The Atari 5200 Wikipedia page

[rhindlethereddragon]

http://en.wikipedia.org/wiki/Atari_5200

 

Apparantly (according to the article), the 400/800 computer started off as a game system to replace the 2600, which Atari thought would be obsolete by 1980. But they decided to make the game console followup into a computer line instead in 1979.

 

Is all of this really true? Never heard any of this before.

[Pilsner73]

http://en.wikipedia.org/wiki/Atari_5200

 

Apparantly (according to the article), the 400/800 computer started off as a game system to replace the 2600, which Atari thought would be obsolete by 1980. But they decided to make the game console followup into a computer line instead in 1979.

 

Is all of this really true? Never heard any of this before.

 

The 400/800 development according to most people was being done to replace the 2600 but when Atari saw the success of Apple computers, they decided to take what was being developed and start building a computer. What I don't understand is why Atari just took the 400 and created the 5200 in 1982, they should have had something else ready by that time.

 

The C64 started life as a potential game machine also.

Edited May 3, 2011 by Pilsner73

[Kr0tki]

Is all of this really true?

Why don't you check the Wikipedia article for the 400/800? It's nicely explained there.

[Nukey Shay]

What I don't understand is why Atari just took the 400 and created the 5200 in 1982, they should have had something else ready by that time.

 

The 10-bit Atari 3200 (which the existing workforce was not accustomed to programming). So Atari was stuck in the next gen war without a next gen system. They did the quickest thing they could - repackage the system that was originally intended to be the VCS successor a few years prior.

[rhindlethereddragon]

It's ironic that the things that Atari did to "disguise" that it was an atari computer are the things that actually helped kill it:

 

A ridiculously complicated, un-durable controller.

 

A ridiculously oversized outer shell.

 

The power supply ridiculously jutting out of the RF box.

 

The cartridges ridiculously and needlessly big.

 

 

But they weren't fooling anybody, really. Word caught on pretty quick that this was an atari 400 computer (which had recently been clearanced out for less than $50..) in disguise. Not that there was anything wrong with that, the Atari computer was a serious piece of machinery. Atari had all of the ingredients to succeed, but made all of the wrong decisions. They should have released the XEGS in 1980, right along side the 400/800.

[Pilsner73]

What I don't understand is why Atari just took the 400 and created the 5200 in 1982, they should have had something else ready by that time.

 

The 10-bit Atari 3200 (which the existing workforce was not accustomed to programming). So Atari was stuck in the next gen war without a next gen system. They did the quickest thing they could - repackage the system that was originally intended to be the VCS successor a few years prior.

 

I know just figured they would have also been working on a new Atari 8-bit computer also, the XL/XE line really were revisions IMO.

[kool kitty89]

What I don't understand is why Atari just took the 400 and created the 5200 in 1982, they should have had something else ready by that time.

 

The 10-bit Atari 3200 (which the existing workforce was not accustomed to programming). So Atari was stuck in the next gen war without a next gen system. They did the quickest thing they could - repackage the system that was originally intended to be the VCS successor a few years prior.

No, it wsn't 10-bit and it wasn't cancelled because of technical difficulties on either end (programming or engineering), but was cancelled for political/bureacratic reasons and eventually replaced by the 5200. (specifically there was a conflict with the home computer division being concerned that the 3200 was going to conflict with or detract from the computers . . . though in the end, I don't see how the 5200 didn't do the same thing but with other disadvantages)

 

Some of that info is on Curt's page, but other things have been disconered since the update early last year. (Marty -wgungfu- addressed this in another recent discussion, I think on the 7800 boards)

 

Those sort of messes are real symptoms of Atari's management problems at the time. (some stemming from the dual Warner/Atari management and others from Atari themselves -albeit some of the later were growths of the former)

 

 

 

Now, there's the separate topic of the original intention of the A8 chipset back in '78/79, and that's totally separate from the 3200.

