Why can't the Atari 2600 display better graphics?

[maiki]

I'm no adept at 2600 programming or the technical details, but it just blows me away at how much of a versatile machine the 2600 really is.I mean look at the 1st games done on the 2600, then look at the fantastic home brews of today.Another question would be,"What other programming tricks can be done on the 2600 that improve graphics, sound etc.?", has the 2600 reached it's limit?It seems this machine never ceases to amaze, truly an amazing piece of video gaming history.The 2600 really is the KING of video gaming consoles IMO.

 

Rather than a king of videogames, which I doubt it is, Atari 2600 is really the system that shows if the person can actually program something or not / or should I say create something from zero. Apart from these days, when there is that Batari Basic thing. I mean if you put that amount of effort into programming let's say Sega Genesis you would get results not seen before on that system, since noone really had to invent things on Genesis when they were producing games. Just about anything was ready for you as for the rendering core.

Edited May 19, 2011 by maiki

[maiki]

It is banking, tho...and you could theoretically redo some of the superchip etc. games by throwing a virtually unlimited (well, more than you could ever use) amount of rom banks at the problem areas. Need the fine scrolling mountians of Defender II without wasting cycles? Just bank to a kernel dedicated to drawing the mountian at that specific location...that sort of thing. (bad example I know...since the game doesn't really rely on the SC's ram to do this).

 

How many banks of ROM is a game allowed before it ceases to be a 2600 game?

 

I didn't know the bankswitching could be used to enhance rendering directly, I was thinking about it as a "larger" floppy disc connected to a system. Yes, this is another sort of cheating then. 4k max.

Edited May 19, 2011 by maiki

[ilovethevectrex]

great question. i was wondering that to. wikipedia says that this console is 8 bit... but then why are nes graphics better? THIS MAKES NO SENCE!!!

[+SpiceWare]

great question. i was wondering that to. wikipedia says that this console is 8 bit... but then why are nes graphics better? THIS MAKES NO SENCE!!!

8 bit refers to the data bus, not the graphics quality. An 8 bit data bus can pass values ranging from 0-255 between the various components in the system.

 

If you look at the original IBM PC, the same computer could display MDA (monochrome), CGA (16 color graphics) EGA(16 out of 64 colors) or VGA (256 out of 262,144 colors) just by using a different video card. Swapping out the video card does not change the data bus, and thus does not change the "bit size" of the computer.

Edited December 16, 2011 by SpiceWare

[+Gemintronic]

The original poster thought he was asking a stupid question. Now that I think of it, I've been stupid to NOT ask this question myself. When I was a kid the 2600 graphics were what they were. Sure, Activision games seemed hi-res, but overall it was a matter of style to me. 2600 games were immediately identifiable compared to the INTV, PC or NES. Now that I know HOW the graphics were drawn and what Atari originally intended I've just been in awe of what's possible.

 

The notable exception is Pac Man. Those graphics have always sucked hot nasty Donkey Kongs.

[potatohead]

The bus simply defines the number of concurrent binary digits that can move through the system. Reading a number, writing a value, etc... all generally happen on a "per number" basis, and the bit size of the bus then determines how big of a number that is.

 

8 bits is really small! Numbers from 0 - 255. That's it. If a bigger number is needed, say 33000, then that's two operations on an 8bit system, where it would be only one on a 16 bit system, generally speaking. (there are always lots of exceptions and edge cases on this kind of thing)

 

Graphics are represented in most systems by numbers! The numbers in binary have 1 and 0 states that all combine to be pixels, on, off, colored.

 

Bits per pixel determines the number of colors available to differentiate pixels. If there is only one bit per pixel, then there are only two colors. On is one color, off is another color.

 

4 bits per pixel would be 16 colors, 2 bits per pixel would be 4 colors.

 

Imagine a sprite on the screen. For the NES, there are multiple bits per pixel. This means the ability to have 4 color sprites, for example. On the 2600, there is only one bit per pixel for sprites, meaning single color sprites. To do NES Mario on the 2600 requires "stacking" the sprites on top of one another so that there are enough colors in that spot on the screen. 2600 only has a few sprite objects and they get used up really quick! The NES has many sprite objects, and that right there has a huge impact on the screen detail one can see.

 

When moving sprites, the bus bit size has an impact on the numbers needed to describe the sprites, position, background graphics, etc... Because the NES has more graphics objects in hardware, more can be done with fewer numbers moving across the bus. So there is that difference.

 

Graphics also generally require a buffer of sorts. The programmer puts the graphics desired into the buffer, and the graphics hardware reads that buffer, drawing the graphics to the display screen, one line at a time. The NES has a buffer, or video memory area where graphics information can be stored. In other words, move all the numbers needed, over that 8 bit bus, to the graphics memory area, then relax as it simply displays them.

 

The 2600 HAS NO BUFFER!! Well, not entirely true. It's got a little tiny one for the playfield, and it's only big enough for one line at a time. Because there is no buffer, and because there are very few hardware objects available, the VCS programmer has to send the graphics to the screen ONE LINE AT A TIME, as they are needed always, constantly never missing a beat. There is no relax period at all, because there is no storage to hold the graphics long enough to do something else.

 

Now that 8 bit bus is working double overtime! All the graphics are streaming over the bus, to the display chip, which then puts them on the TV right then and there.

 

Because of this, the NES generally has a lot more time for game computations than the VCS does.

 

All of those factors together highlight the big differences between the two. The CPU in both of them is very similar. The bus it talks to the various parts on the machine is very similar too. But, nothing else is. The simple hardware of the VCS demands more work be done more of the time, simplifying the possible graphics and game logic, compared to the NES which has more support hardware attached to it's bus, freeing it for more complex tasks.

 

It's not all that simple, but maybe this helps to visualize how things can be different.

 

One last thing!

 

The NES has lots of pins to talk to it's cartridge. Lots of pins means lots of memory addresses, meaning more data. Bigger games, more colors, more sounds, sprites, etc...

 

The VCS has only a few pins to talk to it's cartridge, meaning smaller games, sprites, etc...

[high voltage]

I'm no adept at 2600 programming or the technical details, but it just blows me away at how much of a versatile machine the 2600 really is.I mean look at the 1st games done on the 2600, then look at the fantastic home brews of today.Another question would be,"What other programming tricks can be done on the 2600 that improve graphics, sound etc.?", has the 2600 reached it's limit?It seems this machine never ceases to amaze, truly an amazing piece of video gaming history.The 2600 really is the KING of video gaming consoles IMO.

 

Rather than a king of videogames, which I doubt it is, Atari 2600 is really the system that shows if the person can actually program something or not / or should I say create something from zero. Apart from these days, when there is that Batari Basic thing. I mean if you put that amount of effort into programming let's say Sega Genesis you would get results not seen before on that system, since noone really had to invent things on Genesis when they were producing games. Just about anything was ready for you as for the rendering core.

 

I also think the VCS is KING if not, KAISER of video games.

[Amstari]

The bus simply defines the number of concurrent........

(not going to quote the whole thing )

 

That was a great explanation potatohead!

[Keatah]

Just about anything after the NES has tons of untapped potential, IF the programmers were good enough and worked with the efficiency they did when programming the 2600. It's always been said that if you can program usefulness out of the Apollo CSM/Lunar Module computer or the Atari 2600, you can learn to program anything else.

[SeaGtGruff]

It isn't the data bus that determines how "good" or "bad" the 2600's graphics can be-- it's the TIA's graphics registers. If the TIA had more player registers, color registers, missile registers, playfield registers (i.e., if each playfield bit controlled a smaller pixel), and so on, then the graphics could be "better"-- more objects on the screen, more colors on the screen, and playfields with more detail (smaller pixels). And by "on the screen," I really mean "on a scan line."

 

But the 2600 has no graphics memory per se, just registers, so the only way to get different graphics on each scan line is to write different values to the registers from one scan line to the next, and that takes time. As it is, there isn't really enough time to update all of the existing graphics registers between lines, even if we include the cycles while the scan line is actively being drawn.

 

If there were enough playfield registers to cover the entire width of the screen without having to update the registers at just the right moments to get an asymmetrical playfield, that would help some-- but you'd still need to update the registers sometime, so you'd just be changing the window of time when it's "safe" to update a given playfield register, thus you wouldn't be getting more "free cycles" to work with.

 

If the 2600 had graphics memory (like the Atari 8-bit computers do), you could load up the memory with the shapes and colors you wanted-- except then you'd need another chip that's dedicated to moving the graphical information from the memory into the actual registers (like the Atari 8-bit computers do with the ANTIC chip).

