Talk:Atari 8-bit computers/Archive 1

Archive 1

Amount of RAM in 400/800

I was sure that the 400 had 16KB ram and the 800 had 48MB ram. Can someone doublecheck this? --209.107.95.230

My 800 had 16KB expandable to 48KB. Thought the 400 started at 4KB; not sure. --Williamv1138 11 July 2003

Maybe it's because I bought mine later. At the time I made my purchase, the 400 shipped with 16KB and the 800 shipped with 48KB. --209.107.95.230 14 July 2003

Originally the Atari 400 came with 8K of ram and the 800 had 16K which allowed upgradable modules to be added. Atari later upgraded the ram in the Atari 400 to 16K and over the years, upgraded the 800 to 48K.

The memory module in the Atari 400 is identical to that in the 800 except the 800 had three slots for memory allowing mix and matches of 8k and 16k boards.

Also, the Atari 8bits were not 2mhz processors as stated in the article but 1.79mhz.

Amazing what rattles around in the head after all these years, huh? --James 15 February 2004

Edit summary re XL models

Forgot to add summary of my edit. The 1200XL did not have a built-in BASIC like the subsequently released 600XL and 800XL did, so I deleted the reference to a buggy BASIC as being a problem of the 1200XL. However, the 1200XL did feature a new OS which was different enough from the OS of the 400/800 such that many programs (mostly games as I recall) were not compatible. To address the problem, Atari released an OS patch called "The Translator", which loaded what was essentially the same OS as the 400/800. --4.7.239.195 28 January 2004

The translator disk loaded the older OS ver.B or OS ver.A on the XL/XE computers. Atari only released it on disk, no cassette version was planned or released. The problem with compatiblity was too many illegal calls and jumps were made by the offending programs. Alas, some of Atari's programs did that as well. --Pelladon 03:55, 31 August 2005 (UTC)

Paged 128K memory card for the 800

At least one company (can't recall the name) created a 128 kB memory card for the 800. It paged eight 16 kB pages into one slot's address space using a control register in the I/O space. We used this at NASA to create a proof of concept general aviation cockpit instrumentation suite; the displays could switch in 1/60th of a second, since the Antic display lists were within the paged memory. Does anyone remember the company that made this memory card?Insert non-formatted text here --George F Rice 31 May 2005

800XL expansion

The article does not mention anything about the expansions of the 800XL. I have two Atari 800XLs sitting here. They can be expanded with an MIO (essentially a Active RAM drive that plugs into the wall socket and stays powered). The MIO allows you to boot from memory and even load games and software onto it. I could have my Atari booted and running a game before my screen warmed up. The MIO expanded the RS232 port so that you could hook up to 3 floppy drives stacked, a modem and a printer to it. A guy at HACE Atari computer club in Houston, Texas ran a BBS off of his Atari for YEARS on end. I have the BBS software for it as well, but I never ran it. Called BBS Express. --Kim Nevelsteen 21:34, 15 August 2005 (UTC)

Don't forget the 130XE expansion. Wish I had more info on it. Both 800XL and 130XE had expansion cards made but very few came out.--Pelladon 03:51, 31 August 2005 (UTC)

Just added info about the Parallel Bus Interface (PBI). Feel free to add, develop, etc. Pelladon 07:14, 23 October 2005 (UTC)

Four controller ports: 400/800 a pioneer?

Just a thought: AFAIK, the 400/800 series is the first home computer/game console range to feature a total of four game controller ports built in. Anyone know if this is correct? And if so, has there actually been a single computer or console between the 400/800 and the Nintendo GameCube—introduced 22 years later—with this many controller ports built in? If these suppositions are indeed facts, they should certainly be prominently mentioned in the article(s)! --Wernher 16:14, 28 August 2005 (UTC)

I found a reference to it. I didn't believe it either, but apparently it seems ok. --Kim Nevelsteen 16:35, 28 August 2005 (UTC)

The N64 and Dreamcast both had 4 ports as well. And the N64 article says the first one to have it was the Bally Astrocade. Melodia Chaconne

Thanks! I learn something new on WKP every day! :-) (Of course I could've scoured all the home computer + console articles myself to find it out, but I thought I'd ask instead, in case someone knew more than what WKP had available. And, as always, it's less of an effort to ask someone else than do the research oneself... *blush*). --Wernher 13:12, 30 August 2005 (UTC)

I'm pretty sure that there was no pervious home computer, though, so it's still notable. BTW, the only other home computers with four joystick ports that I know of where the Atari STe and the Atari Falcon. They had two digital, standard Atari joystick ports and two digital/analog, Jaguar compatible joystick ports, the latter of which could each drive two digital, standard Atari joysticks, for a total of six joysticks. --OldSchoolDragon 15:04, 5 October 2006 (UTC)

XE naming

The X in XE stood for XL-Compatible.

Is this actually true? I always remember it as "XL-Expanded". Source please!

Maury 12:34, 6 October 2005 (UTC)

It was me that wrote the original sentence. I just wrote what I recalled from memory. I could not remember at the time what the E stood for. Thanks for pointing this out. I have edited the article to reflect this.

Ae-a 06:30, 9 October 2005 (UTC)

The first XE

BTW, the article mentions that the 65XE was the first XE. In the UK the 65XE didn't come out till the end of 1987, long after the 130XE. Was the reference a typo, or did the XE come out earlier elsewhere? If Atari *announced* the 65XE in 1985, perhaps this could be rephrased. Fourohfour 15:29, 6 October 2005 (UTC)

The same was true here in Canada as well. I assume they had intended the 130 to replace all models, and then introduced the 65 to hit some sweet spot in the market. Maury 12:12, 21 October 2005 (UTC)

If there are no objections in the next couple of days, I will change this so that it no longer implies that the 65XE was the first. If anyone thinks otherwise, can you please provide some evidence? Thank you. Fourohfour 13:14, 21 October 2005 (UTC)

The second generation: XL series

Atari addressed the 1200XL's problems in the hastily-completed 600XL and 800XL, which were largely identical to the original Sweet 16 specifications. These machines also had Atari BASIC built into the ROM of the computer. By the time the new machines replaced it the Commodore 64 had already become the market leader, and Atari was unable to address this.

The 600XL and 800XL were NOT hastily completed. They were planned all along because the 1200XL was too expensive to build. The 600XL and 800XL were part of the family of Atari computers that also included the 1400XL and the 1450XLD. The problem was when Atari moved production overseas, manufacturing problems and delays left them with far fewer stock than they wanted for the all important Christmas season. A lot of companies (TI, Coleco) had production problems too. Commodore, however had plenty of computers on store shelves. THAT'S how Commodore jumped ahead of the others. I'm getting sick of the revisionist history that people constantly spew out again and again... This section needs a rewrite to fix facts and to get rid of the POV slant. Pelladon 19:42, 11 October 2005 (UTC)

I don't understand how you could consider this to be POV, I'm not exactly who's POV you think it could be -- the C64 really was cheaper than the Atari machines, at least in every store I went to see them in (and as a teenage nerd, that was "a lot"). Likewise the market really was being dragged by Commodore, they were always the first to lower the prices during the "war".

Secondly the history is rather suspect. There was a very distinct delay between the introduction of the 1200 and the 600/800/14xx. I have all the promotional materials (although I may have chucked them recently when I moved), there's the 1200 batch, and then the "other" batch. I distinctly remember the 1200 arriving early in the year, and the 600/800 in the winter - by which time the season is over.

Finally the 1200 was a single-board machine (IIRC), like the later 600 and 800. As far as I can tell the only real differences were the keyboard and the cases.

Perhaps I was too hasty when I used the word "hasty", but I believe the historical record is extremely clear: the 600/800 were introduced in order to replace the 1200, which is not only what it did, but what they said it was supposed to do. Maury 22:32, 18 October 2005 (UTC)

That's what I said, the 600XL and 800XL were to replace the 1200XL. They were not hastily completed, actually they were behind schedule. Atari moved engr. resources from the 1400XL and 1450XLD to the 600XL/800XL. I remember because I phoned Atari about the availability of the 1400XL and 1450XLD and their response was always "delayed, don't know when...". Atari moved their production over from domestic factories to overseas. This caused numerous problems that took valuable time and engineering resources. That's what burned Atari during X-mas. Also, Tramiel cut deals direct with retailers, bypassing the computer store owners (and infuriating them) to stock C64 computers. That's why you saw C64 on store shelves. I was there too, pal.

The 600XL/800XL did NOT come out "early" 1983. The 800XL barely made it in Christmas in 1983! No 1400XL or 1450XLD at all. A far cry from the 600/800XL in mid 83, 1400/1450XLD in late 83 as Atari advertised! And forget the 1050 drive, that was nowhere to be found :D

The 1200XL was a single-board machine, but with tons of parts. It was built in the US. Meaning "Expensive". The 600XL and 800XL were built overseas. Fewer parts, cheaper keyboard. Meaning "cheaper". Get it?

The price war was over-rated. How come Apple and IBM didn't get burned?(HINT: they didn't burn their bridges with computer store owners). The real reason was the market was OVERSATURATED. Too many brands of computers, not enought buyers. Remember Spectravideo? How about Osborne? Or MSX? The price war would have happened eventually. Tramiel started it to get at TI, but it snowballed and the rest is history. Pelladon 20:09, 19 October 2005 (UTC)

Apple and IBM didn't get burned because they didn't sell into the same market. They sold primarily to business and scientific users, who were primarily interested in their expansion capabilities. It was only those competing with the C46 in the home computer market that got burned.

As to the dates, you are correct, I have backed them all off by one year by mistake. Fixing...

Apple and IBM had their dealer network and markets. IBM had $$$$$ to spare. The price war really opened IBM's and Apple's markets even further, because companies starting making PC clones and Apple clones instead. And that changed the computer market and eventually drove out Atari, Commodore, TRS-80, etc.

Again, their manufacturing problems hurt Atari computers because they were poised to make a profit, for the first time in their history. (separate from their games and home entertainment divisions, their cash cows) Pelladon 22:14, 19 October 2005 (UTC)

Re: "Not enough buyers". At the time of the price war, sales were booming. Even TI's sales were booming; the problem was that they lost money on each sale. Apple preferred to cede market share and maintain their prices, which they could do because they were an established standard. IBM wasn't a significant player in the home computer market. Mirror Vax 22:59, 19 October 2005 (UTC)

With TI, you're right, they were losing money on each machine. It was their tactic of taking a profit loss in exchange for market share. They finally decided to get out of the home computer market and realign for the business sector. Apple refused to lower prices, true. IBM wasn't a player in the home market until years later. Pelladon 02:06, 20 October 2005 (UTC)

Memory interleaving

Is the middle of this article really the right place to launch into a highly technical discussion of memory interleaving? And why would one possible compare it to the BBC, or a submarine?!? If no one objects, I'm going to remove that portion. Maury 22:32, 18 October 2005 (UTC)

He said the BBC *micro*; that's a more valid comparison than a submarine. Whether the interleaving stuff should be there is another question. *I* found it interesting, but this article is specifically about the Atari 8-bits, so... I personally feel that you can't include everything. If the technique is worthy of in-depth description, it should have its own article (if it doesn't already; please check thoroughly first).

My vote; note the interleaving technique (or lack of it) in the Atari, and provide a link to a more detailed article. You may wish to use the information in the new article (if creating). Please remember to note its origins if this is decided.

You can't include every detail without making the page too unwieldy for a general audience. Good organisation helps, but there comes a time when information has to be separated out, made more concise, or simply removed. I think the "software only" modes section (although of interest to me) is starting to exhibit signs of this bloat. It's great, but.... this isn't a "complete Atari tech manual".

Fourohfour 14:04, 19 October 2005 (UTC)

Changed, let me know how it reads now. Maury 21:36, 19 October 2005 (UTC)

Joystick ports

The joystick ports on the Atari were definitely NOT serial I/O ports. Although enterprising developers used them for this purpose, one could just as easily modulate the monitor port to the same end. I have removed this comment.

Maury 12:10, 21 October 2005 (UTC)

The 6520 PIA was designed with 2 8-bit bidirectional I/O ports. The Atari allow some access to control the 6520 PIA. The PIA was specifically designed to control peripheral devices through its I/O ports (in the Atari, joystick ports). People found uses for it such as a inexpensive printer port connection, robotics control, home security system control, etc. And NO, you could not modulate the monitor port as it was a analog output only. --Pelladon 07:44, 24 October 2005 (UTC)

Early Machines: the 400 and 800

As consumers discovered that the two machines were largely identical, they generally bought the lower-priced 400. All of these problems conspired to force developers to essentially ignore the features of the 800, and develop for the 400 as their target. Expansion cards were already rare, but cartridges for the second slot in the 800 were just as rare.

Hey Maury, how could developers focus on the 400 when the majority of software came on floppy?? As for the second cartridge port, I already explained that it became technically obsolete and was dropped (16kB and larger cartridges came out). Pelladon 16:17, 21 October 2005 (UTC)

Focussing on the 400 essentially consisted of ignoring those features that existed on the 800 alone. Perhaps I should re-word it that way?

The 16k cart was another issue entirely, it required bank switching which means only certain portions of the memory were available at any time. This simply is not as good as having a single, real, 16k cartridge slot. Had Atari done that instead -- which they could have easily -- then Atari BASIC would have been MS (a good thing) for instance, and even early games and other carts could be, well, better. Maury 16:43, 22 October 2005 (UTC)

Atari 16kB carts were NOT bank switched, they were 16kB ROMs. The cartridge slot is just a electrical connection, it wasn't limited to 8kB. Where did you get the idea that the cart slot was limited to 8kB??? Go read Mapping the Atari. OSS had 16kB bank switched carts. Later, the 32kB and larger carts were bank switched. MS BASIC didn't have nearly enough commands that supported the Atari's features and it consumed too much memory. Memory was very EXPENSIVE in the early days. Pelladon 19:57, 22 October 2005 (UTC)

The cartridge slot is just a electrical connection, it wasn't limited to 8kB. Where did you get the idea that the cart slot was limited to 8kB???'

Duh, from the pinout diagram. Notice there are 13 address lines, A0 through A12... 2¹³ = 8096. Yes, it is "just an electrical connection" as you put it, one with enough pins to support 8k. Each of the carts was mapped into a different 8k block, 40959->49151 for the left cart, 32767->40958 for the right. As to why you think Mapping the Atari supports your argument, I can't fathom. This review of the book has a nice ascii chart of the map, and if you look for the words "right cartridge" and "left cartridge" you will note they are mapped into separate 8k blocks, exactly what I have been saying all along.

What Atari and everyone else found out was they could map 16kB directly from just the left cartridge slot. That made the right cart slot obsolete. Gee, on page 103 of Mapping the Atari, it says "It is possible to have 16K cartridges on the Atari by either combining both slots using two 8K cartridges or simply having one with large enough ROM chips and using one slot. In this case, the entire area from 32768 to 49151 ($8000 to $BFFF) would be used as cartridge ROM." And I thought this was self-explanatory. Pelladon 21:04, 23 October 2005 (UTC)

As to Atari BASIC, having spoken to the author personally, I can say with no hesitation whatsoever that what you claim above is just plain wrong. The authors attempted to use MS 8k BASIC, but when MS supplied it to them it was 12k long (a side effect of the 6502s lower code density) and almost completely undocumented. After trying to get it running for some time they gave up, because they couldn't get it to fit onto an 8k cartridge (direct quote). It was not, as you claim, because it didn't have enough commands (which are trivial to add to BASIC, having done so personally), or because of the ROM costs. It was, very simply, because carts were 8k only.

No your just misquoting me. What I said was MS BASIC wasn't popular because one of those reasons was it didn't have a lot of commands that supported Atari's graphics and sounds. No Player/Missile commands, no scrolling, etc. BASIC XL and other programs did. BASIC XL outsold MS BASIC. The buyers decided. Pelladon 21:04, 23 October 2005 (UTC)

You go on to say "OSS had 16kB bank switched carts", as if this invalidates what I am saying. You do understand the meaning of the term "bank switching", right? It means that the 16k of code is switched in and out of the 8k space, making it appear as an 8k cart (which is all you can have) at any one time.

Atari didn't used bank selection in their 16kB carts because they didn't need it. OSS used bank-selection because their programs were languages and they wanted to save memory for programming. Their carts had additional chips (74LS175, 74LS02, 74LS00) which acted like a 4-bit register to track which bank to use. What's the big deal? Pelladon 21:04, 23 October 2005 (UTC)

Frankly I'm equally at a loss as to how you can claim Atari 16kB carts were NOT bank switched and then only a few lines later state OSS had 16kB bank switched carts, thereby invalidating your own statement. Maury 19:36, 23 October 2005 (UTC)

I meant to say 16kB carts from Atari Corp. were not bank selected. I used OSS as an example of a company that did use bank selection in their 16kB carts. If it came across that I implied that ALL Atari 16kB carts weren't bank selected, oops for that. So Maury, why don't you just show me, with EVIDENCE, how Atari Corp bank switch their 16kB carts. I would be very interested. Pelladon 21:04, 23 October 2005 (UTC)

Cartridge info

Just so there is NO confusion:

CARTRIDGES A AND B

There are two 8K regions reserved for plug-in cartridges. Cartridge B, that is the right-hand cartridge slot found only in the ATARI 800 Home Computer, has been allocated memory addresses 8000 through 9FFF. Cartridge A (the left-hand cartridge slot in the ATARI 800 Computer console, and the only slot in the ATARI 400 Computer console) has been allocated memory addresses A000 through BFFF and optionally 8000 through BFFF, for 16K cartridges. If a RAM module is plugged into the last slot such as to overlay any of these addresses, the RAM takes precedence as long as a cartridge is not inserted. However, if a cartridge is inserted, it will disable the entire conflicting RAM module in the last slot in 8K increments.

Taken from Atari 400/800 Technical Reference Notes: Operating System User's Manual, Section 4, page 31. --Pelladon 06:34, 24 October 2005 (UTC)

You failed to quote the very next section. It reads...

Most cartridges consist of two ROM chips on a single circuit board. Moreover, both chip sockets have identical pin assignments. In other words, the chips can be switched to opposite sockets and the cartridge will still work. The difference is in the chips themselves. On one chip, the A12 pin acts as an active-low chip select. On the other the A12 pin acts as an active-high chip select. Therefore the state of the A12 pin selects between the two chips.

What do you call it when a pin is used to select between two banks of memory? Bank switching. Is this evidence enough to answer your question posted at the end of the last section?

Actually the quote you provided is a example of chip-select. Are you being funny or something? --Pelladon 04:21, 25 October 2005 (UTC)

Oh and another thing, the next section reads:

MAPPED I/O

The 6502 performs input/output operations by addressing the external support chips as memory; some chip registers are read/write while others are read-only or write-only (the Atari Home Computer Hardware Manual gives descriptions of all of the external registers). While the entire address space from D000 to D7FF has been allocated for I/O decoding, only the following subregions are used:

This is Section 4, page 32. The very top. Right after the CARTRIDGES A AND B section. I don't know where you got that other quote. --Pelladon 06:24, 25 October 2005 (UTC)

But wait, don't take my word for it. Just to be sure I wasn't talking out of my ass, I contacted the creator of a cartridge copier for the Atari, and asked his opinion" If you are talking about the Atari 400 800 etc, they have a hi and low that can be accessed. So, in a sense there is a form of "bankswitching".

That is NOT bank switching. It describes, if I understand it correctly, using two 4K ROMs to make an 8K cart. The result is identical to using a single 8K ROM chip. Mirror Vax 18:57, 24 October 2005 (UTC)

I cannot believe that a person had to contact someone else to determine if the Atari used bank switching or not, given the available documentation which is explicitly clear. --Pelladon 06:42, 25 October 2005 (UTC)

Thank you for the insult. The reason I contacted the hardware vendor was to find out if he called it that. Otherwise one could get into an endless semantic debate. If people making the cartridges call it bank switching, that's good enough for me. Furthermore, I don't think Vax's comment is correct, note the ATH's description of A13. Maury 12:12, 25 October 2005 (UTC)

Just because you don't understand how something works does that give you the right to change something to fit your idea of what it is. This is not a sematic debate. This is a technical description that is clear and well documented. So why did you lie about the quote? That wasn't in the Tech Reference Notes. Wikipedia is about facts, not opinions or personal analysis. --Pelladon 16:08, 25 October 2005 (UTC)

Vax's comment is correct for a 8kB Atari ROM cartridge. The 16kB Atari ROM cartridge uses two 8kB ROM chips to make a 16kB ROM space. Picky picky. --Pelladon 16:24, 25 October 2005 (UTC)

I have no idea what quote you claim I am lying about, because you didn't bother to note it specifically. I'm guessing you mean the one here [1]. But basically if you're going to call me a liar, then I have no interest in continuing on this topic. IMHO when both the author of OSS basic and the designer of actual hardware say you wrong, you're wrong. Perhaps you should write to them can call them liars instead of me. Maury 18:59, 25 October 2005 (UTC)

My quote came from the "Atari 400/800 Technical Reference Manual:Operating System User's Manual". Part no C016555 Rev. A. Page 31. This is the official documentation for the Atari 8-bit, from Atari themselves. I gave the proper reference for everyone to look it up. Not just you. You go and look up a different site, accuse me of failing to quote the next section from a different site, and then misinterpret that quote as well. I don't know who you are fooling. --Pelladon 20:45, 25 October 2005 (UTC)

Here are some documentation about cartridge types, taken from Atari 800win Help:

Type 1: Standard 8 KB cartridge: Standard 8 KB cartridge, that occupies 8 KB of address space between $A000 and $BFFF.

Type 2: Standard 16 KB cartridge: Standard 16 KB cartridge, that occupies 16 KB of address space between $8000 and $BFFF.

Type 3: OSS '034M' 16 KB cartridge: There are two types of OSS cartridges. Both are 16 KB and occupy 8 KB of address space between $A000 and $BFFF. The cartridge memory is divided into 4 banks, 4 KB each. One bank ('main') is always mapped to $B000-$BFFF. The other 3 banks are mapped to $A000-$AFFF. The current bank is selected by accessing a byte in $D500-$D5FF. Only 4 lowest bits of address are significant.

I don't know how it's achieved electronically. But from programmers' perspective, 16K cartridges between $8000-$BFFF _did_ exist. They were not "bank-switching" because both 8K parts were permanently mapped in. Usage of the address lines for this purpose is beyond my knowledge though. --Cyco130 01:10, 12 December 2005 (UTC)

Attempt to avoid an edit war

I don't understand why you commented on my last edits as vandalism. All I did was re-arrange existing comments, split the price war into two parts, and add a header. Please, detail exactly what it was you feel was "vandalism" in my edits. Maury 12:21, 24 October 2005 (UTC)

You're putting too much focus on Tramiel, Commodore and the price war. This page is about Atari computers. If you want to write about price wars and Tramiel, do so on their respective pages. I commented that it was starting to look like vandalism, I didn't say that it was. I left out Atari's corporate problems because that's a separate topic altogether. --Pelladon 14:56, 24 October 2005 (UTC)

Fair enough, but I am going to re-org again. As it stands the benefits and drawbacks of the 1200 are spread across four paras, one only two sentances long.

But frankly I disagree about your comment on the price war. In the last version it took up only two paragraphs, and yet is the single overriding reason for what was going on in the market. What, you think Atari lowered the price of the 400 to $79 (a loss on every sale) because they WANTED to cut the profit margins? Without mentioning what was going on -- and note that other articles on machines of the era generally comment on it as well -- you end up with a history that reads like Atari was just introducing a bunch of new machines "just because". That is not the case. As you noted, the reason Atari did so poorly in Chrismas '83 was supply, but without mentioning the market we have no idea WHY Atari ran into these problems. It's like studying biology without evolution, just facts with no narative.

I reverted back to the previous version by Maury. I feel the 1200XL section is too large, needs to be trimmed to the essentials. Don't speculate, just stick to facts. The 1200XL is in the same family as the other XLs, so it should not be grouped separately. Finally, details of Sweet 16 should go on a different page, since the 1000 and 1000X were never built. --Pelladon 03:37, 25 October 2005 (UTC)

I didn't mention why the Atari did poorly cause no one really knows. Not enough evidence. Just going on price alone isn't giving an accurate picture. Lots of problems at Atari, but no one talking about it. You can't analyse this, not with so little to go on. --Pelladon 04:50, 25 October 2005 (UTC)

I'm not going to add those sections back in for now. Instead I'll ask for thrid party comments first...

Vote

Third party readers: do you believe the section on the price was was out of place for this article? Maury 16:28, 24 October 2005 (UTC)

• Include. Needs to be there to get the context. --Wernher 21:18, 24 October 2005 (UTC)

Separate out "description" section?

I think the article might read better if the technical parts, notably the chip sections, were removed to a separate section. But I can't figure out if it should be above or below the History section. If it's above then we can simply refer to chips by name in the History section, making it smaller and easier to read. However that also has the side effect of making the history, which I think is the fun part to read, get pushed down the page.

Comments?

Maury 12:26, 25 October 2005 (UTC)

The chips already have their own pages. The chip descriptions are just summaries. You can't describe the Atari 8-bits without describing the chips. That's like saying we'll leave the technical description of calculus out so we can discuss the history, which is not the main focus. --Pelladon 15:58, 25 October 2005 (UTC)

No no, I mean quite the opposite. I'm proposing adding a small section with each of the summaries, along with an overall description of the machine as a whole, all placed in a "Description" major section. Right now there is no real description of the machines, just bits and pieces through the history section. Maury 18:52, 25 October 2005 (UTC)

a paragraph on the 6502

It states "the GTIA had just enough time to read out a line before the processor wanted access again." This is not true, all bus access by ANTIC/GTIA is done when the CPU is halted. Commodore and Apple computers fetched video data and did refresh during the unused portion of the cycle but Atari didn't. Otherwise they wouldn't have been able to get away with using 200ns DRAM. --DamageX

That's right. Atari ran the CPU at 1.79Mhz and used cycle-stealing DMA. The Apple II ran the 6502 at 1Mhz and used transparent DMA. Mirror Vax 20:32, 27 December 2005 (UTC)

Original Atari 400/800 announcement

"The machines were announced in December 1978 as the 400 and 800, "

Everyone everywhere states this, or Nov. 78, or Oct. '78, or Fall 1978, but where is a copy of this announcement? I'd really like to know!

I accept that the machines were introduced at the January 1979 CES. Anyone know the dates in January of that CES?

If no one can turn up a public announcemnt from 1978, it seems that the proper date of public announcement of the 400/800 may actually be January 1979, at the CES.

-hunmanik