The original idea came from Nolan Bushnell's idea that the VCS (or any such console or computer) would only last about 3 years on the market before it died (that assumption crippled apples management as well). Luckily, Warner management didn't buy into that and invested in the VCS in the long haul (and became far more profitable for it) . . . though other problems compromised them. (including not repositioning the VCS's market as it aged, of course the market was new, so the concept of an aging system and a new/high-end system that would be marketed in parallel would have been experiemental -in hindsight, we can see the standard for success tends to be releasing a new system as such while gradually transitioning the old product into the budget market -at which point the high-end system becomes mainstream and no longer niche)

 

It should be noted that the originally planned VCS replacement was far more primitive than the eventual CTIA/GTIA+ANTIC+POKEY chipset. (iirc,it would have been more like just having CTIA and a simpler pokey without the SIO or keyboard functionality, with the CPU having to handle much more overhead for display generation, more like with TIA, so it would have been a much smaller step over the VCS)

[kool kitty89]

OK, here's the other thread I was thinking of:

http://www.atariage.com/forums/topic/177726-the-xe-game-system-as-successor/page__p__2248445#entry2248445

http://www.atariage.com/forums/topic/177726-the-xe-game-system-as-successor/page__st__25__p__2265087#entry2265087

 

What I don't understand is why Atari just took the 400 and created the 5200 in 1982, they should have had something else ready by that time.

 

The 10-bit Atari 3200 (which the existing workforce was not accustomed to programming). So Atari was stuck in the next gen war without a next gen system. They did the quickest thing they could - repackage the system that was originally intended to be the VCS successor a few years prior.

 

I know just figured they would have also been working on a new Atari 8-bit computer also, the XL/XE line really were revisions IMO.

Yes, the XL/XE were just revisions. (some things like the cross A8/PC DOS compatible 80186 based machine like the 1600XL would be an arguable successor, but perhaps closer to the C128 -except aiming at PC/DOS compatibility rather than CP/M and going for a higher-end form factor from day 1 given the mock-ups)

[Pilsner73]

OK, here's the other thread I was thinking of:

http://www.atariage.com/forums/topic/177726-the-xe-game-system-as-successor/page__p__2248445#entry2248445

http://www.atariage.com/forums/topic/177726-the-xe-game-system-as-successor/page__st__25__p__2265087#entry2265087

 

What I don't understand is why Atari just took the 400 and created the 5200 in 1982, they should have had something else ready by that time.

 

The 10-bit Atari 3200 (which the existing workforce was not accustomed to programming). So Atari was stuck in the next gen war without a next gen system. They did the quickest thing they could - repackage the system that was originally intended to be the VCS successor a few years prior.

 

I know just figured they would have also been working on a new Atari 8-bit computer also, the XL/XE line really were revisions IMO.

Yes, the XL/XE were just revisions. (some things like the cross A8/PC DOS compatible 80186 based machine like the 1600XL would be an arguable successor, but perhaps closer to the C128 -except aiming at PC/DOS compatibility rather than CP/M and going for a higher-end form factor from day 1 given the mock-ups)

 

Just wondering what if Atari had shrunk the 400/800 down ala the 600xl for the 5200 so it had a nice sleek case, worked out the controller issues and then released it with out the RF/power thing. Would this of been a much bigger success compared to the 5200 they did release?

[kool kitty89]

Just wondering what if Atari had shrunk the 400/800 down ala the 600xl for the 5200 so it had a nice sleek case, worked out the controller issues and then released it with out the RF/power thing. Would this of been a much bigger success compared to the 5200 they did release?

Atari already had that in 1982, but they canceled it in favor of only releasing the 1200XL:

http://www.atarimuseum.com/computers/8BITS/1200xl/600_proto/600proto.html

 

I already mused/rambled below about the 5200's possibilities for being a sleeker/more cost effective form factor (inside and out), and other things like directly releasing an XLGS of sorts instead of the 5200. (honestly that would have made far more sense than releasing the 5200, no reason to launch a new A8 derived but incompatible format if it wasn't going to be VCS compatible, inherently cheaper to manufacture, or had lockout to actually allow licensing control over 3rd party publishers -for content control and royalties; the 5200 did none of that, and updated/special controllers could easily have been done on the A8 -let alone VCS, from digital with added buttons to analog sticks, etc, etc)

 

Edit:

On another not (something I've argued before, but didn't mention below), is that well after the fact, in 1984, just after the liquidation of Atari Inc and mess left by Warner ruining Morgan's reorganization progress and making the transition to Atari Corp total chaos and anarchy (it's a wonder Tramiel managed to pull together some of the things he did), the 7800 was in that mess and left in limbo over negotiations. (Warner hadn't included it in the sale -or the GCC contract rather- and, due to the generally sloppy transition/sale Warner's part, that fact hadn't emerged until somewhat after the fact of the sale/split of Atari Inc)

Negotiations went back and forth for several months until Tramiel finally ended up paying GCC's contract in early 1985.

