The BEST 8 Bit Gaming Computer

[Spector]

Fair enough - I'm not arguing that. It was just a suggestion from the C64 guy that the Spectrum was faster only because of a lack of colour, yet with many games the C64 has a similar lack of colour but is still slower. The C64 is also weaker with animated sprites, isn't it? Less frames of movement for some reason.

[potatohead]

The two share an 8 bit data bus. I'm not sure I see the distinction.

They share an 8 bit external data bus, but the 65816 certainly has a 16 bit internal data bus. I'm not sure about the 6809 though as I don't have a schematic to hand.

 

Edit: I found one and the ALU in the 6809 appears to only output to the 8 bit bus, so I'm guessing it uses a similar trick to the Z80.

 

Ok, that nails it. They are 8 bitters, in that regard.

 

Fair enough.

[potatohead]

On the matter of the best, I still say the Atari 8bitters rule because of how the video system worked. Games that were keyed to the screen display, in particular those that took advantage of the video sub-system, ended up with a crisp real-time feel, other systems did not reproduce as well.

First of all that's opinion and video generated interrupts aren't unique to the Atari. The coprocessor was but the things it did could be duplicated just with the CPU on other machines.

 

Yeah, it's opinion. More below!

 

This is interrupt, event driven programming, and the Atari really emphasizes it because of how it operates. Display list interrupts, vblank interrupts, latched inputs, etc... all combined to allow some low-level linking directly to the metal that affected the overall game design in a unique way. The 2600 is the extreme of this, with the 8bitters being a nice hybrid. On the C64, these kinds of things were not done as often because of the focus on more detailed on screen graphics.

Interrupt driven programming is pretty common in game development on all machines. Even on the C64.

 

No disagreement there. It's the overall structure of the Atari that makes a difference. Again below.

 

I suppose one could argue the types of color displays on the C64 are more general case, allowing for more colors in more parts of the graphics screen without the vertical limitations. However, that also means building your display has less overall effect on the game itself. That's where the Jay Miner magic is --and it's a superior experience when game and hardware become one for the player.

Less overall effect on the game?

You mean I wouldn't have to program around hardware limitations?

 

Yes! This is it exactly. The hardware limitations provide an interesting canvas for games. This means, some games will be just excellent. IMHO, this is largely because the game evolved around the hardware. It's the whole, "I created this cool effect, let's make a game of it." method of game conception. By nature, other machines with a different canvas are likely to reproduce the same game, but will lack some precision in general that affects the monor details of how it interacts with the player overall.

 

What's really great about the Atari is the reality that in many cases, the display is an intergal part of the game itself. It's an extreme on the VCS, but still very true on many great 8 bitter titles. When one is talking about the greatest 8 bit gaming computer, the types of games that really appeal have a lot to do with that. I've always been in the Atari camp because I believe there is more art involved to building both a great game and a great display.

 

The C64, made great displays easier for a lot of scenarios. Less art in many cases as it's simple competency in execution that matters.

 

Fast forward to today.

 

A really great game on the PlayStation, has some degree of art to it where the same title can be easily reproduced on the newer consoles with ease. The differences are not as stark as they were for the 8 bit era, but they are still there.

 

Getting back to the Atari vs other 8 bitters, the hardware design presented a lot of interesting tradeoffs that provided a rich playfield for those wanting to explore it and game on it. If one thinks this stuff is great, then the Atari was the greatest...

 

(Sorry for sloppy post, I'm outta time) Will try to get back and edit later.

[JamesD]

Fair enough - I'm not arguing that. It was just a suggestion from the C64 guy that the Spectrum was faster only because of a lack of colour, yet with many games the C64 has a similar lack of colour but is still slower. The C64 is also weaker with animated sprites, isn't it? Less frames of movement for some reason.

If there was anything lacking on the C64 color wise I would say it was the programmer's fault.

The problem with the C64 goes back to clock speed. Even if DRAM refresh, screen refresh, etc... take 30% of the Atari CPU clocks that still leaves more clock cycles for the Atari than the C64 has to begin with!

 

The C64 sprites steal some clock cycles to generate but there is certainly nothing slow about them. You can move them faster than a person can see them if you want. I'd guess something else in the games were slowing them down.

[vdub_bobby]

The C64 is also weaker with animated sprites, isn't it? Less frames of movement for some reason.

The C64 sprites steal some clock cycles to generate but there is certainly nothing slow about them. You can move them faster than a person can see them if you want. I'd guess something else in the games were slowing them down.

If there was a hardware reason for less frames of movement for animated sprites, it would probably have to do with the amount of memory, wouldn't it? And how much memory each frame of a sprite takes up?