 

So you could argue that the 2600 has just about the right number of graphics registers in light of its CPU speed and lack of a dedicated graphics co-processor chip. Yes, it has a primitive design that limits how fancy its graphics can be-- but it was created at a time when chips weren't as cheap to buy as they became just a few years later, and it was intended to have a relatively short shelf life and then get replaced by a fancier video game console. Atari had no idea the 2600 would end up having such a long shelf life. They *tried* to replace it with fancier consoles, but a lot of people kept playing their 2600s and buying new 2600 games, rather than replacing their 2600s with a 5200, or a 7800, etc. So in some ways the 2600 became a victim of its own runaway success.

 

The amazing thing is that-- despite the "crudity" or "primitiveness" of its design and its consequent graphical limitations-- the 2600's design also gives it such incredible flixibility that even today we're still seeing new 2600 games that push its graphics to new heights. The "Boulder Dash" port is a perfect example of this.

[+Andrew Davie]

The "Boulder Dash" port is a perfect example of this.

 

At this point I'd like to point out that Boulder Dash® has no flicker whatsoever, and it achieves that by drawing everything except for Rockford using playfield graphics.

The only "extra" is some RAM, mostly used for the rather large playing field. The rest is magic.

Cheers

A

[SeaGtGruff]

The rest is magic.

You left out all the years of hard work that were put into it! It's a fantastic achievement, and I for one am glad you kept at it and were able to finally finish it. Much-deserved kudos to you and Thomas!

 

The fact you're drawing almost everything with the 2600's playfield graphics just goes to show that lo-resolution graphics don't preclude fun and challenging games. It isn't a matter of how advanced or detailed the graphics are-- it's more a matter of what you do with them.

[ilovethevectrex]

great question. i was wondering that to. wikipedia says that this console is 8 bit... but then why are nes graphics better? THIS MAKES NO SENCE!!!

8 bit refers to the data bus, not the graphics quality. An 8 bit data bus can pass values ranging from 0-255 between the various components in the system.

 

If you look at the original IBM PC, the same computer could display MDA (monochrome), CGA (16 color graphics) EGA(16 out of 64 colors) or VGA (256 out of 262,144 colors) just by using a different video card. Swapping out the video card does not change the data bus, and thus does not change the "bit size" of the computer.

ah i understand. i always thought it just meant how good the graphics were

[Aloan]

I personally love the blocky LEGO graphics of the Atari 2600!!! It is its charm! if it were high res, it wouldn't feel the same!

http://aloan.site90.net/atari_graphics_simplified.html

[Inky]

Something I've wondered, and has probably been answered, I keep hearing talk of how the 2600 can do this that or the other with more memory. What about (theoretically) putting in a second cpu? Would that have any practical benefit?

[Keatah]

Like a 2nd 6507, somehow, onto the main board? I don't know about why it hasn't been done or talked about. I would suspect and guess the complexity of doing so, the bottlenecks, and the amount of awesomeness returned aren't in balance.

 

As far as a co-processor goes, one need not look further than ARM enhanced games built around Harmony/Melody.

[Keatah]

This is Robot Tank, released in 1983. Look at the extreme detail. The background isn't pixelated like it is in Double Dunk, and the Activision logo looks perfect. If a game released in 1983 has decent graphics, why can't a game released 4 years later have it?

 

It can, it's a matter of the programmer actually doing it.

 

 

Anyway, those are all my questions. I know the Atari hardware is limited, but it seems like part of the reason the graphics on these games were so bad was because they weren't work on at full potential. Anyone know? Thanks in return.

 

I think you answered it yourself. Slack and laziness on the developers part. It also takes real hard time and work to make a good VCS game. A game whipped up in a week or programmed in a high-level language will play bad and look cheap.

[Supergun]

I just stumbled upon this thread, and I completely agree with the original topic starter regarding the 2600 Centipede title screen. I felt the same way the first time I saw it as well 30+ years ago. You can give me all the factual and technical explanations you want, and at the time, limited by 1982 technology and cost efficiency, I will absolutely accept it. But as for present day and/or future technology, I see absolutely no reason why it could not eventually be done.

Also...

If you can plug it into the Atari VCS cartridge slot and play it, it's an Atari 2600 game. Doesn't matter if it has 20 gigabytes of RAM and a magical half unicorn fairy living inside of the cartridge.

 

The above statement is not only 100% factual & accurate, but it also perfectly describes & encompasses the ongoing controversy among video game collectors surrounding what does & does not constitute a true complete collection for any given game console. In the case of Atari carts, the use of bank switching, (larger roms), extra ram, special chips, (like sara & DPC), etc. may be viewed as "cheating" by some old school purist programmers, but if it plugs & plays on a stock 1977 console, and was available for purchase during the systems supported lifespan, then it counts as an official cart. Period! Granted, I understand the spirit of what their saying, the 4K pac-man homebrew effort impresses more then say a 32k bank switched extra ram version of pacman might.

 

As for the supercharger, while I agree that the supercharger's 12 games do not count as official atari carts, and never will, I see no reason (especially given today's modern technology where memory is no longer a limitation) why all the games couldn't be implemented into independent stand alone carts like the Rabbit Transit game was. (and save your breath if you plan to attack that statement because at worst, one could purchase 12 Harmony cartridges, and slap 12 homemade labels on them & there you have it)

 

I think the most recent example that comes to mind for me was the SNES Road Avenger (Road Blaster FX) game that was programmed & implemented onto a snes cartridge. Granted, it's a whopping 6,320 Mbit, and the largest SNES carts were 48Mbit. Granted, it may be spendy at $200 to buy a snes SD cart and a memory card for just that one game, but it is a stand alone cartridge, and it does plug into a stock snes and plays!

 

So is using today's available extra (virtually unlimited) memory to make games larger or better for older systems "cheating"? Yes, if you were trying to count the game as an official part of the released collection, because we are well passed the supported lifespan of the system. But it is not cheating if your merely making a homebrew game.

 

So while I agree that plugging in a specially modified Atari 2600 cartridge with wires & cables coming out of it, which then hook up and connect to and interface with an external independently powered laser disc player in order to allow someone to play Dragon's Lair on the Atari 2600 is cheating and does not count as an actual Atari "cartridge" game per say, if however, when the technology allows you to micro miniaturize all of that to fit inside a cartridge shell, then YES, it counts, and it's not cheating!

[Supergun]

 

It can, it's a matter of the programmer actually doing it.

 

 

 

I think you answered it yourself. Slack and laziness on the developers part. It also takes real hard time and work to make a good VCS game. A game whipped up in a week or programmed in a high-level language will play bad and look cheap.

 

For the record, in an interview, the programmer of Double Dunk admitted himself that he prefers the original one on one 2600 Basketball to his own game! (as do I)

[ZylonBane]

It can, it's a matter of the programmer actually doing it.

I think you answered it yourself. Slack and laziness on the developers part. It also takes real hard time and work to make a good VCS game. A game whipped up in a week or programmed in a high-level language will play bad and look cheap.

 

You do realize you're responding to a post that was written FOUR YEARS AGO, yes?

[Keatah]

You do realize you're responding to a post that was written FOUR YEARS AGO, yes?

 

And posted to on Mon Aug 31, 2015 10:58 PM

Go pick on someone else.

 

ADDED:

This is a retrogame thread and forum. Old is cool. Any question and any answer is valid regardless of when it was asked and/or responded to.

 

If you feel the need (which I don't believe you do) to clean up this thread, apply for moderator status and begin deleting with post number 64 onwards.

Edited September 3, 2015 by Keatah

[+SpiceWare]

And posted to on Mon Aug 31, 2015 10:58 PM

Go pick on someone else.

I see he's still trying to dictate to others on the acceptable uses of their time. I added him to my ignore list back in 2013 for that reason.

[Thomas Jentzsch]

I see he's still trying to dictate to others on the acceptable uses of their time. I added him to my ignore list back in 2013 for that reason.

But then you miss a nice troll training opportunity.

[AtariTexas]

I hope I'm not on anyone's ignore list. I love you all !!

[+SpiceWare]

But then you miss a nice troll training opportunity.

Have better things to do than that, like work on Bus Stuffing. This week's been a bit hectic though so I've yet to start on the demo.

 

I do on occasion click the View it anyway? link to see if my list can be shortened - removed somebody from it yesterday in fact as they were participating nicely in a discussion. Last time I clicked on one for ZylonBane it resulted in this reply.