 

Now, that the context, but my point would be that, with the 7800 and general mess at the time (and the 5200 already officially discontinued -meaning tricky marketing to bring it back if they wanted to), that could have been the perfect time to push something like the XEGS (more likely a 16k direct derivative of the 600XL) and drop the 7800 entirely with the mess it was causing. (with their weakened position and the general mess, trimming down to as few products/formats as possible would make the most sense; the A8 already had lots of support and tons of stockpiled units and components -not to mention software, so building on that format, even in 1984, could have made far more sense than screwing around with trying to bring the 7800 over -they didn't have the resources or time to launch the 7800 like Atari Inc could have, and focusing on the A8 format could have been their best bet)

 

And looking more in hindsight, towards 1985/86 and onward, Katz's strategy to push computer ports (in leu of Nintendo locking in most Japanese Arcade developers), the A8 would have made even more sense in general. (digging into European developer support should still have been a good option too -something I'm not sure why they didn't do with the 7800)

And looking even more in hindsight, a game "computer" platform would totally sidestep Nintendo's licensing lock-in contracts . . . so actually, they wouldn't need to limit themselves to computer ports either. (Nintendo couldn't afford to change their policies to block computer publishing as well since the conflict of interests would have been too much for the time, even in the Japanese computer market -a lot of developers on the MSX and PC-8801 that supported the Famicom, let alone the US or -especially- Europe)

 

 

 

 

It's ironic that the things that Atari did to "disguise" that it was an atari computer are the things that actually helped kill it:

I agree in general, but I disagree on your examples.

 

A ridiculously complicated, un-durable controller.

I'd say moderately overcomplicated, and certainly un-durable (though not nearly as bad as some make out). It's also a bad mix of cut corners and expensive/inefficient design.

 

More buttons were necessary, and the keypad was necessary for some ports of more complex computer games. (albeit a keypad accessory could have been a good alternative, especially a more ergonomic keypad than was offered for the VCS)

The separate start, reset, and pause buttons were a bit unnecessary too, but one or 2 option/start/function/select buttons were a good idea.

There's also potential for compatibility with the normal Atari joystick pinout. (the 2 analog lines possible for added functionality, or the 5 digital lines for games using analog controls)

 

Analog was nice (would have been nice to have analog sticks offered for the A8 and VCS too), but not good for standard use. (the exception, not the rule)

Either using pull-up resistors or some other possible set-ups (especially if limited to 2 controller ports) could have allowed pseudo-digital or true digital joystick support. (preferably keeping with the VCS/A8 pinout standard)

A digital (or even pull-up resistor) based stick should have been cheaper and more reliable, on top of better suited to the majority of games at the time.

 

A ridiculously oversized outer shell.

Not to mention an oversized circuit board actually slightly lager than the 1200XL. (should have been slightly smaller than the 600 prototype)

Having a more general purpose expansion port would have been a lot better too. (like a pin-swapped PBI -if they were dead-set on separating it from the A8 line)

 

The power supply ridiculously jutting out of the RF box.

Initially, yes, though the auto-switching RF box idea was novel and eventually became standard, just not using that mechanism.

Changing from a simple co-ax connector to a DIN plug was a bit odd for the A8 power supplies as well. (especially since the entire system was 5V, so even an externally regulated power supply would only need 2 pins/poles)

 

The cartridges ridiculously and needlessly big.