[JamesD]

To give you an idea of what is possible on a Plus/4:

Xe03

[JamesD]

If there was a hardware reason for less frames of movement for animated sprites, it would probably have to do with the amount of memory, wouldn't it? And how much memory each frame of a sprite takes up?

Well, I'm not a C64 guru so I don't know exactly what was up... however, there are a couple ways animated sprites are implemented in hardware and how you would have to update them depends on the hardware.

1. The sprite hardware uses a fixed buffer and you copy each sprite's new frame into the buffer to display the new image.

2. The sprite's hardware uses a pointer to the buffer and you just change the pointer to the new buffer to display the new image.

 

On the first type of sprite hardware a space game with only a small amount of sprite animation could be absurdly fast, but with heavily animated characters it could slow considerably.

On the second type it doesn't matter, it's always fast.

 

Just remember, this also depends on what color mode is used too. The more colors in the sprites, the more data you have to move if you change them on type 1.

 

Now, whenever you aren't doing much sprite animation either one can be very fast but as soon as you have to move a block of data... CPU speed or a more efficient algorythm is king.

If only part of the sprite changes you could update just that part of the sprite by using a software state machine. More tests but lots less load, save, increment and loop to do to update the sprite. But I think you'll find a lot of programmers would just update the entire sprite because it's easier to do.

[MacbthPSW]

If there was a hardware reason for less frames of movement for animated sprites, it would probably have to do with the amount of memory, wouldn't it? And how much memory each frame of a sprite takes up?

Each sprite is 3 bytes (24 pixels) wide and 21 pixels tall, which works out to 63 bytes per frame. They're stored at 64 byte intervals (the extra byte "wasted" is actually pretty handy for storing e.g. colour). The VIC-II can see/use 16k of RAM at any given time (this can be pointed to any of 4 banks starting at $0000, $4000, $8000 or $C000) so that allows for as many as 256 sprite frames available at any moment.

 

In practice, 2k will be dedicated to a character set (256 8 byte characters) and 1k (or possibly 2k if the game is double-buffered) for character mode video memory, so that would leave 192-208 frames. Or if the game is using a bitmap mode, that will consume 9k of memory, leaving 112 frames.

[MacbthPSW]

2. The sprite's hardware uses a pointer to the buffer and you just change the pointer to the new buffer to display the new image.

 

On the first type of sprite hardware a space game with only a small amount of sprite animation could be absurdly fast, but with heavily animated characters it could slow considerably.

On the second type it doesn't matter, it's always fast.

The C64 uses the 2nd type - each hardware sprite has a one-byte pointer to one of the 256 available slots. A single byte store can completely change the shape of the sprite, or move it clear across the screen, or whatever. C64 sprites are very fast, and very powerful.

 

As for C64 sprite animation being slow, that's completely unfounded. Did anyone have any particular examples? Just check out any of the classics to see how good the animation is - Impossible Mission, Summer/California/Winter/Whatever Games...

[JamesD]

The C64 uses the 2nd type - each hardware sprite has a one-byte pointer to one of the 256 available slots. A single byte store can completely change the shape of the sprite, or move it clear across the screen, or whatever. C64 sprites are very fast, and very powerful.

I thought it did but I didn't want to say unless I was sure. It, the SID and memory size are why I voted for the C64 over the Atari. A faster clock would have been nice though.

 

I also think the Atari is closer to the first type. You even have to move the sprite data up and down in the Player Missile buffer to move it up and down on the screen instead of just changing the Y co-ordinate to place the sprited at. A lot of vertical sprite action or animation on screen at once can take thousands of clock cycles and the extra CPU speed is definately needed.

 

As for C64 sprite animation being slow, that's completely unfounded. Did anyone have any particular examples? Just check out any of the classics to see how good the animation is - Impossible Mission, Summer/California/Winter/Whatever Games...

I'd say if there was a limitation it was imagined or on the programmer's part.

The only way sprites should be slow is if the game logic is making a huge number of calculations... and frankly... the ONLY thing I've seen needing that number of calculations is 3D and a lot of that is in drawing and has nothing to do with sprites.

[Spector]

The C64 uses the 2nd type - each hardware sprite has a one-byte pointer to one of the 256 available slots. A single byte store can completely change the shape of the sprite, or move it clear across the screen, or whatever. C64 sprites are very fast, and very powerful.

I thought it did but I didn't want to say unless I was sure. It, the SID and memory size are why I voted for the C64 over the Atari. A faster clock would have been nice though.

 

I also think the Atari is closer to the first type. You even have to move the sprite data up and down in the Player Missile buffer to move it up and down on the screen instead of just changing the Y co-ordinate to place the sprited at. A lot of vertical sprite action or animation on screen at once can take thousands of clock cycles and the extra CPU speed is definately needed.

