A1200 ran rings around the A500 speed wise and could do all sorts of things the A500 simply couldn't.

Yes, I admit I downplayed the A1200 a little for the sake of my argument. BTW, I was a poor kid and bought a C=64 with datasette when the neighbourhood kids got their A500s. My first Amiga was an A600 which I bought new for 299 DM (!!!) in 1993 because it was so cheap and I thought that I had always wanted one, so why not? It didn't live much more than a year and then was replaced by a used A1200 which I immediately upgraded with a 1230/882 @ 50 MHz, 8 MB of fastmem, 330 MB 2.5" hdd and an internal HD floppy drive. (As you can tell, I had more money by then... ). Anyway, I certainly wasn't disappointed with the A1200, I didn't even notice its existence until quite a while later. And I still think it is the best of the bunch as the big box Amigas (that certainly were more capable) didn't do it for me. But just because I (as many of us) liked that computer it doesn't make it the right market strategy.

It's machines like the Falcon and the later Archimedes models (which were more expensive, but also (much) more powerful than the A1200) failing as well that has me so convinced it wouldn't have changed anything.

I really can't say much about Ataris or Archimedes because I lack knowledge about them and their market impacts. I think Commodore was in a much better position than these companies. Commodore once was so mighty that they bought MOS, the company that produced the CPUs that Apple used up to the first Macintosh that preceded the Amiga by only very little. The last thing that Commodore did right was to buy the Amiga. And it may have been better for the Amiga if this had not happened.

Apple is... Interesting. I'm not sure I'd call what they did in the 1990's a sign of them doing well though. They only just made it.

Yes, but they made it and nobody else. They made it because they did a few things right between 1984 and 1994...

Anyway, I think you and I have a different opinion here and that's fine. I can see your side of the argument, hopefully you can see mine.

I sure can and I have enjoyed this discussion. The Amiga kindled my interest in assembly programming, then microprocessors and eventually microelectronics. This made me become an electronics engineer and microchip developer. The young man in me just refuses to accept that it all was in vain. But the rest of me with the greying hair can see your points just fine...

As nice as Amiga OS is, I've never been convinced it was truly optimised for drawing speed.

Oh, it definitely is! Without the OS efficiently offloading a lot of jobs to the blitter it wouldn't have been possible to have this sophisticated OS and GUI running on such puny hardware.

As an example: I remember having a 'toy program' that let you move windows across the screen fully drawn on the Amiga and that ran surprisingly well compared to the normal OS window drawing and updating.

Yes, but the OS always targeted a 7 MHz 68000/OCS and when you had that "toy program" it was probably running on a better Amiga. They never even bothered to compile AmigaOS kickroms for AGA computers for 68020 which was the lowest CPU grade there have ever been in AGA Amigas. All kickroms are plain 68000 code.

[roondar]

Originally Posted by sandruzzo

If A1200 was in all part 4x A500, who cares about Chunky mode!

Originally Posted by Foebane

Because Chunky mode would be faster, as it is on PC. Have a look at this article which describes why Doom just wouldn't work on a vanilla A1200.

It's 100% true that Doom would not run on a vanilla A1200. And even if it somehow fit in memory and we'd have been crazy enough to run the game, it would still have run worse than it does on a 386SX and that ran Doom very, very poorly.

But there's two things to add here:

1. If you put a CPU in the A1200 that is about as fast as the recommended PC hardware for Doom and use a good c2p routine, it actually runs rather well. Now that does mean using a 68040/40Mhz accelerator or better and the equivalent PC will still likely run it at a higher frame rate, but it can be done.
2. If we take the base A1200 and give it a chunky mode and 'unlimited bandwidth' to chip memory, the 68020 in the A1200 would still not be able to do Doom any better than a 386 of the same speed. Doom requires a lot of CPU grunt and the actual chunky display pixel writing Doom does is only a small part of the overall workload. Doom spends quite some time calculating the values of the pixels it draws, and this would not be sped up at all by using a faster Blitter or a fast chunky mode.