This I disagree with. They aren't too big IMO, big enough to be impressive (with a very cool look to them), but not so big as to really be inconvenient. (and unlike NES carts, they make better use of that space for the label)

Lacking end-labels was a problem though (activision carts sort of almost do, but not in a useful way) . . . at least they made the carts face forward unlike the 7800 and VCS. (many A8 carts also lack end labels, so equally inconvenient)

 

The biggest problems with the 5200 was its overall marketing plan, cost ineffectiveness (and non competitive price point compared to the CV), weak launch lineup and initial pack-in, the bulk (complicating various expanses and shelf space logistics), controller issues (though they were still better than the CV or IV controllers in many respects -namely ergonomics), and lack of VCS compatibility on top of all that. (no adapter at launch either, and an inefficiently designed adapter -a tight low-cost optimized derivative of the A8 chipset withotu native VCS support should at least have had provisions for a convenient and efficient adapter module via an expansion port -on the front like the CV was quite convenient- best case probably would have meant using the 5200 CPU as the 6507 and perhaps have passthrough controller ports to interface with RIOT -unless they opted not to use POKEY for all I/O and added a RIOT in place of PIA, so only TIA and a cart slot would be needed in the adapter . . . sort of a modest cost savings by that point though, probably more likely not to have any PIA or RIOT onboard and repurpose POKEY+GTIA for all primary digital and analog I/O)

 

 

But they weren't fooling anybody, really.

The idea wasn't to disguise it at all, it was to be incompatible with and indirectly competitive with the A8 line, that was a big part of canceling the 3200. They'd also moved away from attempting a successor to the 2600, and aiming more towards a high-end companion system. (albeit that IS pretty much what most successful consoles have ended up doing, but with plans to shift that high-end niche console to the role of a successor platform after a year or so, depending on the case)

 

Really, though, if they didn't want to have a conflict with the game console and computer divisions as such, they could have left the VCS to the console side for the time being, and opened the computers more to the lower-end consumer range (albeit with a better low-end machine than anything else on the market had the 600 launched alongside the 1200XL in '82 -something between the prices of the VIC and C64, with better games than either for the time -C64 had yet to get any decent support).

 

Hell, given the cost inefficiency of the 1982 5200, the 600 probably would have been cheaper to manufacture, let alone more attractive to consumers via computer utility and expandability with PBI and SIO peripherals. (perhaps even a special game-oriented 600 with a built-in game rather than BASIC in ROM . . . maybe also make that a lower-cost version with a membrane keyboard, more like the 400 -game console with computer support- except finally down to a competitive game console price point -which the 400 was barely by late 1982, and not properly marketed)

 

Word caught on pretty quick that this was an atari 400 computer

In the computer community, I'm sure, but I doubt most average consumers/Atari users knew about that. (ie normal people, not the hardcore computer crowd of the time)

 

(which had recently been clearanced out for less than $50..)

Umm, iirc the 400 had just barely been dropped below $200 in late 1982, and only through mail-in rebates. They were still being sold in late 1983 at around $99 from articles I've seen. (the 600XL was oddly listed as being released at $199 in some late '83 and early '84 adds posted a while back, rather odd given the pricing of the competition and the 400 itself, especially for a 16k machine in 1983/84 . . . the 800XL was correspondingly listed at $299 iirc; of course, just a couple months later, Tramiel slashed the 800XL to $99 -not sure if that was at a loss, but it was probably lower than they should have gone in any case, mainly due to limited supply/distribution meaning they'd likely have ended up selling a similar amount at significantly higher prices -probably should have just dropped slightly below average C64 pricing of the time -which was hovering between $170 and just above $200 in mid/late 1984)

 

Atari had all of the ingredients to succeed, but made all of the wrong decisions.

Yes, wrong decisions for multiple market sectors and multiple regions. (they missed out potential both on the lower and higher end market segments, and in Europe -where low cost was even more important in general, not to mention prominent tape media support)

Kassar pushing the "appliance computer" approach and refusing the Apple II-like expansion features Atari engineers wanted was a mistake for sure. Not pushing for Class-B complaint cost-reduced machines by late 1981 was another mistake -probably leave all controller ports but remove the pretty much useless 2nd cart slot. (don't bother with any added features like the 1200XL, just the same machine in a smaller, sleeker, less expensive form factor) Or pushing a totally non-RFI restrained Euro-specific cost-reduced model (or line) prior to Class B even being established. (perhaps using that as the basis for the updated US models as well)

 

