The answer recently posted to Did Atari make more money from arcade games or consoles? quotes a New York Times article from 1982 https://www.nytimes.com/1982/12/19/business/the-game-turns-serious-at-atari.html

Atari's future may lie with its home computer division. The market for home - as opposed to small-business computers - has taken off this year and Atari has seen its sales quadruple, to an estimated $300 million. Yet the division is still losing money, perhaps $25 million $50 million this year.

Wait a minute. How the heck was Atari still losing money on computers in 1982?

They had great hardware; the Atari 800 was Jay Miner's penultimate masterwork before he went on to design the Amiga chipset; it was still considered good as late as the mid-eighties.

The great hardware didn't languish in obscurity; Atari was one of the best-known brand names in the industry.

Nor did it lack software support; Atari was perfectly positioned to support their machines with cartridge versions of arcade games, and Star Raiders is still cited as a killer application for home computers.

The company did not lack skills in low-cost manufacturing or distribution, nor was it without retail presence, as demonstrated by their decisive victory in the second generation of game consoles.

It's not like they underpriced their hardware; the 800 was released at more than $1000, and even the 400 at over $500.

The 1979 release got them well ahead of most of the competition.

By the same token, the accounting can hardly have been still paying off R&D cost in 1982, particularly since they flatly refused to invest in a follow-up 16-bit home computer. (That's what led to Jay Miner and Joe Decuir going off to join a startup named Hi-Toro to work on a shiny new machine called the Amiga.)

So with all those advantages going for them, at a time when home computers were one of the hottest industry segment, the sort of greenfield market entrepreneurs would kill for:

How on earth did they manage to lose money on home computers?

  • 11
    I had always assumed Atari was a casualty of the "Home Computer Wars", which is usually thought of as a battle between Commodore and TI. TI tried to compete on price, and it was said at one point they were losing $50 per computer sold. The joke answer to "How could they survive?" was "Volume!". If your competitors are low-balling price, whether or not they can afford to, it was bound to hurt everyone in the home computer market.
    – RichF
    Feb 11, 2020 at 22:23
  • 3
    I always suspected they simply made another mistake on the scale of ET and with the upcoming competition they couldn't take it.
    – Mast
    Feb 12, 2020 at 7:03
  • 2
    Many companies can't make the decision to kill their cash cow and go for the new thing until it is too late. Atari's business was arcade and video games. Home computers (separate from video consoles) were a sideline, and a money losing one at that. Actually committing to home computers didn't happen until too late (if it really did at all at the high levels determining strategy).
    – Jon Custer
    Feb 12, 2020 at 13:35
  • 7
    The Smithsonian did a podcast about the failure of Atari recently; a transcript is here: 99percentinvisible.org/episode/the-worst-video-game-ever/… -- the TLDR is: though they certainly lost a lot of money on ET, that was a symptom, not the disease. The disease that killed Atari was clueless management who prioritized short-term sales of shoddy products over a long-term strategy of ensuring quality and value. Feb 12, 2020 at 20:29
  • 1
    "It's not like they underpriced their hardware" That's pretty much the definition of "losing money". The hardware was too expensive to make to sell profitably in the numbers they needed to amortise the fixed overheads.
    – JeremyP
    Oct 15, 2021 at 7:56

5 Answers 5


In 1982 the original 400 and 800 were on the market. These were expensive machines to implement. Even the low-cost 400 was significantly more complex and expensive than something like a VIC-20. Here is an image of the 400's parts:

enter image description here

It has:

  • a mobo (left) with two plug-in cards and sockets, one for the 10k ROM and one for 16k RAM, and the cartridge slot (center of board, black).
  • a separate and rather complex PWR card (right), which also handles the SIO port and holds the RF modulator. The RAM was PMOS and needed +12/5/0/-5, thus the complexity of this card which took a single 9V (IIRC, maybe 6?) input.
  • So a total of four circuit cards and eight connectors!
  • a seriously big chunk of cast aluminum for the RF shielding
  • an upper case with a door (with the word Atari on it) that had a spring-loaded latch (the triangle thing on the RF shield) and another spring to pop and hold it open. It also has its own aluminium shield on the bottom of the door, not visible here.
  • manually installed keyboard, speaker and cartridge holder, dozens of screws.
  • and look at all those diodes and resistors on the mobo!

Now let us compare this with the VIC-20:

enter image description here

This is the entire system, albeit without the upper case. It is a single card. The shielding is stamped metal sheet (and I believe the bit at the very bottom of the image?). It uses edge connectors for expansion, along with a single DB9.

