Was anyone else disappointed with the A1200?

[dreadnought]

Originally Posted by grond

But Commodore was too cheap to do that. They happily settled for "good enough" and "works (sort of)".

How much more costly would be implemeting this feature? Is it the cost of research or more expensive chip or something?

I'm not saying they didn't make mistakes, but sometimes it's easy to forget how super tight was the market environment they were operating in at the time.

[freehand]

If you snooze you loose, if they had implemented as said above the amiga would of been very special all over again.

[eXeler0]

Originally Posted by freehand

If you snooze you loose, if they had implemented as said above the amiga would of been very special all over again.

You’d think the 256 chunky pixel color mode in VGA standard from 1987 would be a hint where things were headed… but no…

[oscar_ates]

Planar was suited to low ram computers successfully. You can use 5bits per color not 8 bits. It was not very suited to 3d fps games and by 1992 ram chips became large and fast enough for chunky graphics. So planar technology became obsolete. Snes had mod 7 where they could rotate backgrounds to give pseudo 3d effect. Together with superfx chip they survived the 3d fps era. Commodore was not prepared for anything nor they have a solution

[Promilus]

Originally Posted by oscar_ates

Planar was suited to low ram computers successfully. You can use 5bits per color not 8 bits. It was not very suited to 3d fps games and by 1992 ram chips became large and fast enough for chunky graphics. So planar technology became obsolete. Snes had mod 7 where they could rotate backgrounds to give pseudo 3d effect. Together with superfx chip they survived the 3d fps era. Commodore was not prepared for anything nor they have a solution

Planar was suited to low ram computers and obviously gaming consoles which are inherently low on RAM anyway. It did make some aspects of modifying 2D easier as well plus introduction of new modes was straightforward - just add few new planes and et voila. It was not bad idea with blitter and copper doing heavy lifting there. But did become PITA once (S)VGA became dominant standard on PC. So that's 90s. Obviously AGA tried to fight it off own way (still bitplanes) but once 3D was introduced all hope was lost.

[AestheticDebris]

Planar makes sense right up to 7 bitplanes. It's more flexible in terms of how many colours you can have (the alternative is byte packing which really only works for 2,4 or 16 colours without wasting RAM). It's also easier to do blitting operations with, shifting left and right a few pixels is just done with simple barrel shifting etc.

But AGA crossed the line and went to 256 colour. At that point using 8 bitplanes instead of just encoding a pixel per byte doesn't really make any sense. It starts to be more awkward to plot individual pixels and the advantages of barrel shifting no longer apply when you could have just moved entire bytes. What's more it becomes a case of diminishing returns to increase colour depth, there was little call for anything between 8 and 16-bit colour depth so you were never going to want 10-bitplane displays etc.

[grond]

Originally Posted by dreadnought

How much more costly would be implemeting this feature? Is it the cost of research or more expensive chip or something?

Zero or close to zero. You need a switch (like FMODE register bits and so many others that were introduced for AGA) and 8 multiplexers (about 8 transistors each depending on implementation) to reshuffle the bits from the shiftregisters before palette lookup. With slight changes to the chip layout (which was newly done for AGA anyway) we are talking about engineering days at most.

[lmimmfn]

Even if AGA had chunky mode, how much DMA would be needed to fill the screen? It might have needed VRAM to boost the bandwidth.

[NorthWay]

With chunky pixels you could do a blitter that uses implied mask and can do a cookie-cut with 3 memory accesses instead of 4. And no memory needed for the mask either.
(And there is an optimization of not fetching background source (just advancing pointers) if mask is all-1(i.e. do not show any of the background) that triggers much more often and would be worth spending transistors on.)

[hammer]

Originally Posted by lmimmfn

Even if AGA had chunky mode, how much DMA would be needed to fill the screen? It might have needed VRAM to boost the bandwidth.

Refer to Individual Computers' Graffiti which effectively enables chunky pixels from Amiga's Chip RAM bandwidth.

http://wiki.icomp.de/wiki/Graffiti#U...he_chunky_mode

[ Show youtube player ]
Graffiti graphic box with Amiga's Doom port with 68030 @ 50 MHz CPU Ventisca 1230 III accelerator. It's A1200's AGA Chip RAM bandwidth.