 

As for C64 sprite animation being slow, that's completely unfounded. Did anyone have any particular examples? Just check out any of the classics to see how good the animation is - Impossible Mission, Summer/California/Winter/Whatever Games...

I'd say if there was a limitation it was imagined or on the programmer's part.

The only way sprites should be slow is if the game logic is making a huge number of calculations... and frankly... the ONLY thing I've seen needing that number of calculations is 3D and a lot of that is in drawing and has nothing to do with sprites.

I didn't really explain myself there - I didn't mean the sprite animation was slow, I meant it was poor in many cases, in the sense that when the Spectrum woul use five or six frames of animation, the C64 would use just two or three. Rolling Thunder is a good example of how the sprite animation on the C64 version is awful compared to the Spectrum one. Just look at the effect when the main character jumps over the bannister onto the top floor and compare it to the C64 version, where he walks like Mickey Mouse.

[JamesD]

 

I didn't really explain myself there - I didn't mean the sprite animation was slow, I meant it was poor in many cases, in the sense that when the Spectrum woul use five or six frames of animation, the C64 would use just two or three. Rolling Thunder is a good example of how the sprite animation on the C64 version is awful compared to the Spectrum one. Just look at the effect when the main character jumps over the bannister onto the top floor and compare it to the C64 version, where he walks like Mickey Mouse.

Blame the artist that drew them or the programmer.

Perhaps they turned it out in a hurry or didn't want it to look as good.

There's no technical reason it couldn't be better.

[Matt_B]

Blame the artist that drew them or the programmer.

Perhaps they turned it out in a hurry or didn't want it to look as good.

There's no technical reason it couldn't be better.

 

Yes, I think we should try and judge the machines in question on their absolute best performances rather than the ones where they came rather short of the mark. Botched conversions produced by programmers who'd bitten off more than the could chew and/or working to ridiculous time limits happened on all machines. It could be a lottery which computer ended up with the best implementation of a game and rarely came down to things as mundane as the capabilities of the hardware.

[Spector]

Blame the artist that drew them or the programmer.

Perhaps they turned it out in a hurry or didn't want it to look as good.

There's no technical reason it couldn't be better.

 

Yes, I think we should try and judge the machines in question on their absolute best performances rather than the ones where they came rather short of the mark. Botched conversions produced by programmers who'd bitten off more than the could chew and/or working to ridiculous time limits happened on all machines. It could be a lottery which computer ended up with the best implementation of a game and rarely came down to things as mundane as the capabilities of the hardware.

 

Well it's true that rolling thunder wasn't the best C64 conversion ever made. But doesn't more frames mean more raw processor power, something which the C64 lacks when compared to Spectrum?

[JamesD]

Well it's true that rolling thunder wasn't the best C64 conversion ever made. But doesn't more frames mean more raw processor power, something which the C64 lacks when compared to Spectrum?

More frames on what? One game?

NO! Too many variables.

 

The speccy uses 1 bit / pixel and the X coordinate fits in 1 byte.

 

ANY machine that has a 320 pixel wide display must us 2 bytes to represent the X axis.

THE MATH ROUTINES MUST ACCOUNT FOR THIS UNLESS YOU TRIM THE DISPLAY AREA DOWN TO 256 PIXELS!

 

If a version of the game for a different computer is using a display mode with more than 2 colors it must do much more work! 4 colors = 2 bits / pixel. 16 colors = 4 bits / pixel.

 

The other machine is having to do more work than the Z80 in the Speccy.

 

Even on the same CPU the difference between screen resolutions and graphics modes would make a big difference.

You couldn't expect anything else.

[MacbthPSW]

But doesn't more frames mean more raw processor power, something which the C64 lacks when compared to Spectrum?

Changing sprite frames on the C64 doesn't involve processor power at all - like I said above, the C-64 just needs a single byte changed to display any one of (up to) 256 different frames available.

 

The Spectrum requires a lot more processor power to display a "sprite", though it's a function of how often the sprite moves or changes, not the number of frames available.

[Spector]

Well it's true that rolling thunder wasn't the best C64 conversion ever made. But doesn't more frames mean more raw processor power, something which the C64 lacks when compared to Spectrum?

More frames on what? One game?

NO! Too many variables.

 

The speccy uses 1 bit / pixel and the X coordinate fits in 1 byte.

 

ANY machine that has a 320 pixel wide display must us 2 bytes to represent the X axis.

THE MATH ROUTINES MUST ACCOUNT FOR THIS UNLESS YOU TRIM THE DISPLAY AREA DOWN TO 256 PIXELS!

 

If a version of the game for a different computer is using a display mode with more than 2 colors it must do much more work! 4 colors = 2 bits / pixel. 16 colors = 4 bits / pixel.