Say what you will about the VIC-20, it was cheap. And it proved that the #1 selling point for a computer was its price. And then came the 64. It was cheaper than the 400 and more powerful than the 800. So as Commodore started to crush Atari in the market, Atari responded by lowering its prices and preparing a new high-end machine to differentiate itself.

So basically it was due to the attempt to keep product flowing into a market that was rapidly moving down the price scale. This has killed lots of companies in the past and will in the future too.

It is difficult to understand how Atari got itself into this situation. They were aware, from the start, that the FCC shielding that drove up the costs were a significant problem. There were also concerns about the price of the SIO. And by 1981, the complexity of the motherboard could be greatly reduced - by this point NMOS RAM was widespread and removed all the 12V complexity. And the company had purchasing power far in advance of even Commodore, they sold 8 million 2600s in 1982 alone.

Now any competent management should have had a back-burner project to take advantage of these changes, but it seems Atari didn't even start considering such changes until early 1982. At that point, the VIC was eating their lunch. This is not a technical issue; the 600XL could have been in the market and would have been significantly more cost-effective than the 400.

I assume they simply got distracted by the success of the 2600. It seems like a massive waste of talent.

  • 6
    Atari also kept spending lots of money on R&D which didn’t necessarily end up delivering, and on pushing into markets where they didn’t have much success (especially outside the US). Feb 11, 2020 at 16:14
  • 4
    @StephenKitt Did Atari spend lots of money on R&D? I've seen a figure floating around of $100 million when they had $3 billion in revenue. That's 3% of revenue on R&D, which as I understand it is considered low for a box shifter, let alone a technology company.
    – rwallace
    Feb 11, 2020 at 17:05
  • 6
    But Maury has a very good point. The 400 had much higher manufacturing cost than the Vic-20. If they were trying to compete with the Vic on price, that would certainly do it; that's what sank the TI 99/4a. Yes, they should have designed a cost-reduced version much earlier than they did, as a crash priority project. There seems to have been a lot of general ineptitude in Atari management in the early eighties...
    – rwallace
    Feb 11, 2020 at 17:13
  • 3
    Where Ray Kassar oscillated between on the one hand insulting his best game designers and driving them out of the company, and on the other hand hiring Alan Kay to do long-term research (good) but failing to make any arrangements for such to be turned into products (bad). The middle ground of product development seems to have been lost. Maybe the need for a cost-reduced 400 was part of that missing middle ground.
    – rwallace
    Feb 11, 2020 at 17:15
  • 3
    Yeah, do a teardown on an Atari 800, it was an exquisitely built machine. Like Mac Pro tier detail and complexity. Not a single board machine like a C64, it had a cast aluminum chassis. Feb 13, 2020 at 20:54

I can only add to the other answers with observations and memories, some of which come from owning an Atari 800 in the time period you're asking about.

First, notice the names flying around in the space of only a few answers: IBM, Apple, VIC-20, Commodore 64, TI. Add to that list TRS-80, Timex Sinclair, and the Coleco Adam (released in 1983), and more. In 1982, the home computer market was glutted with options and a wide range of price tags.

IBM was the choice for business & scientific applications, which bled into the home market and defined it as a serious computer. Apple promoted their line to U.S. educators, particularly K-12, and had established a reputation as a serious home computer as well. (By serious I mean, "a computer for doing real things, not merely video games." I'm talking perception, not reality.) Commodore's aggressive price-cutting and success in retail chains led to the VIC-20 and C-64 being the choice for cost-conscious buyers. (These are broad brush-strokes, obviously.)

For most consumers, their purchase decision was dictated by tastes, income-level, and experience with or access to a machine outside the home. (People would even buy a machine because their neighbor had the same model, making it easy to, ah, share software.)

The early marketing for the Atari 800 was a sincere attempt to frame the line as an all-around consumer solution for home finance, education, games, and so on. (This Atari dealer video, while dated, shows the breadth of applications Atari was promoting early on.) They failed to follow through, and over time the Atari line came to be regarded as merely a game machine with a higher price tag. Their R&D attempts to find a toe-hold in the education market bore no fruit, and they simply had zero story to tell the business world. (After all, this was a 40-column computer requiring a television for its display.)

