After seeing yet another topic on AtariAge about why the Commodore 64 (C-64), released in 1982, succeeded in both sales and software support, where the Atari 8-bit series, released in 1979, didn't, I thought I'd offer up my usual thoughts on the matter in a more formal manner. To my mind, it's pretty simple. While the Atari 8-bits had a roughly three year headstart, in those three years, Atari wasn't able to make much headway in the market despite having the best audio-visual potential of the time, bar-none. The missteps with the lovely, but initially flawed, Atari 1200XL, didn't do them any favors, and by the time the C-64 started picking up significant momentum in 1983 when its retail price started dropping to the point where no one was able to compete effectively with its value proposition and still turn a profit, Atari was already done, particularly since they lacked Commodore's supply chain advantages.
Certainly price was a factor in the C-64's success in the US, but in the rest of the world, particularly Europe, price was often the primary driver (e.g., long after the US standardized on reliable, but expensive disks and drives, Europeans were still using unreliable, but cheap cassettes and tape decks), making Atari's inability to produce a low cost 8-bit in a timely manner particularly devastating. The influx of talented European programmers to the C-64's software pool can't be underestimated as the Atari 8-bit line struggled to make it into homes there. It also didn't do Atari any favors that they had multiple models out in the wild with 16K - 64K of memory at that time, making it difficult to target the higher spec. We can't underestimate the value of every Commodore 64 having 64K from its first day on the market to its last, making ports to platforms without a significant user base of guaranteed 64K-spec machines less likely. [Read more]
Now, you might ask about the Apple II (1977), which had a similar spec spread as the Atari 8-bit line, right down to the original run of computers not supporting the same graphics capabilities as the newer computers (same difference between the TI-99/4 and 4a). The Apple II was always a different market with different expectations. The typical Apple II owner was a more affluent purchaser who wasn't afraid to upgrade, and there were a lot of developers who used that machine, which made a significant difference despite never being the top selling computer (many originals appeared on the Apple II that were later ported to other platforms). There were a lot more factors in the Apple II series' success than just sales numbers, which obviously weren't anywhere near a match for the C-64, but that's a discussion for another day.
In short, I'd summarize this topic's theme as Atari being unable to press their advantage in the three years when it would have made a difference, and then having far too many models of computer to choose from. If, after the 400/800, and even after the 1200XL debacle, they standardized on one 64K spec model for a good price (say, the 800XL), they might have moved more units and might have been more competitive. As it was, Atari was probably their own worst enemy. Hindsight is 20/20, though, and it's clear when we look back that most computer companies made some really poor decisions, so at least Atari wasn't alone in that regard (and yes, this includes Commodore). Naturally, back then, everyone was learning on the job and there was no rule book to follow, so what we see as an obvious bad decision now might not have been so obvious back then. If it was, they would have surely acted on it.
Of course (plug time!), the histories of these computers and more are detailed in my latest book, Vintage Game Consoles, which also goes into some of the reasons why the various platforms did or did not have success and in what areas, and what the general competitive environment was like at the time. The concept of success is relative of course, as today we still get to enjoy these wonderful platforms in countless ways, so, in their own way, they've all managed to succeed. In fact, even the least successful platforms have their supporters today. There's something very "personal" about this older technology that lends itself to that (in fact, I talked about that idea in this blog post). In any case, while the reasons I've outlined in this blog post merely touch on the countless factors that went into one platform succeeding where the other didn't (again, relative), this should at least give you a general idea of why things happened the way they did.
Agree? Disagree? Sound off in the comments!
Atari BASIC also was extremely slow. The benchmarks in Creative Computing show the 1.75mhz Atari was an order of magnitude slower than the 1mhz Commodore. Using a 3rd party BASIC solved this, the Atari coming out ahead. That being said, it is important to remember how strangely important the built-in BASICs were back in the 80s.
Just another factor.
Just another factor.
I'm not sure I necessarily agree with that. I don't know how much of an impact BASIC speed made in a purchasing decision. By and large, the C-64 was considered to have a very odd BASIC implementation and of course was mated to a legendarily slow disk drive (thanks to a flaw in the serial interface), and that had little impact on its sales, obviously.
C64 BASIC was the old and rather primitive Commodore PET BASIC. Graphics and sound were achieved by using POKE statements so you really had to know what you were doing. (To the uninitiated, POKE statements place a specified value directly into a particular address in memory - real low level stuff - e.g. POKE 1000,23 would put the value 23 into mem location 1000. A BASIC POKE statement interprets or compiles directly onto just two instructions in machine code - so chances are it ran pretty quickly)
Special locations were reserved for making sounds and displaying images. The values to be POKEd had to be carefully selected to achieve the desired result.
BASIC c64 listings in magazines also made a lot of use of the special graphical character set - so between that and the ubiquitous POKE (and PEEK) statements they were really hard to read and more error prone than listings for other machines.
C64 certainly appeared to do better in the UK than the Atari - but I always hankered (and still do) after an old shape, pre-XL Atari 800. The UK software houses seemed to take to the C64 for some reason. Not sure about the hardware features of the Atari hardware , but the hardware sprites of the c64 were much touted as an advanced feature (even though they required lots of POKE statements to make them work - it wasn't for the faint hearted even in BASIC).
The Atari 800 is indeed a striking looking computer. I love the four joystick ports (only the second time they appeared on a computer after the VideoBrain) and two cartridge slots (which I believe was the first time and not repeated again outside of add-ons until MSX machines), as well as the general boxyness (in a good way). I'd argue that the 1200XL has a better keyboard and, for some, a better design, but overall, the 800 will always be the class of the series.
Back in the day I actually was going for a nice 800XL setup but that got intervened with when my parents got a C64. In hindsight I am glad they did as the C64 proved to be a much more popular machine and it had a lot more software available for it than the Atari range of machines over here in the. Netherlands.
Of course now I own the Atari range of machines as well and both have their merits.