Is it true the prototype 2600/vcs used a 6502 and not a 6507

[carmel_andrews]

According to the text in the follwing link, they claim the proto 2600 used a 6502...interesting, so why did they switch over to the 6507 (or were warner's like tramiel, i.e cheap and cheerfull)

 

_________________________________________________

 

 

http://www.cpushack.com/tag/6502/)

 

 

And the text

 

Wired has an interesting article about several prototypes of rather historical devices. Of much interest are the Apple 1, and the Atari 2600 although the doorbell powered Moog is pretty classy as well.

 

Take a look at the Atari 2600 prototype and notice that they used a MOS 6502 in it. The final version used the lower cost (and smaller) MOS (or Synertek) 6507.

 

Atari 2600 Motherboard - 6507 CPU

 

When designing a product, it often is easier to use the standard full featured version of an IC for development work, and then as you refine the design, trim down to the least, and smallest components you can.

We also learn how Foxconn got its idea of low wages. Steve Jobs himself paid his sister a mere $1/board to assemble the Apple 1.

Its interesting to see how prototypes can be so vastly different from the finished product. A fact that design engineers know all to well. “I have to put all of THAT into what?”

 

__________________________________________________________

 

 

 

Were there any advantages of Atari going with the 6507 (apart from cost) then the 6502...Or was it Atari's intention all along to make an updated version of the 2600 using the 6502 (a'la the Atari 8bit)

[Syntaxerror999]

I think cost was the main drive. Bushnels idea was that a home console had to be as cost effective as posible. The TIA itself was costly to design, but once it was completed was extremely cheap to manufacture.

[atari181]

At that time spending the extra $ on the 6502 really wouldn't make sense . Especially as it was only made to play the first few Atari titles that were produced. It wasnt meant to be the juggernaut it became

[kskunk]

The "number of pins" is a substantial part of a chip's cost. With a tiny, high yielding, chip like the 6502, the number of pins can be the most expensive part of making the chip! Every pin you add increases the risk the part will fail, requiring costly rework or more likely just being trashed. This was especially true in 1977, when wire bonding technology was still primitive.

 

The 6502 was famous for costing $25, which is like $90 in modern money. I'm sure Atari got a better deal than that, but let's assume they got a 30% price cut for cutting off 30% of the pins in the 6507.

 

That's no small savings. Over the first couple of years, that kind of savings would pay the salaries of dozens of engineers.

 

- KS

[SoundGammon]

Here's a pic of the 2600 prototype, thanks to Curt's museum!

[Wickeycolumbus]

Were there any advantages of Atari going with the 6507 (apart from cost) then the 6502...Or was it Atari's intention all along to make an updated version of the 2600 using the 6502 (a'la the Atari 8bit)

 

I think it was all in the cost. Having the A8 and 5200 use 6502 variants didn't stop them from sticking a 6502 in the 7800.

 

Here's an older post where Joe Deciur commented on things he wished were done differently on the 2600:

 

http://www.atariage....ost__p__1684074

 

Excerpt:

 

Examples of 'wish we had done it differently':

- use 40 pin 6502 (cost 50 cents more, all in packaging)

- connect IRQ pin to 6532, so that we could run two threads: display, and game processing

- use 30 pin cartridge connector (cost 50 cents) to add top three address lines, R/W, clock, decode back

- hposition comparators (see above) instead of hmove register/counters

- if we could have afforded the chip area, have 40 bits of playfield instead of 20, used in one of three ways:

1) 40 bits across (cover the whole line with 4 clocks/bit)

2) 80 bits across, reflected (40 then 40)

3) 80 bits across, repeated (40 then 40)

[johannesmutlu]

Well the 6507 chip might have be more cheaper then the 6502 chip and it can only access 8KB rather then 64KB of ram/rom,BUT with the invention pf bankswitching,it became possible to overcome that 8KB limit,however that still means that if you want to port a game ment for the 6502 chip wich uses 64KB of data, to a 6507 chip,you have to moddifie that rom (or sections of 64KB rom from an even bigger rom)in such a way that it can be splitted into sections of 8KB,,wich will will be a tedious task because imagine if a level alone is 64KB then you have to split it up into sections of 8KB and load each section one after another,but if the game’s AI is 64KB then you need to split that too and make sure that each ai’s function can still function with 8KB and then you also could and should only load parts of the ai’s function when needed and ignoire the rest,but i can imagine that that will be hard and theres no garantee that every section will still work within 8KB especial if even 1 of the ai’s function does for instance require a minimum amount of 16KB ,so porting a 64KB game to a chip wich could only proceess 8KB of ram or rom and even splitting 64KB data into chunks of 8KB is a hit or miss porting situation,but still it’s incredible what can be pushed out of such infirior chip,take atari 2600 for example,it only has 4KB of rom (thus even for lower then what the 6507 chip inside it could handle) BUT with bankswitching it’s possible to averagely exceed that to 32KB,64KB or even as high you want,without bankswitching most systems from that era among later systems would,ve become still very primitive and useless.