The 5200 appears to be Atari just buying some time, ie get something more advanced the aging 5200 out there as competition (judging by US magazine ads i've seen) were busy comparing their versions of games to the versions the 2600 recived and the 2600 was looking rough, whilst they worked on a 'true' follow up (going purely from what i've taken from the video), would this be fair to say?.
If the video promotes that, it's completely false. We have the full story in our book.
First off, what became the 7800 was developed by the company GCC in response to the release of the 5200. Not because Atari was doing a stopgap because they knew they were moving to it.
The 5200 was specifically designed to box out the Intellivision, being placed as Atari's "higher end" offering with the 2600 being moved down to the lower end. It was not intended as a replacement for the 2600, there's no way marketing would ever allow that.
The road to a followup to the 2600 actually started way before that. What became the Atari 400 was intended to be a next-gen game console, and in fact was still designed as one up until 6 months or so before the introduction in 1979 when a keyboard was slapped on it because marketing didn't want to interfere with the 2600 sales (sales of the console had been dismal in 1978 and caused massive losses for the company. So the focus for 1979, per Warner's direction, was to get the 2600 on it's feet). So the 400 became a game computer instead. (This sets up the relationship to the 5200).
Then in 1981 the effort began on three separate products: Stella, a repackaging of the 2600 with no changes. Sylvia (also called Super-Stella), intended to directly compete against the Intellivision. PAM (Personal Arcade Machine), a high end console that Atari was going to pour a lot of market research into.
Sylvia was at one point also referred to as Project X, and was in essence an enhancement of the 2600 concept - almost a hybrid between the 2600 and 8-bit computers. It featured a full 6502 CPU, an enhanced version of the original TIA chip, and a version of the ANTIC chip called FRANTIC. It also featured a full 2K of ram and a speech synthesis chip. It was to be packaged in a case similar to the Atari 2700 (and whose case was used in the Sears Video Arcade II).
PAM was purely a designed by market research console. Starting in 1981, Atari did exhaustive research asking gamers what they wanted in a new console. When it became clear that three products at three price points would be rejected by retailers and consumers, Sylvia was dropped and PAM became the focus - taking on both high end and compete against Intellivision roles. PAMs design was purely targeted towards early 80s game buffs, heavy coin-op players and those who would always buy top price products. The features of PAM were all designed on the findings of this research and what gamers at the time said they wanted.
The all in one power/auto-switch RF chord (last seen on the RCA Studio II) was put there because gamers requested a cut-down on the amount of wires, since most entertainment centers already had a rats nest of wires behind them. Just a single button to turn on the power and switch the tv.
As far as compatibility with the 8-bit lines, there were several stories we were told by ex-employees regarding tension between the home computer department (HCS) and PAM group, with HCS not wanting PAM to step on their territory. Hence the different memory map. There was also talk about a keyboard originally being attached through the extra two joystick ports (the reason for having 4). The expansion port contains most of the lines for an SIO, so other peripheral were planned but scrapped. We didn't include this material in the current revision of the book because we're still holding out for more concrete documentation (i.e., a directive from marketing, which is who usually stuck their noses into projects to change them).
There was also initially a speech synthesis module expansion for the 5200 (a holdover from Sylvia and being done on contract with Milton Bradley), and the 2600 adaptor was planned from the beginning. The module was cancelled, and the 2600 adaptor was delayed because of a snafu that did not make the adaptor electrically compatible with the 5200s cartridge port. Gary Rubio came out with a fix for the 4-ports already out there with the problem while the final runs of the 4-ports had the modifications built in.
Also, regarding the joysticks, Atari was aware focus groups evaluating System X before the launch were complaining about the sticks being non-centering. Engineering wanted to delay the launch to have time and fix the issue, but Marketing wouldn't allow it. So the boot around the stick was upgraded to a thicker/heavier rubber to provide a modicum of centering while they moved on to design self-centering sticks.
The Video System X they originally had planned may have been a better system but, due to the difficulties with programming for it, they scrapped it and just used existing parts of their 800 computers to make the console.
No, it was the reason I stated above.
Atari got it backwards on this one. If they wanted to use the ANTIC/GTIA/POKEY chipset in a console, it should have been done before they
did it in a computer. Better yet...release it as a console with computer add-ons...ie the SIO, Keyboard
and such already built in to the 5200. Then offer all the add-ons later.
I'm sorry, but that comes off as an opinion based on hindsight rather being familiar with the timeline and projects mentioned above. That just was not possible given the circumstances presented and where these projects evolved from.