So by 1982, Atari's 8-bit line had one avenue for immediate income, the home market, and that meant competing on price and software titles. With Apple's and Commodore's success, Atari found themselves as the platform being ported-to rather than ported-from. Worse, they were sometimes overlooked entirely, leading Atari to license or write their own titles to fill gaps in their catalog (such as with AtariWriter, a surprisingly great word processor). And since some home consumers simply wanted to play video games, Atari found itself competing with its own 2600 console.

The company wasn't in a position to fight a consumer price war, but by 1982 lacked the cachet of Apple or IBM to stand above the fray. In 1979, a 16K Atari 800 cost $1680 and was sold via an Atari authorized dealer; by mid-1983, a 48K Atari 800 sold for $165 on the shelf of a Sears store, a situation they absolutely did not want to find themselves in. Atari overpriced the 1200XL (leading to a rush on the 800, which they'd dropped in price to clear inventory), meaning they didn't have a machine to cost-compete with the C-64 until a year after the latter's introduction. Commodore simply dropped their price in response.

Between the trifecta of Apple, IBM, and Commodore, Atari found itself wedged out trying to differentiate itself. Management kept toying with dropping the line and returning to a pure video-game strategy, while the other three companies were obviously committed to the personal computer. When 2600 sales collapsed at the end of 1983, Atari found itself without their cash cow, and therefore out of time and options.

  • 1
    BBC Micro and Amstrad were also quite popular around that time I recall. Feb 12, 2020 at 2:10
  • I wasn't attempting to be exhaustive, just trying to indicate the sheer number of choices a consumer had at the time.
    – Jim Nelson
    Feb 12, 2020 at 2:15
  • excellent references ... especially the sales statistics
    – eagle275
    Feb 12, 2020 at 15:19
  • 2
    I agree with the “serious computer” part in particular. Offices wanted computers without graphics. The less graphical capability, the better. Atari computers were known for their impressive graphics.
    – VGR
    Feb 12, 2020 at 19:35
  • 1
    Atari's video game image didn't help in our house when it came to computers. I received a TI99/4A in part because Dad had TI a calculator on his desk and I "already had an Atari" (the VCS/2600 of course). Jul 22, 2022 at 17:56

I'm going to make a guess.

For starters, Maury makes a good point about the manufacturing cost of the 400/800 series. I certainly think that is a key point, but not the whole story.

Based on data found here which is a spreadsheet of sales numbers from the era, you can see that between 1979 and 1982 things are picking up for the home computer market in general. Looking at the Apple II, Atari, TRS-80 and to some degree Commodore (it only lists the C64, the VIC is presumably in the 'Other' category), sales are increasing steadily in that date range. Good, things are moving upwards and Atari is actually doing better than other specific models, though there is this pesky 'Other' category which is doing very well.

If we then focus on the 1983-1986 range, there are shifts we can see:

  • The PC is obviously emerging as the business platform of choice. We know how this story goes.
  • The Apple II did well with the introduction of the //e in 83/84, obviously giving it more steam over the now dated II/II+ platform that existed up that point. In the mid-80s, the II's capabilities were now dated and the only reason to really buy one was because of the existing software library.
  • The C64 hit the market as a low cost competitor and is outselling everyone except the IBM PC.
  • The Atari 400/800 starts to decline and effectively disappears by 1986.

So let's narrow the focus to Commodore and Atari.

The VIC-20 and Commodore 64 were cheap. IMHO, it competed with the Atari 400/800 since the Apple II and IBM PC simply weren't in the same class of "game machine", so people who wanted a game computer would go for the Commodore or Atari machines. But considering cost, the C64 was cheaper than the Atari so we could assume that this was a major factor. When Jack Tramiel went to war on cost, he would win because of his vertical integration of components.

But thinking that cost alone shouldn't be the answer, we can consider three other things:

  1. At Atari it seemed that a significant portion of its management didn't realize they needed to keep pushing the envelope and innovate. Sure, they reduced cost with the XL line but they appeared to not know what they were doing with what they had. This made them fall behind.
  2. They didn't make documentation publicly available until De Re Atari which was about 2 years after release of the 400/800. This surely didn't help matters.
  3. The video game crash in 1983 had Atari's name all over it, simply because Atari was synonymous with video games. That doesn't make the company look like a good bet when comparing against the Commodore.

Though I can't cite a source for it, I've read that piracy was easy on the Atari because of the smart 810/1050 disk drive design. This would obviously be a deterrent for software publishers. Sure, the Commodore 1541 was in the same boat, but look at the number of units sold - the market is obviously more lucrative there even if a percentage is lost to piracy.