VGA has 4 bitplanes (16 colors) and 8-bit (256 colors) chunky pixels.

Commodore could have integrated an additional chunky pixel raster for Lisa instead of a 16 million color palette.

[hammer]

Originally Posted by Gilbert

Was anyone else disappointed with the A1200?

Most Amiga users and magazines seemed to be very happy with the A1200 when it came out. I wasn't at all, and a look at the first games pretty much ended my association with Amiga gaming. I just saw the same games with more colours and a bit smoother. There was no wow factor. After that I stuck with the Amiga 500 (with half meg memory expansion) and my Super Famicom (Jap SNES).

Here's what Commodore got wrong in my opinion

1. Too much focus on creating higher-res screen modes with more colours (and also making the blitter work in these different screen modes) and not enough on enhancing gaming(8 or maybe 16 sprites when the comparitively old Megadrive and SNES could manage 64 and 128 respectively). It's a bit like the original Amiga - yes it can display 4096 colours on screen, but the majority of the games for the system were 16 colours (Albeit some had added some Copper magic) and most didn't even run at 50/60 fps. That was fine back in 1985 but 7 years(!) later you expect a significant upgrade.

2. There was a mild improvement to dual playfield mode. Great!... when the SNES had 5(?) playfields and could scale and rotate whole screens. Commodore seemed to have no sense they were competing here....

2. Sound chip needed 6 channels to get a decent track playing with sound effects. Again SNES and Megadrive have 6 channels each. Using the same sound chip from 1985 was ridiculous!

3. Like the original Amiga, if you wanted to get a good number of objects on screen with a lot of colours and scrolling, you had to spend ages using hardware tricks or specific techniques. Time = money and developers aren't going to want to spend 2 years making an arcade quality game on the A1200 when simpler systems exist....

I do have a CD32 now, but it's not very impressive from a technical point of view, even the mighty Banshee is bettered on both the SNES and Megadrive. The reason I like it is because it offers something a bit different and it's an Amiga It's fairly obvious it had no hope of competing long term. I just find it hard to see what Commodore was thinking with the AGA architecture??

Without 32-bit Fast RAM, A1200's hardware performance potential is halved.

[pixie]

Originally Posted by hammer

Refer to Individual Computers' Graffiti which effectively enables chunky pixels from Amiga's Chip RAM bandwidth.

http://wiki.icomp.de/wiki/Graffiti#U...he_chunky_mode

[ Show youtube player ]
Graffiti graphic box with Amiga's Doom port with 68030 @ 50 MHz CPU Ventisca 1230 III accelerator. It's A1200's AGA Chip RAM bandwidth.


VGA has 4 bitplanes (16 colors) and 8-bit (256 colors) chunky pixels.

Commodore could have integrated an additional chunky pixel raster for Lisa instead of a 16 million color palette.

It could have been the ultimate dongle!

[sandruzzo]

Originally Posted by NorthWay

With chunky pixels you could do a blitter that uses implied mask and can do a cookie-cut with 3 memory accesses instead of 4. And no memory needed for the mask either.
(And there is an optimization of not fetching background source (just advancing pointers) if mask is all-1(i.e. do not show any of the background) that triggers much more often and would be worth spending transistors on.)

Blitter needed destinations' byte alignment.

[sandruzzo]

With a 14-28 mhz chipset and Ram, a copper that could move 2-4 moves with one instruction(depending on FMODE) 16 bit palette, blitter with byte alignment, you could have all the chunky that you wan

[hammer]

Originally Posted by pixie

It could have been the ultimate dongle!

Graffiti box and 68030 @ 50 Mhz accelerator card are two Amiga add-ons without economies of scale.

68030 and 68020 have similar effective IPCs (instructions per clock).

Commodore's Ramsey 32-bit memory controller has a 25 Mhz clock speed.

Due to Commodore management's "read my lips, no new chips" directive, Commodore engineers weren't allowed to build raster graphics IP on Ramsey 32-bit memory controller infrastructure during Amiga 3000's development.

[lmimmfn]

Originally Posted by hammer

Refer to Individual Computers' Graffiti which effectively enables chunky pixels from Amiga's Chip RAM bandwidth.

http://wiki.icomp.de/wiki/Graffiti#U...he_chunky_mode

[ Show youtube player ]
Graffiti graphic box with Amiga's Doom port with 68030 @ 50 MHz CPU Ventisca 1230 III accelerator. It's A1200's AGA Chip RAM bandwidth.


VGA has 4 bitplanes (16 colors) and 8-bit (256 colors) chunky pixels.

Commodore could have integrated an additional chunky pixel raster for Lisa instead of a 16 million color palette.

The Grafitti is still hampered by chipram bandwidth, even at 320x256 on OCS you're still stuck with a maximum framerate of around 33FPS, double that on AGA machines.
I dont think it's a sufficient solution for the amiga range, yes it solves the chunky problem, answers the bandwidth question but it highlights that basic chunky mode in AGA would not have been sufficient without throughput, I.e. VRAM.

Which also raises the question that by 1992 end of when the A1200 released shouldn't they have released it with 2MB VRAM, the priced were only 20% more than DRAM by that time.

Don't get me wrong, the FPS would be fine for the likes of Doom but would be completely inefficient for high rez PC games. Perhaps the conversation becomes irrelevant with the arrival of the juggernaut that was the PS1.

[logo]

@hammer

Gilbert wanted so bad to play japanese games (I admit some of them were great) and made the choice to buy a "cheap' SNES to play the same games as his friends. Gaming was his main concern.
Most of his technical arguments are questionable. The funny thing is that 30 years later, many people who left the Amiga in the 90s or refused to support AGA systems, still try to justify it somehow. On the other hand, those who remained and acquired AGA machines, seem only rarely to regret it.

Now, I would like ask one thing: In 1992 which other computer was able to offer 2D action games better than the Amiga ones ?
Only japanese had one, and it was an expensive one...

[dreadnought]

Originally Posted by logo

Most of his technical arguments are questionable. The funny thing is that 30 years later, many people who left the Amiga in the 90s or refused to support AGA systems, still try to justify it somehow. On the other hand, those who remained and acquired AGA machines, seem only rarely to regret it.

What a bizarre take. So I "Refused to support AGA", as if it was some important political cause, and now I'm trying to "justify" moving on? Oh, boy

I don't mind our usual squabbles about hardware & software, these can be fun of sorts, but when it all veers into cultism - or slagging off other platforms/games - it's just sad (yep, especially considering that 30 years have passed and it's old farts who should know better typing away).

[grond]

Originally Posted by lmimmfn

yes it solves the chunky problem, answers the bandwidth question but it highlights that basic chunky mode in AGA would not have been sufficient without throughput, I.e. VRAM.

Which also raises the question that by 1992 end of when the A1200 released shouldn't they have released it with 2MB VRAM, the priced were only 20% more than DRAM by that time.

I don't think that AGA's 7MB/s were definitely too slow for 1992. AGA could scroll and the 7MB/s were only a limit for screen updates for which updating the entire screen as in FPS games only was a special case. With a 32bit wide blitter (also a relatively simple update/addition), the bandwidth would have been ok for another few years. People would have accepted to get worse results than a top-of-the-line PC for some games if those games (or types of games) had run on their Amigas at all.

Changing DRAM for VRAM, on the other hand, likely would have required far more work because it would have messed up all the timing-sensitive OCS games that depended on certain sequences of DMA slots and CPU chipmem cycles.

[logo]

@Dreadnought

Why do you feel concerned ?
I didn't say "all of them" but "many of them".
Let's me try to explain myself better. The first post of this thread try to explain why it was the right move to refuse to adopt an AGA Amiga back in the days.
For me it was the contrary, and it's one of the reasons Commodore died (not enough customers) and why we didn't see a lot of great AGA games.
Most systems don't get their best software on the first years. For example, the Megadrive and of course the A500 early adopters had to wait before to see the best of what their machines could do.
I must admit I would have more consideration for the opinion of someone who had an AGA computer, and didn't like it (I'm still talking about Gilbert).

About the importance of the choices we made 30 years ago, no it was not a political cause, if you're happy with what computers world today is. For some people like me, who dislike the most part of it, to stay an Amiga user was in some way a political act, a try and a modest contribution to keep the control of home computers (I put Atari users in the same bag, it wasn't Amiga exclusiv). Of course, we failed...
For the ones into gaming only, a choice between Mario and Sonic, hadn't the same meaning, I do agree.