 

The other machine is having to do more work than the Z80 in the Speccy.

 

Even on the same CPU the difference between screen resolutions and graphics modes would make a big difference.

You couldn't expect anything else.

 

Well that's all technical stuff and I don't know about that. I'm more interested in what actually happened rather than what happens on paper. You say about differences between screen resolutions and graphics modes, but virtually all isometric games on the C64 that are in identical mono colour to the spectrum version are slower. Is that just because of the C64's bigger screen display? Because it can't be about colour.

[JamesD]

I'm more interested in what actually happened rather than what happens on paper.

Translation: You are more interested in a comparison which favors your favorite machine.

 

You say about differences between screen resolutions and graphics modes, but virtually all isometric games on the C64 that are in identical mono colour to the spectrum version are slower. Is that just because of the C64's bigger screen display? Because it can't be about colour.

Just having to support 16 bits in the math would be more than enough to slow the machine down.

 

The only way you'd ever really be able to compare them head to head is with a large number of benchmarks designed so that they wouldn't favor one or the other across the board.

[liveinabin]

Spectrum. Totally.

 

I did have a C64 as well but, although it had the graphics and the sound, the games didn't interest me as much. Perhaps it was that with it's chunky display, the machine was more locked into 2D sprite based games, where the Spectrum seemed just as happy with vectors (well, bitmapped vectors, you know what I mean).

I liked the BASIC better (anyone could write a little game, and user defined graphics were really easy), the machine was more ubiquitous (in UK anyway) which matters and I liked being able to choose my own tape deck. Those blasted C2N (or whatever) Commodore decks kept needing head realignment (well, they did for me anyway).

[Spector]

I'm more interested in what actually happened rather than what happens on paper.

Translation: You are more interested in a comparison which favors your favorite machine.

 

As opposed to you.

[pocketmego]

I'm more interested in what actually happened rather than what happens on paper.

Translation: You are more interested in a comparison which favors your favorite machine.

 

 

I don't know how true that is. This discussion has gotten so technical, i'm not even sure what the debate is about anymore. Not all of us are tech heads you know. I judge my gaming machines by how well they fulfill my needs as a gamer.

 

I'll stand right up and say that I like the Spectrum as a machine. I think it has some unique games that only IT plays to the very best of their design. I've even gone so far to say that if you are not playing Manic miner on a Spectrum, you are not really playing Manic Miner.

 

However, I can't take anything away from what has clearly shown itself to be the winner of this poll. The C-64 is a magnificent gaming machine that stayed strong until the 90's 16 bit console era shut it down. But, even then it was still keeping pace with C64 versions of things like Final Fight, Forgotten Worlds, Stryder, etc.

 

I have no experience with Atari machines , other than playing 5200 games which are supposed to be the same stuff that was on the Atari 400 and 800 machines. Unfortunately I played a 7800 before I ever played a 5200 and I am spoiled. However, i know the 5200 version of Mario Bros. is pretty freakin' wretched when compared to either of the C64 versions of the game by Ataria nd Ocean.

 

Just a few random thoughts.

 

-Ray

[JamesD]

Translation: You are more interested in a comparison which favors your favorite machine.

As opposed to you.

Never even owned a C64 and I don't really care one way or the other.

I just think a lot of the ways the two machines are being compared here is pretty biased. (go figure)

 

Speccy owners focus on a 3D games and discount the sprite & collision detection or character generation capabilities of the C64 but I'll bet every Speccy owner had a bunch of sprite oriented games in their collection as a kid.

Should I go through the top games lists for the Speccy and point out how many were sprite oriented vs 3D?

 

Each machine has it's advantages and saying one or the other is better depends on what you are doing.

Even within a specific area you can't make an assumption that one or the other is better based on a single game because you don't know how the game was implemented and if it's a fair comparison.

[JamesD]

I'll stand right up and say that I like the Spectrum as a machine. I think it has some unique games that only IT plays to the very best of their design. I've even gone so far to say that if you are not playing Manic miner on a Spectrum, you are not really playing Manic Miner.

Well, the same could be said for almost every other computer. There are games specific to each that take advantage of some feature of the original platform that gives it a certain feel that is almost impossible to duplicate unless you do a line by line translation of the code. Many ports were done without a complete line by line translation and they don't "feel" the same.

 

On the other hand if a translation is too close to the original code you end up loosing possible improvements for other platforms. But then some ports end up surpassing the original because the author does take advantage of some feature of a different maching. It almost always comes back to who was writing it..

[Bill Brasky]

OMG, the C64 is winning in a landslide! What a shocker!

[Curt Vendel]

The vote is really unfair, if you are going to include Atari, then it should be the Atari 800 not the 400, unless you want to change the selection of the C64 to a VIC20

 

 

 

 

Curt