So putting this all together, I'd surmise that it is really down to two major points: lack of documentation from the start and lack of a cost reduced model at the start of the Commodore price war. Those two points would certainly fall under the fault of management.

That all being said, the one chance that Atari had to fix this was the 1200XL, but it was less than spectacular in comparison to the wonderful 800 and was also perceived as too expensive at release. If they had released the 800XL instead at that time (early 1983), I think things would have been different, but that didn't get announced until summer 1983.

My two cents :-)

  • Lack of a cost reduced model at the start of the Commodore price war, yes, I can see that. Lack of documentation from the start, that is true, they were still figuring out the difference between the closed console platform and the open computer platform. Are you saying that led to a lack of third-party games, which in turn made the computers less attractive to consumers, relative to the Vic?
    – rwallace
    Feb 12, 2020 at 1:15
  • 1
    @rwallace from what I've read/heard, documentation certainly hurt the platform at first since the "home programmer" simply didn't have the ability to make anything like Star Raiders. If it wasn't the cause, it certainly contributed to it. Perhaps the 5200 confused the market as well. But ultimately, I think it came down to cost of entry with the platform and lack of a compelling advantage over the C64 (to the less technical buyer) that did it - then it was just a game of market share which drove it out.
    – bjb
    Feb 12, 2020 at 13:39

Historical perspective: IBM announced their PC in late 1981, Commodore announced the C-64 in late 1982, just before Atari destroyed their retail channel relationships by grossly overfilling the channel with bad product (E.T., et.al.)

So the low-end retail channel was taken over by Commodore, and the higher-end retail channels were in the process of switching from Apple to IBM, leaving a rapidly shrinking middling retail presence for Atari. Lower sales and higher marketing costs than targeted probably ate their profit margins.

In part, it may have been due to Atari's (and Jay Miner's) original design target being at or above the Apple II+ instead directly at the VIC-20 in terms of general capability and pricing. e.g. the 800 ended up having been designed for a different market segment than it ended up actually competing in.


Having lived through this time period, my first computer in 1983 was the vic-20 because the price was affordable for a young sailor in the Navy. By 1984, my mom had bought a Commodore 64 because of the price.

One part that seems to have been absent from the discussion is the rise of the clones. From roughly 1985 onward there is little chance for any home computer to survive that was not intel based and running DOS as the operating system.

Evidence of this is the publication in 1984 of "Computer Shopper." In 1984 it was running about 350 pages each month. by the late 1980s it was running at about 600 pages and by 1990 it peaked at 1000 pages of adds for parts and complete computers. All priced lower than the Apple products and all priced lower than the other computers like the Atari and the Radio Shack TSR-80.

On top of that they were all running the same operating system MS-DOS or a flavor of DOS. Dozens of software titles were coming out every month that all ran on a DOS clone. It is that wave of machines from all of those manufacturers that ultimately did in all of the computers that were not running DOS. The exception is the Macintosh because it had the GUI and people were having their first contact with computers at school. Once Windows 3.1.1 came out the wave almost overwhelmed Apple too.

In 1986-87 the market was being filled with dozens of hardware manufacturers and it will be another 10 years before the massive consolidation takes place among the manufacturers. Yet through all of that Microsoft just kept growing and growing. There was a reason that Bill Gates for a long time he was the richest man on earth. He rode the crest of the clone wave.

  • 3
    Windows didn't "almost overwhelm Apple", it put Apple out of business. Remember Apple went bankrupt in 1997, and Microsoft spent over $150million to bail them out. So without Bill and Microsoft, there wouldn't be iPhones. For anyone interested, google "Apple Bankruptsy" - it's kind of an interesting story...
    – Decker
    Feb 14, 2020 at 20:48
  • 1
    Your right, I forgot about that part. Of course Bill had his own motives too. At the time there were already public statements that Microsoft was going to be sued by the feds for monopoly practices. He probably helped Apple to avoid even stiffer penalties from the feds.
    – Ken Banks
    Feb 14, 2020 at 22:59
  • 3
    "From roughly 1985 onward there is little chance for any home computer to survive that was not intel based and running DOS as the operating system." -- In the late 80s, the Amiga and the Atari ST did quite well, given the PC clone's high prices and their not being well suited for games. From the early 90s onwards I'd agree. May 17, 2020 at 0:40
  • @Decker Apple didn't go bankrupt. It was very close in 1997 but was saved by the deal with Microsoft before they had to file.
    – JeremyP
    Oct 18, 2021 at 13:23

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .