7800 - what did Atari wrong?

[zzip]

31 minutes ago, SmittyB said:

The 7800's graphics are really built around the 160 modes with (as is my understanding) the 320 modes being not so much afterthoughts but extra modes that were easy to implement as extensions of the 160 modes.

That's why all of the object placement happens with a resolution of 160, and why the transparency doesn't work as expected in 320 modes.

thanks for the clarity.   It's a shame they couldn't have made a more usable 320 mode, because with the release of NES/Amiga/ST,  160 mode was looking very dated by 1986

 

33 minutes ago, SmittyB said:

As for mixing graphics modes, yes it's possible and several games do this. It's done by changing it mid-screen but this means waiting until the right moment and that's time that could be used for processing other things.

Interesting.  Antic Display Lists on the 5200/8bit make this super-simple.  But if you have to waste CPU cycles on the 7800 trying to time out the mode change, it probably isn't worth doing in most cases.

[RevEng]

50 minutes ago, zzip said:

Interesting.  Antic Display Lists on the 5200/8bit make this super-simple.  But if you have to waste CPU cycles on the 7800 trying to time out the mode change, it probably isn't worth doing in most cases.

The mode can be changed with a display interrupt on the 7800, but the trade-off is that you can't then push DMA near the maximum, or else the interrupt won't reliably change the mode in time. Lots of 7800 games change modes mid-screen via interrupt, but there is one disadvantage compared to the A8 here - the sprites are either in one mode or another, and if they vertically cross into a mode they weren't designed for, they display weird. PM graphics don't suffer this problem. So typically the 7800 mode change is just used for title screens, score areas, etc.

 

One change I made with the 7800 Serpentine port was to use 160 mode graphics for the score area, instead of the 320 (antic 2) in the A8 version... reason being the 160a frog hops up there on occasion.

[TailChao]

3 hours ago, zzip said:

I forget the exact issues with the 7800's 320 mode,  but in general it has color limitations that makes using it a pain for many types of games.

The color limitations aren't that bad, in my opinion the bandwidth requirements are much more of a problem. Going from 160A to 320B doubles the number of pixels to move which effectively halves your fillrate. Using 320A will net you identical performance but is much trickier to stylize - you'll likely have to layer draws to get appealing visuals. But in any mode you're always dealing with Maria eating Sally's cycles.

 

Adding more RAM can help simplify Display List management significantly, just from being able to cache them. They take significant CPU time to generate and if you have mapper hardware which helps partition them into different segments (i.e. background and foreground) that's also a huge boost.

 

My ideal would be nearing MMC5 level craziness where you'd have a mapper negotiate all interactions with the cartridge's EXRAM and be able to "nudge" select Display Lists' horizontal positions. So then you could just write to a register to move your entire backdrop left or right by (let's say) eight pixels. Drop the fuss of horizontal scrolling to nearly zero.

[Greg2600]

I've always read that the 7800 scrolls poorly, and that combined with the advantage in larger sprites, I wonder if the hardware design was simply to maximize the arcade ports of the day?  Parallax scrolling was almost unheard of at the time.

[Geoff Oltmans]

On 7/13/2020 at 9:24 PM, Greg2600 said:

Wholesale staff changes is not a shocker, but Jack didn't have the kind of capital that Warners did nor his competitors.  That being said, as I mentioned, he thought he was getting a fully ready to go retail product in the 7800 and instead walked into a firefight between Warner and GCC.  Not sure what Epyx could blame on Tramiel?  That company shot itself in the foot time after time.  We all know his main interest in Atari was the home computer division, which was his background.  However, while I once viewed Tramiel as the villain, yet with proper research he was simply the victim of buying a veritable lemon of a company.  Let's face it, they did okay with the ST computer line, before being crushed by competitors who adopted Microsoft's OS.  However, the Lynx was a bargain basement development in that it was finished when they bought it.  The 7800 was a crippled system when released in several ways.  Less "tightwad" and just simply lack of revenue. 

Well, if the Lynx FAQ is to be believed then Atari stiffed them a sum of money for royalties that they took them to court over and several of the Epyx folks didn't sound too happy with Atari.

[zzip]

On 7/13/2020 at 10:24 PM, Greg2600 said:

Wholesale staff changes is not a shocker, but Jack didn't have the kind of capital that Warners did nor his competitors.  That being said, as I mentioned, he thought he was getting a fully ready to go retail product in the 7800 and instead walked into a firefight between Warner and GCC.  Not sure what Epyx could blame on Tramiel?  That company shot itself in the foot time after time.  We all know his main interest in Atari was the home computer division, which was his background.  However, while I once viewed Tramiel as the villain, yet with proper research he was simply the victim of buying a veritable lemon of a company.  Let's face it, they did okay with the ST computer line, before being crushed by competitors who adopted Microsoft's OS.  However, the Lynx was a bargain basement development in that it was finished when they bought it.  The 7800 was a crippled system when released in several ways.  Less "tightwad" and just simply lack of revenue. 

Yeah part of the problem was Tramiel Atari didn't seem to have enough capital to continue to be a big player in Videogames, unlike Warner, which is why they always seemed to cheap out.   Obviously the company was in bad financial shape when he bought it and he had to restructure it.   Warner would have had to continue to make hard choices had they held onto it.   But I think once the console business recovered, Warner would have still been in a better position to capitalize on it than Tramiel was.

 

Also I think they viewed the videogame division as a cash cow they could tap into to help fund the computer vision.   They didn't put enough money into R&D, they didn't put enough into developing new IPs, they were happy to repackage and sell what had been developed previously along with a library of games that were last popular in 1983 well into the late 80s.

[Greg2600]

3 hours ago, Geoff Oltmans said:

Well, if the Lynx FAQ is to be believed then Atari stiffed them a sum of money for royalties that they took them to court over and several of the Epyx folks didn't sound too happy with Atari.

Atari Corp's position was that Epyx failed to deliver on project timelines with Lynx software.  As a result, they held back payment to Epyx which obviously had that company scrambling.  Was it the greatest idea for Tramiel to punish his first party software developer?  Probably not, but I'm sure these penalties were quite clearly established in the contract.  This is not uncommon in business, if your supplier fails to deliver on time, you don't pay them.  Tramiel was not a good CEO, he was too much of a micromanager.

 

2 hours ago, zzip said:

Yeah part of the problem was Tramiel Atari didn't seem to have enough capital to continue to be a big player in Videogames, unlike Warner, which is why they always seemed to cheap out.   Obviously the company was in bad financial shape when he bought it and he had to restructure it.   Warner would have had to continue to make hard choices had they held onto it.   But I think once the console business recovered, Warner would have still been in a better position to capitalize on it than Tramiel was.

 

Also I think they viewed the videogame division as a cash cow they could tap into to help fund the computer vision.   They didn't put enough money into R&D, they didn't put enough into developing new IPs, they were happy to repackage and sell what had been developed previously along with a library of games that were last popular in 1983 well into the late 80s.

Warner Bros. ran Atari into the ground, so had they not sold the company, it would probably have been out of business by the time you first heard of Link or Samus.  They were losing astounding amounts of money.  Tramiel closed some 80 Atari locations around the world, he had no choice.  Atari was sued by Walt Disney, Steven Spielberg, and other partners for failing to live up to contracts to produce games.  Commodore sued Tramiel and new employees for "trade secrets."  Some of these bad deals were covered by WB, like Alan Alda's $10 million dollar spokesman contract!

Jack liquidated the inventory to pay for the 16-bit Atari ST platform development, which was to compete obviously with Amiga for many years.  He and his partners basically spent around 150 million for 325 million in assets all told.  Again, his focus was on computers not consoles.  As I said above, he was the worst kind of CEO for a company producing retail toy products.  He definitely skimped, and was not compatible with that marketplace like he was the more mature, business-oriented computer scene.

[+Pat Brady]

2 hours ago, Greg2600 said:

As I said above, he was the worst kind of CEO for a company producing retail toy products.  He definitely skimped, and was not compatible with that marketplace like he was the more mature, business-oriented computer scene.

 

Tramiel gets a bad rap around here, but I think the worst kind of CEO for this sort of company would be one that lays off the hardware engineering team, undervalues and antagonizes the software developers, and ruthlessly eliminates any loyalty from retailers and customers. Under this kind of CEO, the company might make a lot of money for a short duration but will then face total ruin.

[Frank.Bullitt]

15 hours ago, bizarrostormy said:

Tramiel gets a bad rap around here, but I think the worst kind of CEO for this sort of company would be one that lays off the hardware engineering team, undervalues and antagonizes the software developers, and ruthlessly eliminates any loyalty from retailers and customers. Under this kind of CEO, the company might make a lot of money for a short duration but will then face total ruin.

I watched a YouTube documentary about him by KimJustice a while back. He was quite the interesting guy. I had no idea he was a Holocaust survivor.

 

There's also one on Robert Maxwell that I'd recommend too.

[ZylonBane]

On 7/14/2020 at 9:31 AM, zzip said:

One example is Karateka.  On the Atari 8-bit, it uses 320 mode with color artifcating.

Wrong. 8-bit Atari Karateka uses the 160 wide 4-color mode.

 

 

[zzip]

On 7/18/2020 at 2:41 PM, ZylonBane said:

Wrong. 8-bit Atari Karateka uses the 160 wide 4-color mode.

Interesting.. the character has blue pixels at its edges like it would if using artifacting for colors.  Also the title and intro screens are 320 and I guess that's why I assumed the entire game was.   I still think it looks better than the 7800 version

Edited July 20, 2020 by zzip

[patroclus99]

I actually think that I wouldn't have sold Atari if I had just taken a vacation. . . all the sudden someone comes along and offers you more money than you thought you'd ever be worth, and you say, "Eh, why not?"

--Nolan Bushnell

 

 

In my opinion, Atari's decline began with its rise; Bushnell was an incredibly talented innovator in electronics; but he preferred an RV-like career itinerary in which he traveled from one pursuit to another.

 

He had business savvy:  he understood how to market his product; he knew how to corner supply lines to get the best price and maintain a successful presence in the console-market; to a certain extent, he understood the relationship between incentive and productivity; and he had a primal/archetype understanding of the gaming demographic. . . in a time in which there were no "gamers".

 

I believe if he really wanted to, he could've attracted investors, collaborative arrangements, and partnerships to launch an IPO; Pong had been an incredible success for over 3 years, during which time I'm sure he had established himself predominantly in the budding arcade-electronics industry.
But, I really don't think he had a passion, or really an interest, for VCS development or distribution; it wasn't fun anymore.  This is someone who always wanted to work at Disneyland, who liked setting up his coin-ops in bars, pool-halls, restaurants, and whatnot.   And really, I think he was far more interested in robotics than electrical engineering.  I think he liked the three-dimensional relationship of electronics and entertainment, hence Pizza Time.  Designing a machine to execute very rudimentary programs to bounce things around on a TV screen just bored him; couple this with having to run a division of electronics and software development for a corporation who didn't know shit from Shinola about computing.

And so, you have a brilliant innovator, who is whimsical about his commitments, leaves a company who doesn't really have a clue.  This leaves a vacuum of leadership and managerial technical competency replaced by operations-officers, managers, or other insignificant MBA minutiae, during which time  I think most of the game-developers, artists, etc. there were just doing their thing, clock-punching the day away until the next memo. . . or until they took some initiative of their own.

 

The Atari that I remember, the heavy-sixer on which my grandfather and I played played Video Bowling and Golf. . . by myself, Yar's Revenge, Video Pinball, Adventure, Night Driver, etc. . . Bushnell is still there, I think, but he's also, presumably, deeply preoccupied with Pizza Time in which Warner wasn't interested.  Was he even actively engaged in game-development at Warner?  Was he even at work most days?   I remember reading an article where he maneuvered Jobs and 'Woz' to finish Breakout (CX2622) expeditiously; but the VCS was released in '77-'78, Bushnell left in '78.  He was out of the picture when we were actually playing these games.

 

Ostensibly, Atari floundered because Nolan didn't take a vacation. . .  zzz.

 

As far as the 5200, 7800, 400/800, XE/XEGS, Lynx, Jag, 520/1040 ST. . .  I'm sure I left out something. . . these devices were doomed before they were even conceived.  If you think about it, the whole damn thing was just a big fluke.

 

 

[Goochman]

I dont think Atari had a lack of Technical Competency - they were overflowing with sharp technical talent.  However they lacked vision and the ability to make good use of this talent.  Bushnell had that in spades, but if he didnt sell Atari it wouldve died on the vine from a revenue standpoint.  Warner couldnt have picked a worse candidate than Ray to run the company..............

[The Mr. Video]

I think the big reason why the 7800 failed was because Atari was too slow to release the 7800. They test marketed the 7800 in 1984, but didn't give it a wide release until '86. By then NES was getting a nationwide release and was catching on like wildfire. Another factor was their lack of third-party developers. Nintendo's policy of not being able to develop games for competing systems upon becoming a third-party developer allowed them to snatch up all of the prestigious developers and leave Atari and Sega next to nothing, and while Sega managed to mostly get by with their own releases, Atari just couldn't compensate.

[8th lutz]

23 hours ago, The Mr. Video said:

I think the big reason why the 7800 failed was because Atari was too slow to release the 7800. They test marketed the 7800 in 1984, but didn't give it a wide release until '86. By then NES was getting a nationwide release and was catching on like wildfire. Another factor was their lack of third-party developers. Nintendo's policy of not being able to develop games for competing systems upon becoming a third-party developer allowed them to snatch up all of the prestigious developers and leave Atari and Sega next to nothing, and while Sega managed to mostly get by with their own releases, Atari just couldn't compensate.

Atari was slow, but it was complicated on why. The owner of Atari Inc. sold Tramiel the Assets of Atari Inc's Home Consumer division on July 2nd, 1984. The problem was Warner Communications did not pay GCC for development of the 7800. There was a heated argument for who was supposed to pay GCC with taking Tramiel May of 1985 to pay GCC. Tramiel than had to do additional negotiations for the initial launch titles that GCC did. Tramiel did not have a video game division either at the time. All Tramiel was doing was selling leftover inventory of the 2600 from a video game retail side of things before the deals with GCC with the Atari 7800 were. 

 

When Tramiel bought the Atari Inc. Consumer division assets from Warner, Tramiel had his own company. Tramiel renamed his own company Atari Corporation. Tramiel did not have the deepest pockets and could keep a lot of Atari Inc. employees that were under contract for Warner.  Those pockets did affect the cartridge size for the 7800.

 

The Nes had a worldwide launch in September of 1986 and the 7800 Nationwide  launch happened May of 1986. That means the 7800 Nationwide was released. The Nes did a bunch of test markets in 1985 and 1986 before doing the Nationwide. The test market were at the largest cities in the United States in 1985 and 1986.

 

The Nes was test Marketed in New York City in 1985.  Los Angeles was the 2nd NES  test market with it being February of 1986.  Chicago and San Francisco were the 3rd and 4th test markets. The NES than was test marketed in another 12 top United States Markets before doing the Nationwide launch in September of 1986. 

 

Doing to the time frame, It is given the 7800 was released in a lot places before the NES. The 7800's problem was the games themselves in 1986. You are dealing with games that were supposed to be released for the 1984 launch, not a 1986 Nationwide release. 

 

Even without Nintendo's 3rd party policy, Atari 7800 was not going to get a lot of  3 party games. The 7800 was not released in Japan and that is going to affect if any Japanese developers want to develop and publish games outside of Japan only. The Master System would've been a big benefactor without Nintendo's 3rd party policy. I think Atari could've got Borderbund Software and Tengen for the 7800 as 3rd parties.

 

Edited July 31, 2020 by 8th lutz

[Zonie]

Don't get me wrong, I love the 7800. It's my favorite console. But in 82, all of my peers and I saw the 5200 for what it was right away the very first time we saw it in a store. A 400 without a keyboard and a goofy controller. It was my contention that Atari should have just taken the 400, Painted it black, gotten rid of the cartridge lid, added Ram, perhaps give us a two button joystick, and called it done. Basically give us a membrane keyboard version of the XEGS then, and do it for the 5200's price.

 

If they had done that, they would have been ahead of the game, the existing line of games both cart and disk would be compatible, and with the SIO port, expandable to no end. People already knew how to code for it, and the 5200 debacle would never have happened. A better 400 was all we needed.

 

The XEGS was a great idea. It just came too late, and confused the market. Too bad.

[The Usotsuki]

The XEGS is what the 5200 should have been.

[Zonie]

12 hours ago, The Usotsuki said:

The XEGS is what the 5200 should have been.

Yep. Pretty much what I was trying to say.

 

[Greg2600]

Look in hindsight Atari should have done the Nintendo deal and sold the NES as its system.  If not, they should have done the Sega deal, and sold the Genesis as its system. 

[zzip]

On 7/30/2020 at 8:41 PM, 8th lutz said:

Doing to the time frame, It is given the 7800 was released in a lot places before the NES. The 7800's problem was the games themselves in 1986. You are dealing with games that were supposed to be released for the 1984 launch, not a 1986 Nationwide release. 

Not just that, but a lot of these original games were already old by 1984 standards!   Dig Dug, Joust, Ms Pacman, Asteroids, etc.

[zzip]

23 hours ago, Zonie said:

Don't get me wrong, I love the 7800. It's my favorite console. But in 82, all of my peers and I saw the 5200 for what it was right away the very first time we saw it in a store. A 400 without a keyboard and a goofy controller. It was my contention that Atari should have just taken the 400, Painted it black, gotten rid of the cartridge lid, added Ram, perhaps give us a two button joystick, and called it done. Basically give us a membrane keyboard version of the XEGS then, and do it for the 5200's price.

Or sell the keyboard as an add-on.   It was around that time when every console was announcing keyboard accesories, anyway.   The 5200 could just turn into a full-fleged computer just by plugging in a keyboard, without all the hoops the other consoles were doing, because they were never designed to be computers.   The 5200 could be better than an ADAM at a fraction of the cost.

 

16 hours ago, The Usotsuki said:

The XEGS is what the 5200 should have been

I don't think they'd have 64K RAM in a console in 82 though.  Too expensive!   Maybe if it was expandable like the 600XL.   I always thought the 600XL was a better 5200 than the 5200 anyway.

[Zonie]

6 hours ago, Greg2600 said:

Look in hindsight Atari should have done the Nintendo deal and sold the NES as its system.  If not, they should have done the Sega deal, and sold the Genesis as its system. 

Good point

[Zonie]

2 hours ago, zzip said:

Or sell the keyboard as an add-on.   It was around that time when every console was announcing keyboard accesories, anyway.   The 5200 could just turn into a full-fleged computer just by plugging in a keyboard, without all the hoops the other consoles were doing, because they were never designed to be computers.   The 5200 could be better than an ADAM at a fraction of the cost.

 

I don't think they'd have 64K RAM in a console in 82 though.  Too expensive!   Maybe if it was expandable like the 600XL.   I always thought the 600XL was a better 5200 than the 5200 anyway.

No, it had to come with a keyboard for 100% game compatibility with the existing 8-bit line. True that People were making full keyboard replacements for the 400 by then, but the membrane was cheap. The O2 had an even cheaper membrane. Only Commodore was able to provide a real one on the Vic-20 because they saved so much cost having their own fab for the chips. I actually like my 400 and use it as a console.

 

As for memory, 16K was fine back then. Expandability would be even better, but that is getting into 800 territory. 

[zzip]

12 minutes ago, Zonie said:

No, it had to come with a keyboard for 100% game compatibility with the existing 8-bit line. True that People were making full keyboard replacements for the 400 by then, but the membrane was cheap. The O2 had an even cheaper membrane. Only Commodore was able to provide a real one on the Vic-20 because they saved so much cost having their own fab for the chips. I actually like my 400 and use it as a console.

They could have mapped Start/Option/Select and Space (Paulse) to buttons on the console and that would have taken care of 99% of the cartridge games I think.

 

I don't know what a pack-in keyboard would have done to the cost or retailer/consumer perception (is it a game system or computer?) But you make a good point about the O2 having one.

[Zonie]

Didn't Star Raiders use a lot of keys? It was the game I wanted one for.

But yeah, those four buttons are on the XEGS.