For Doom to run well, you want fast access to graphics memory in a chunky format (which is the problem everyone acknowledges on Amiga forums and was a big problem on the Amiga due to the extra step) and a fast CPU to calculate what pixel needs to be what colour and run the game (which is what many people seem to ignore). It's that last thing that makes Doom have a recommended system that is a fast 486, as a 25MHz 386 can quite happily push out over 20MB/sec. Which is more than enough to do 60FPS 256 colour graphics in 320x200. And yes, I do know the 386 would've been limited by the video bus - the point I'm making is that it's not writing the pixels that was the problem here.

So, you'd still need a fast CPU for Doom even if we had AAA. I doubt that Commodore would've released an A1200 style computer with a 68040 (or even a high end 68030) even if they did have AAA.

But I do agree to a point: if the A1200 had more memory bandwidth to chip memory and a chunky mode - it would've been easier to get it to run Doom as long as you installed a proper turbo card with a 68040+ on it. And who knows, maybe such an 'A1200' would've done better in the marketplace and give us Doom. But that would've meant convincing the Amiga crowd to pay for those turbo cards and not just get the SNES version instead.

Originally Posted by grond

Not only get more but also to be allowed to give them more of our money for a better product.

I understand your point and you know my position so I'll leave it there

Yes, I admit I downplayed the A1200 a little for the sake of my argument. BTW, I was a poor kid and bought a C=64 with datasette when the neighbourhood kids got their A500s. My first Amiga was an A600 which I bought new for 299 DM (!!!) in 1993 because it was so cheap and I thought that I had always wanted one, so why not? It didn't live much more than a year and then was replaced by a used A1200 which I immediately upgraded with a 1230/882 @ 50 MHz, 8 MB of fastmem, 330 MB 2.5" hdd and an internal HD floppy drive. (As you can tell, I had more money by then... ). Anyway, I certainly wasn't disappointed with the A1200, I didn't even notice its existence until quite a while later. And I still think it is the best of the bunch as the big box Amigas (that certainly were more capable) didn't do it for me. But just because I (as many of us) liked that computer it doesn't make it the right market strategy.

Oh, I definitely agree there were problems with the AGA product line.

For a start, as I've always understood it, AGA was essentially born of a 'panic reaction' by Commodore. At the time, AAA was not yet anywhere near ready and the A500 had been selling really well*. Then 'out of the blue'**, somewhere in 1992 A500 sales suddenly dropped dropped through the floor and Commodore was in trouble. So they decided to very rapidly get AGA out the door to stave of said trouble. Given how little time they apparently spend on it, it's a miracle it turned out as well as it did.

The key take away here is that Commodore had waited too long and now their product was not meeting the demands of their potential customers any more. The AGA machines were a pretty big improvement, but also still had some of OCS/ECS's limitations right in the chipset which turned out to be problematic moving into 1993 and onwards (such as low CPU->Chip Memory bandwidth limits and no chunky graphics).

*) as in: sales of the entry level Amiga (A500/later A500+ & A600) were up year on year with double digit percentage growth (or more, one year saw their sales doubling IIRC) every single year since the A500 was on sale. Until mid-1992. Then it suddenly sharply declined.
**) to be precise, a shade of blue known as 'Segatendo 16 bit' or was it 'Nintenega 16 bit'?

I really can't say much about Ataris or Archimedes because I lack knowledge about them and their market impacts. I think Commodore was in a much better position than these companies. Commodore once was so mighty that they bought MOS, the company that produced the CPUs that Apple used up to the first Macintosh that preceded the Amiga by only very little. The last thing that Commodore did right was to buy the Amiga. And it may have been better for the Amiga if this had not happened.

Market impact of the Falcon was very poor, the Archimedes did better in this regard but was still a small player - Acorn actually soldiered on for quite a while with the Archimedes. Specs wise, both the Falcon and the later models of the Archimedes offered quite a few of the things named in this thread on the 'A1200 wishlist'. The Falcon had a chunky mode and the Archimedes had that, a very fast CPU (not in MHz, but it was a true RISC processor and punched above it's weight category as a result) and a really high overall system bandwidth (something on the order of 48MB/sec in 1992). Both also offered at least 256 colours and the Archimedes got an official Doom port (albeit only in 1998 ). IMHO, the Archimedes has been underrated in retro land due to it's obscurity outside of the UK and it being squarely aimed at the education market at the time.

As for Commodore buying MOS and Amiga, they did indeed do that. But they also did both for much less money than you might have guessed. As for the Amiga, the only other alternative was Atari buying them and they wanted to make it into a simple console. IIRC, R.J. Mical one said he felt that Commodore had done a lot of good and that it would've been much worse if Atari had won.

Yes, but they made it and nobody else. They made it because they did a few things right between 1984 and 1994...

As I understand it, they got bailed out by Microsoft in 1997 and would've gone bust otherwise (https://www.wired.com/2009/08/dayintech-0806/) - apologies in advance for the ad-heavy page.

I sure can and I have enjoyed this discussion. The Amiga kindled my interest in assembly programming, then microprocessors and eventually microelectronics. This made me become an electronics engineer and microchip developer. The young man in me just refuses to accept that it all was in vain. But the rest of me with the greying hair can see your points just fine...

Like you, I didn't have much money. I started out in 1988 (I think, may have been 1989) with a C64 and Datasette. Upgraded to an A500 later after saving up an entire year for it and then on to the A1200 (later with my beloved Blizzard 1230MKIV). The Amiga essentially got me into studying computer science. I absolutely loved my A1200 at the time and wanted to learn how to properly program in large part because of it (and my A500 and C64 before it). Without Commodore, I'd have had a very different career path.

Oh, it definitely is! Without the OS efficiently offloading a lot of jobs to the blitter it wouldn't have been possible to have this sophisticated OS and GUI running on such puny hardware.

I'm not saying Amiga OS is inefficient, merely that drawing lines is a pretty big performance hit on the Amiga even with the Blitter accelerating them (8 cycles per pixel per plane is pretty painful). It seems to me like something you could mostly avoid, instead getting by with simpler operations.

Yes, but the OS always targeted a 7 MHz 68000/OCS and when you had that "toy program" it was probably running on a better Amiga. They never even bothered to compile AmigaOS kickroms for AGA computers for 68020 which was the lowest CPU grade there have ever been in AGA Amigas. All kickroms are plain 68000 code.

It was running on my A1200, indeed.

As for the Kickstart, are you absolutely sure? I seem to recall WinUAE refusing to load the A1200/A4000/A3000 3.0+ Kickstarts on an A500/A600/A2000 environment with a message that it requires a 68020 to run these Kickstarts.

[Daedalus]

Originally Posted by roondar

As for the Kickstart, are you absolutely sure? I seem to recall WinUAE refusing to load the A1200/A4000/A3000 3.0+ Kickstarts on an A500/A600/A2000 environment with a message that it requires a 68020 to run these Kickstarts.

I'm pretty sure it was all 68000 code, but perhaps the issue with WinUAE is simply a check it does when you're using machine-specific environments? Have you tried using a non-specific setup with a 68000 and a 32-bit ROM? Or maybe it's hardware initialisation stuff... Anyway, I expect the Kickstart modules were compiled with backwards compatibility in mind, so the same modules can be used to upgrade previous, non-AGA machines.

3.5 and 3.9 were the only ones actually compiled with 68020 optimisations AFAIK.

[Toni Wilen]

At least A1200 3.1 ROM crashes if not 68020+. It executes DIVL.L. I guess 3.1 utility.library is 68020+ only.

3.0 works with 68000.

[grond]

Originally Posted by roondar

If you put a CPU in the A1200 that is about as fast as the recommended PC hardware for Doom and use a good c2p routine, it actually runs rather well. Now that does mean using a 68040/40Mhz accelerator or better and the equivalent PC will still likely run it at a higher frame rate, but it can be done.

For 030-class CPUs you'll lose something like 20 to 30% to c2p at low framerates. The loss will be higher for increasing framerates. E.g. if you run Quake at 1fps the added effort for c2p is negligible. If you want to run Wolfenstein at 40fps, well, you can't.

Having to deal with c2p is not very attractive to a coder. It has been a sport for Amigans to find fast c2p routines and, as already stated, it took several years to arrive at the solutions we have had since the late 1990s. None of them make the planar disadvantage disappear. It is a nuisance.

[*]If we take the base A1200 and give it a chunky mode and 'unlimited bandwidth' to chip memory, the 68020 in the A1200 would still not be able to do Doom any better than a 386 of the same speed.

That's why Commodore should have offered 020 and 030 CPU cards right from the start just like they had the 030 and 040 for the A4000. And looking at how Ambermoon runs OKish (for an RPG) with floor and ceiling textures on the unexpanded A1200, I think a chunky mode would have meant a considerably larger 3D window.

Doom requires a lot of CPU grunt and the actual chunky display pixel writing Doom does is only a small part of the overall workload. Doom spends quite some time calculating the values of the pixels it draws, and this would not be sped up at all by using a faster Blitter or a fast chunky mode.

I tend to disagree. One 320x200 frame are 64KB. At 30 fps you would spend about 30% of the total available CPU time just for copying the data to chipmem. Even a fast 030 cannot do 8bit c2p at copy speed, so the percentage becomes even larger. If, on the other hand, chipmem had been as fast as normal RAM of the time minus the screen DMA, you would simply have drawn the buffer in chipmem, updated the DMA pointers and be done with the frame.

Furthermore, Doom really doesn't need much CPU grunt, it needs memory bandwidth above all. All the renderer does is to collect pixels from textures and put them in their place. Since the binary space partition tree algorithm is so good, it hardly ever needs to draw a pixel twice. But it needs to set each and every pixel once per frame. The added stuff is a DIV per pixel column on the walls and ceilings. That's not really computing intensive.

Doom as long as you installed a proper turbo card with a 68040+ on it.

Doom can run well enough on an 030.

For a start, as I've always understood it, AGA was essentially born of a 'panic reaction' by Commodore. At the time, AAA was not yet anywhere near ready and the A500 had been selling really well*. Then 'out of the blue'**, somewhere in 1992 A500 sales suddenly dropped dropped through the floor and Commodore was in trouble. So they decided to very rapidly get AGA out the door to stave of said trouble.

That's not how I remember it. Originally the AGA features were already planned for the A3000 but Commodore pushed the A3000 out of the door before it was finished giving us "ECS". Then nothing happened for another few years...

The A600 was a total desaster and actually resulted in heavy losses. It came in 1991. When it was demonstrated inside Commodore, the staff went "wtf, even more of the same old, same old?"

Commodore even made it for another three years after that. AGA+chunky should have come with the A3000 in 1989, then things might have gone differently. Commodore wasted all that time and development budget on the stupid 8bit computers and then couldn't compete when there were better machines on the market. Those machines didn't come out of nowhere, they had been in development for some years, too...

The key take away here is that Commodore had waited too long and now their product was not meeting the demands of their potential customers any more.

Yes. The question is how much too long they had waited. They didn't understand that the evolving computer business absolutely required them to start developing the next big thing the very moment they had their newest product out the door.

IMHO, the Archimedes has been underrated in retro land due to it's obscurity outside of the UK and it being squarely aimed at the education market at the time.

To me the Archimedes always was a mystery machine that was supposed to be vastly superior to everything else but that nobody had ever seen. It was a unicorn.

As I understand it, they [Apple] got bailed out by Microsoft in 1997 and would've gone bust otherwise

Yes, because they were the biggest of the remaining bunch and Microsoft needed a living competitor to avoid monopoly related troubles. Commodore could have been the biggest remainder. They once were the No. 2 in the computer business behind IBM.

As for the Kickstart, are you absolutely sure? I seem to recall WinUAE refusing to load the A1200/A4000/A3000 3.0+ Kickstarts on an A500/A600/A2000 environment with a message that it requires a 68020 to run these Kickstarts.

I remember reading a discussion about this. Thomas Richter knows for sure. I seem to remember that he said something along the lines that compiling for 020 would only give a few percent more speed but would increase the maintenance and testing effort.

EDIT: I have just seen what Toni Wilen has written. I guess we can conclude that at least parts of the available 3.0+ kickstart versions are compiled for plain 68000.

[Daedalus]

Originally Posted by Toni Wilen

At least A1200 3.1 ROM crashes if not 68020+. It executes DIVL.L. I guess 3.1 utility.library is 68020+ only.

3.0 works with 68000.

Interesting... So there must be a different version specially compiled for the 68000 versions of 3.1? The V37 utility.library is supposed to detect the CPU and use appropriate instructions for 68000 and 68020+ setups, so it seems an odd decision to drop that functionality.

[roondar]

Originally Posted by grond

For 030-class CPUs you'll lose something like 20 to 30% to c2p at low framerates. The loss will be higher for increasing framerates. E.g. if you run Quake at 1fps the added effort for c2p is negligible. If you want to run Wolfenstein at 40fps, well, you can't.

Surely higher frame rates do not make c2p relatively more expensive? They should make drawing graphics more expensive at about the same rate (irrespective of chunky or planar). The limits of c2p on the Amiga have a lot to do with chip memory bandwidth and not with frame rate per se: there is only a maximum of about 7mb/sec available to the CPU (in a best case scenario) no matter how much or little GFX you wish to draw.

Having to deal with c2p is not very attractive to a coder. It has been a sport for Amigans to find fast c2p routines and, as already stated, it took several years to arrive at the solutions we have had since the late 1990s. None of them make the planar disadvantage disappear. It is a nuisance.

I never said it wasn't a nuisance, I said you could still run Doom on an Amiga with a fast CPU with reasonable results even with c2p. That's not actually the same thing as there not being at a disadvantage. The point here wasn't that it was no problem, but that it also had to do with other things than just the AGA chipset - merely changing AGA in the A1200 into a fast chunky display would not have given the vanilla A1200 Doom.

That's why Commodore should have offered 020 and 030 CPU cards right from the start just like they had the 030 and 040 for the A4000. And looking at how Ambermoon runs OKish (for an RPG) with floor and ceiling textures on the unexpanded A1200, I think a chunky mode would have meant a considerably larger 3D window.

Sure. I never said there would be no advantage to chunky. I was only talking about Doom on a base A1200.

I tend to disagree. One 320x200 frame are 64KB. At 30 fps you would spend about 30% of the total available CPU time just for copying the data to chipmem. Even a fast 030 cannot do 8bit c2p at copy speed, so the percentage becomes even larger. If, on the other hand, chipmem had been as fast as normal RAM of the time minus the screen DMA, you would simply have drawn the buffer in chipmem, updated the DMA pointers and be done with the frame.

A genuine question: if what you say about the game not actually needing CPU grunt but rather memory bandwidth is true, then why is spending 50 or so % of the available CPU time drawing pixels a problem? You clearly have memory bandwidth to spare at that point?

Furthermore, Doom really doesn't need much CPU grunt, it needs memory bandwidth above all. All the renderer does is to collect pixels from textures and put them in their place. Since the binary space partition tree algorithm is so good, it hardly ever needs to draw a pixel twice. But it needs to set each and every pixel once per frame. The added stuff is a DIV per pixel column on the walls and ceilings. That's not really computing intensive.

I'm not so sure I agree. Certainly, give Doom more memory bandwidth and it runs a lot better, but even then... Well, take a look at these four videos of Doom running first on a 386 & 486 with the same ISA VGA card and then on a 386 and 486 with a VLB graphics card (which has lots of bandwidth). Edit: for fun, I've added Doom on the A1200 with a 68040@50MHz.

See here for the ISA 386: [ Show youtube player ]
And here for the ISA 486: [ Show youtube player ]
Here for the 386 VLB: [ Show youtube player ]
And here for the 486 VLB: [ Show youtube player ]
Edit: Lastly, here is the A1200 with a 68040: [ Show youtube player ]

Now, it's clear that neither ISA card runs the game as well as the 486+VLB does, but there's still a pretty visible difference between the two ISA versions - the 486 clearly does better. More to the point, the 386 with the VLB card is visibly slower than the 486 using ISA.

Clearly, running the VLB 486 version is by far the best - so bandwidth does help. But without a fast CPU to back it up, Doom just does not run that well even if it has bandwidth to spare. The 386 VLB version ran very poorly, even though it had a ton of video bandwidth. Only when the fast memory bandwidth was coupled with a fast CPU did we see frame rate go up to much higher levels.

And lastly, the A1200 with the fast 68040 does a reasonably good job as well: it's clearly faster than either 386 even with the c2p penalty and it seems to be faster than the 486+ISA as well. Edit: I know the A1200 does not run full screen, but it seems to be 320x200, which is identical to the PC version.

Doom can run well enough on an 030.

I've never seen that happen myself. Just take a look at Doom on a 68030 Mac (which has chunky graphics and fast processor access to video memory*): [ Show youtube player ]

Maybe the Mac version is horrible, but I've never seen Doom run well on a 68030. That said, if you have a better example I'll be sure to take look
*) Compared to AGA at any rate: the 68030 in the Mac IIvx can push 4 bytes to VRAM every 8 cycles = 16MB/sec write speed.

That's not how I remember it. Originally the AGA features were already planned for the A3000 but Commodore pushed the A3000 out of the door before it was finished giving us "ECS". Then nothing happened for another few years...

Interesting, I remember it very differently. I don't remember anything about AGA features on the A3000 at all - only the A3000+ that became the A4000 instead. Now, I'm more than willing to accept I'm wrong, but I would like some more info to be sure.

Do you know anywhere I can find this stuff?

The A600 was a total desaster and actually resulted in heavy losses. It came in 1991. When it was demonstrated inside Commodore, the staff went "wtf, even more of the same old, same old?"

Commodore even made it for another three years after that. AGA+chunky should have come with the A3000 in 1989, then things might have gone differently. Commodore wasted all that time and development budget on the stupid 8bit computers and then couldn't compete when there were better machines on the market. Those machines didn't come out of nowhere, they had been in development for some years, too...

A small correction: the A3000 was first sold in 1990 and the A600 went on sale march 1992 (only a couple of months prior to the A1200 ). I do remember stories about the staff being disappointed with the A600 but as I understand that had more to do with the result than the original plan: the original plan was the A300, a cut down A500 on the cheap.

But the A600 was more expensive than the A500, while not offering any real benefits over it to those who didn't want an (expensive) hard disk, so it failed. Some of the first A600 motherboards actually say A300 on them, which is kind of funny I guess.

Yes. The question is how much too long they had waited. They didn't understand that the evolving computer business absolutely required them to start developing the next big thing the very moment they had their newest product out the door.

To be fair, the generation of computers they had made their money on all had unusually long cycles. Take for example the NES: a 1982 design still on sale in 1992. The upgrade to that was released in 1990, eight years after release. Or the 286, a 1982 design still put in new products (that managed to sell) in 1990.

This is not to say you don't have a point, merely that it was not just Commodore that wasn't putting new stuff on the market quickly and that the market mostly seemed to accept older stuff. This did rapidly change around the early 1990's and they should've seen that coming with the specs of some of the newer stuff on the market.

So I'm not giving them a free pass or anything.

To me the Archimedes always was a mystery machine that was supposed to be vastly superior to everything else but that nobody had ever seen. It was a unicorn.

It was indeed very rare outside of the UK, but the specs are simple to find as are plenty of examples of the machine in action that show that it indeed is rather interesting and fairly powerful for the time. Note that I never owned one. I got interested in them after hearing the Amiga developer of Pacmania comment in an interview that the Archimedes version of that game was best.

And then I started looking it up and checking YouTube videos and it is indeed an impressive machine.

Yes, because they were the biggest of the remaining bunch and Microsoft needed a living competitor to avoid monopoly related troubles. Commodore could have been the biggest remainder. They once were the No. 2 in the computer business behind IBM.

They were once No. 1, though that did not last very long. It is a giant shame what happened to Commodore and mostly self inflicted.

A good cautionary tale of what happens when you let the owners of a company focus purely on personal and shareholder gain and not consider the rest much. Sounds sadly familiar to how Apple is being run today.

I remember reading a discussion about this. Thomas Richter knows for sure. I seem to remember that he said something along the lines that compiling for 020 would only give a few percent more speed but would increase the maintenance and testing effort.

EDIT: I have just seen what Toni Wilen has written. I guess we can conclude that at least parts of the available 3.0+ kickstart versions are compiled for plain 68000.

I certainly accept that. It was merely me wondering as the kickstart refused to start and now I know why.

[grond]

Originally Posted by roondar

Surely higher frame rates do not make c2p relatively more expensive?

Of course, they do. The c2p always takes the same time, e.g. 20 milliseconds. If you want to do 40fps, you will spend 0.8 seconds of each second just doing c2p. If you want to do 1fps, you have 0.98 seconds of each second of the processor all for calculating the screen contents.

merely changing AGA in the A1200 into a fast chunky display would not have given the vanilla A1200 Doom.

Yes, I agree (I actually didn't think this point needed discussing).

A genuine question: if what you say about the game not actually needing CPU grunt but rather memory bandwidth is true, then why is spending 50 or so % of the available CPU time drawing pixels a problem? You clearly have memory bandwidth to spare at that point?

I don't understand this question.

Well, take a look at these four videos of Doom running first on a 386 & 486 with the same ISA VGA card and then on a 386 and 486 with a VLB graphics card (which has lots of bandwidth). Edit: for fun, I've added Doom on the A1200 with a 68040@50MHz.

I think this pretty much underlines my point. The 386 works against a superslow graphics bus which speedwise is pretty much like AGA. If you can calculate each frame in almost no time (486 with ISA), the CPU will spend most of its time stuffing graphics data down ISA's throat but will come up with a reasonable overall experience. A 386 with VLB would probably compete easily. The 486 with VLB flies.

Now for the A1200 example: I'm not sure it compares right. The PC demos are running the timedemo where gameplay is slowed down or sped up with the processor speed (on a modern computer the timedemo would finish in a split second making the dude run at almost light speed). This makes the slow examples appear slower than they were perceived at the time because you're essentially running through honey.

The A1200 is just somebody playing at what seems to be a reduced resolution on a PAL screen (less DMA, more AGA bandwidth). We can't tell it's not using EHB mode which speeds up c2p a lot.

Do you know anywhere I can find this stuff?

I read the A600 story and about the longstanding desire within Commodore for something new and exciting to supersede OCS/ECS in a "book" by a German support staff for Commodore that was recently linked somewhere for free download. I read it during my easter vacations and it was overall so confusingly written and overall bad that I threw my printout away after finishing it. But somebody here will surely know who that guy was.

I'm sorry I can't support the A3000/AGA story with facts.

But the A600 was more expensive than the A500, while not offering any real benefits over it to those who didn't want an (expensive) hard disk, so it failed.

Yes, this is another reason why everybody but the head management knew it would flop.

To be fair, the generation of computers they had made their money on all had unusually long cycles.

Yes, this is true. The business was changing and they didn't notice. They thought they could keep their ways forever. But then they could have known better: they saw how important it was to make the C=128 (EDIT to add: and C=65) compatible to the C=64 but more powerful at the same time. Why did they not see that all Amigas had this ability to be (mostly) downwards compatible right from the start? They could have made it more powerful all the time without breaking compatibility and without having to make much of an extra effort.

A good cautionary tale of what happens when you let the owners of a company focus purely on personal and shareholder gain and not consider the rest much.

It's more a problem of short-term bonuses. If I cut the dev budget now, I will get better quarterly numbers and thus a higher bonus while I continue to cash in on yesterday's product. When everything crashes and I haven't got a product to sell, I'll have another job at another company...

[vulture]

@roondar

from my experience, Doomattack is very playable on a 68030@50mhz

[roondar]

Originally Posted by grond

Of course, they do. The c2p always takes the same time, e.g. 20 milliseconds. If you want to do 40fps, you will spend 0.8 seconds of each second just doing c2p. If you want to do 1fps, you have 0.98 seconds of each second of the processor all for calculating the screen contents.

Oh, I see. I got that wrong, sorry - I forgot you need to do c2p as an extra on top of simply pushing the pixels.

I don't understand this question.

What I meant is: if Doom is limited primarily by video memory bandwidth and not CPU speed, but drawing the screen only takes 50% of the CPU including the c2p pass - shouldn't it run pretty well then?

I think this pretty much underlines my point. The 386 works against a superslow graphics bus which speedwise is pretty much like AGA. If you can calculate each frame in almost no time (486 with ISA), the CPU will spend most of its time stuffing graphics data down ISA's throat but will come up with a reasonable overall experience. A 386 with VLB would probably compete easily. The 486 with VLB flies.

But it really doesn't. The third video I showed is a 386 with VLB and a VLB card and it clearly runs much slower than the 486 with ISA card.

Now for the A1200 example: I'm not sure it compares right. The PC demos are running the timedemo where gameplay is slowed down or sped up with the processor speed (on a modern computer the timedemo would finish in a split second making the dude run at almost light speed). This makes the slow examples appear slower than they were perceived at the time because you're essentially running through honey.

Ah, that does make the comparison to the A1200 less valuable.
That said, I still think the A1200 video I showed had it running fairly well, time demo or not.

The A1200 is just somebody playing at what seems to be a reduced resolution on a PAL screen (less DMA, more AGA bandwidth). We can't tell it's not using EHB mode which speeds up c2p a lot.

Fair enough, I don't know if it's in 64 colours. The resolution however, does seem to be 320x200, which is an exact match for the PC version. It's probably PAL though, but that's pretty much the standard for the Amiga - the vasty majority of Amiga's were sold in Europe. Requiring it to run NTSC mode seems rather strange for a game that itself doesn't run at more than 35FPS on PC's (hard cap).

I did try to find a different version with a 68040 above 25MHz, but that turns out to be quite difficult. There's a bunch with a 25MHz 68040 and a bunch with a 68060 but no 68040/40MHz ones.

I read the A600 story and about the longstanding desire within Commodore for something new and exciting to supersede OCS/ECS in a "book" by a German support staff for Commodore that was recently linked somewhere for free download. I read it during my easter vacations and it was overall so confusingly written and overall bad that I threw my printout away after finishing it. But somebody here will surely know who that guy was.

Would be interesting indeed. And it does fit with the AAA development timeline, which allegedly started in 1989.

I'm sorry I can't support the A3000/AGA story with facts.

Understand, I'm not saying you're wrong - just that my memory differs.

Yes, this is another reason why everybody but the head management knew it would flop.

Indeed.

Yes, this is true. The business was changing and they didn't notice. They thought they could keep their ways forever. But then they could have known better: they saw how important it was to make the C=128 (EDIT to add: and C=65) compatible to the C=64 but more powerful at the same time. Why did they not see that all Amigas had this ability to be (mostly) downwards compatible right from the start? They could have made it more powerful all the time without breaking compatibility and without having to make much of an extra effort.

I agree that they should've and could've done better.

It's more a problem of short-term bonuses. If I cut the dev budget now, I will get better quarterly numbers and thus a higher bonus while I continue to cash in on yesterday's product. When everything crashes and I haven't got a product to sell, I'll have another job at another company...

Originally Posted by vulture

@roondar
from my experience, Doomattack is very playable on a 68030@50mhz

I found only one video that specifically said it ran Doomattack (I found plenty of ADoom and others though) and it doesn't seem to run much better than on a 386 to me. Everybody is different and I do admit that I might have played Doom like that on my A1200 back in the day, but I personally don't think it looks very playable. Doom is all about fast action and immersing yourself into the game and that, to me, requires a better frame rate than that. That's subjective though, I know.

Here's the video: [ Show youtube player ]