Hell, with an Apple II-like model at launch, they could aim that system at class A and save the cost/bulk/weight of the aluminum casting. (have it aimed at a monitor-specific "serious" computer, perhaps with a separate keyboard more like Tandy's machines -more so after IBM released the PC, to cut-in on that "serious" form factor)

They could have launched a home-oriented model alongside that, perhaps both the low-end 400 and an in-between model lacking the full expansion slots but with a proper keyboard and monitor port on top of RF, perhaps have then labeled the 400, 600, and 800. (probably have a PBI like connector on the 400 and "600" with a 1090XL style expansion module to add full "800" functionality -of course, priced to make the full 800 more attractive than even a 400+expansion -which would also be facilitated by the fact that the "800" would actually be cheaper to produce as a single-board design with very few RFI restrictions -until Class B came into play)

 

They should have released the XEGS in 1980, right along side the 400/800.

Maybe, or at least as soon as they could after the class B standard was in place. Maybe not so much the XEGS form-factor, but maybe something more like a 600/600XL with a cheap membrane keyboard built-in, preferably with controller ports on the front (maybe all 4, even though only Asteroids supported 4 player simultaneous), perhaps with an external keyboard port like the XEGS for a "full" keyboard. (with the membrane keyboard just there for minimalistic compatibility for games using keys and basic computer usage -the space bar's often-supported pause functionality would be one of the most frequently used features)

 

If not back in 1980 or '81, at least in 1982 (as with the canceled 600). 1983/84 could have seen an even bigger cost reduction with 16kx4-bit DRAM chips becoming available (so a 16k A8 console/computer could use just 2 DRAM chips), let alone if CGIA had entered production (I can only imagine Atari Corp lost the necessary documentation/plans/staff, or simply had too many remaining ANTIC+GTIA chips stockpiled -which wouldn't have been an issue with major sales of consoles/computers using those chips ). The next step beyond CGIA might have been consolidating the DRAM interface/refrsh logic and/or further consolidating POKEY or the CPU. (there was FREDDIE, but that doesn't seem to have actually saved cost/space over the older MMU -and the faster/more flexible bank-switching support seems to have been basically unused at the time -a new custom chip merging the DRAM logic and MMU logic might have been a good idea, or with POKEY, or some combination of those 3)

 

 

Again, given the fact that 5200 didn't manage to be cost cut relative to the computers and didn't add any lock-out (the main other reason to be incompatible), I can only imagine it was Atari's video game vs computer division "feud" and/or upper management/marketing making odd decisions that led to the 5200 rather than a direct A8 compatible computer/game system. (and the whole separate set of mistakes that led to them missing out on the European market)

 

 

 

Actually, the other possible (useful/logical) reason to use the A8 chipset (or part of it) in an incompatible manner (aside from cost reduction or added lock-out like the 7800 got), would have been to implement VCS compatibility. A hybrid VCS/A8 machine may have even had cost advantages over the Sylvia design (definitely R&D savings and the advantage of using chips already being stockpiles), with the main disadvantage being net cost of manufacturing the chips and long-term costs of the system (albeit consolidation could have mitigated that).

Given the 3200/Sylvia plans, the closes "kludge" based on existing hardware in 1982 probably would have been SALLY+ANTIC+GTIA+TIA+RIOT, probably with 16k DRAM (definitely cheaper in the long-run, though the 2k SRAM planned for Sylvia could have had some performance advantages and cost savings due to less board space -1 SRAM vs 8 DRAMs and added interface logic, but long-term would have more cost advantages with 16kx4-bit DRAMs and embedded interface logic). No POKEY sound, but they could at least enable RIOT interrupts to allow modulation of the 1-bit GTIA toggle (for pulse/square wave -at far higher resolution than TIA or an 8-bit POKEY channel even) or for modulating a TIA channel. (if you wanted volume control or for 4-bit PCM playback) Maybe they could even have even used the RIOT scratchpad for some added performance. (especially if mapped as zero page, albeit still only 1/2 the useful space zero page offers . . . I don't know too much about the VCS's memory map, but it would definitely have made sense if RIOT's scratchpad was mapped to the zero page address range)

Edited May 10, 2011 by kool kitty89

[ATARI7800fan]

I have wondered this myself before.

[kool kitty89]

I have wondered this myself before.

Which part? Are you responding just to the first post or the topics expanding from that?