chepe

Commodore rant (which also applies to Atari somewhat): Commodore had many successful machines but never really seemed to understand why. They were obsessed about 'price brackets' not unlike say car manufacturer: "We need a mini car, sedan, mid-size...". When they had a successful model they produced it as long as possible, then developed a successor for it, even if they already had a better machine in production. Because they were obsessed about the idea that there were markets for computers in many price ranges, including very cheap ones. And to certain extent this was true but Commodore took it too far: when you are already producing cheap C-64, what is the point of even cheaper C-16 since almost anyone can already afford the C-64? Same idiocy led to Amiga 300/600 debacle and many others. Commodore wasted resources designing machines which were already obsolete, instead of designing ever better computers and letting the old models take over "cheap price" bracket. Pricing was certainly important, that is how ST got a head start after all. However, people were ready to pay a little extra if it genuinely brought them better capabilities. My dad bought me 520STFM ca. 1989. At the time C-64 was still in sale and was much cheaper. However nobody I knew in my hometown bought a C-64 after that even if it was only half the cost of ST. Everybody bought Amiga, ST or PC. And that is what went wrong with Atari and Commodore in their last years: they saw their ever more powerful competitors and correctly noted that their marketing advantage was that their computers were cheaper. But they became too obsessed about pricing, it doesn't matter how cheap your computer is, if it's obsolete nobody will want it even for free. By the time Atari figured it out with Jaguar they no longer had the resources to pull it off. I'm sure Jaguar would have worked out much better if Atari had used few more million of dollars and 3-4 months more to figure out the architecture, work out bugs etc. But they didn't have that time or money and rushed the product out. Pretty much same for CD32, which IMO could have worked too.

Sure, but ST brand was probably unsalvageable by the time Falcon came out. The timeframe to do something to seriously update ST was in 89-90 and Atari didn't. Commodore was arguably also late with A1200, but managed to get at least initial, promising burst of sales which guaranteed some chipset support from software producers, before everyone realized the machine was obsolete. I think Commodore might just have made it, or at least survived for 4-5 more years if they had managed A1200/CD32 thing bit better. In fairness, home computer market of 80's/early '90s was notoriously dynamic and hard to predict. Few people in 1990 would have thought, at least in Europe, that in 4 years Commodore and Atari would both be kaputt.

Atari almost seemed like hostile to the idea that customers added RAM by themselves...in Amiga 500 RAM upgrade was the simplest thing, and eventually almost everyone had 1mb, and most games were designed to take advantage of it. Unlike STE, Amiga 1200 was big enough upgrade that it got people replace their old Amigas with it. I bought one myself because I wasn't ready to invest to PC world yet and wanted a transitional machine. However I could see writing on the wall already. In 1993, average good PC game took 10 to 15 megabytes and biggest games took over 20. Even if A1200's hardware could run games like this, putting them to DD floppies was hopeless. So what happened was that Amiga diehards rushed to buy A1200, once they had one, nobody else was interested in the machine because it couldn't play PC games and was too pricey just to play console grade games.

So that's why A1200 didn't get a HD disk drive? Reason I heard was that HD drive wouldn't have fitted to the selected chassis - which sounded stupid, but Commodore did stupid decisions like that all the time so it sounded believable. They also left hard drive out of the baseline A1200 on the basis that "we couldn't decide what size it should have been", or at least that is what they said at the time. I forgot that Mega STE had 16mHz 68000. Maybe that would have been plausible for regular STE as well if changing to 68020 was too expensive. Or did it require cooling fan?

Hmmm, I've never heard that one. I do remember 4mb RAM limit, strange if they underclaimed it? I mean, it's not like many people had even 4mb but it's always good for bragging rights. What did Atari System use for graphics? Was it entirely ran by CPU? I do note that CPU was downtuned to 7.16mhz, apparently to make it compatible with sound chips. I don't think this architecture would have been too useful for ST, even if Atari Games had been willing to license it to Corp. Developing new custom chips was expensive, for everyone equally. However, PC world had economies of the scale working for them. In 1988, PC sales were something like 10 to 11 million worldwide - TEN TIMES more than ST & Amiga sales combined. Even if only fraction of them equipped their machines with VGA cards, it still meant that much more PC graphics chips were made than (already inferior) Amiga/ST chips...Commodore & Atari needed to hit a home run on their next attempt because they had resources to develope new chipset only once. Both of them missed, wasting resources on 'deck chair rearrangement' upgrades like STE and A600. I remember when Amiga 1200 came out, it was quite attractive for its seemingly cheap pricing. However, it came with no monitor, no hard drive, only 14mhz CPU...if you added up monitor and hard drive, for about same cost I could have bought 386 w/SVGA which was superior to A1200 in pretty much every respect. Commodore and Atari had not only lost their performance advantage, they had lost their price advantage. Hence I arrive to my original point - Atari should have concentrated more on upgrading CPU and memory rather than matching Amiga's chipset. I dunno what compability problems 68020 would have brought, presumably some. But it would have been great marketing value in 1988-89 timeframe. And 1mb RAM should have been minimum, even for STE. Already in ~1990, Amiga 500's were seldom even sold without 1mb RAM. Another issue was mass storage. DD floppies were becoming obsolete in around 1990. I remember playing some late Amiga games, it was horrible pain even with 2 disk drives. Commodore made a huge mistake when they failed to upgrade mass storage for Amiga 1200.

Hey, I've got nothing against STE - I never had one myself but I gather its sound capabilities were great at the time. I just don't think it was the best possible move from Atari to release: it wasn't great enough upgrade for existing ST users to change into (I personally never considered it myself), and it costed more than old ST models, thus losing its prime advantage over Amiga. My point was just that these 'what-ifs' always seem to get fans excited about what powerful custom chips they would have liked to see in ST, and Atari itself sorta went down that route in later years, but that was not what made ST popular, and IMO Atari should have stuck with the original philosophy. Keep it simple, keep it cheap.

The YM chip was probably the best commonly available one which could be easily bolted on ST. Sure custom chip like Paula would have been nice, but Atari was under huge pressure to get the machine out to generate cash flow because even with injection of Tramiel capital, they only barely made it. Those times, you needed to quickly build up 'critical mass' of customers to wake up software developers. One of the best ways to ensure that was to make sure that the machine was 1) available 2) affordable. It only went seriously wrong later, when they began to worry about matching Amiga, when they really should have been worried about PC and consoles. Somewhat strangely, Commodore fell into exact same trap. Both companies fell victim to the tunnel vision.

Maybe I'm in a minority but I think ST was pretty fine as it was, originally...sure few minor things should have been different or at least modernized bit more quickly. But it's not like other contemporary machines were perfect, all of them were hurriedly developed and contained various quirks and shortcomings - original IBM PC was a horrible contraption which would have died a slow, agonized death if it wasn't for IBM brand name. Atari's punchline for ST was 'Power without a price' and that's the philosophy they should have followed with STE, or in fact, there never should have been STE. They should have kept it simple. Stick a 68020 on ST, 1Mb RAM minimum and HD disk drive. Maybe some minor upgrade on graphics. Put this out ca. 1988, call it 'ST020' or something to maintain brand continuity. Keep custom chips at minimum. Should be possible to have pricetag comparable to Amiga 500. At the time Macintosh II costed like $5000 so it would compare very favourably.