Custom 2600-multicart using EEPROM

[Sdw]

I've been looking into buying a socketed PCB and some EPROMs from the store, and swap the EROMs back and forth for different games/demos.

 

Then I talked to a friend who's rather knowledgeable about electronics, but not familiar with 2600.

 

He suggested that it could be possible to use an EEPROM (note the extra E) giving the ability to erase/reprogram without using an U/V-eraser.

Also since the EEPROM would be larger than 4 or 8kb he suggested hooking up some switches to the extra address pins and thus getting a crude sort of multi-cart, where you by manually setting the address pins access different regions of the EEPROM.

 

Anyone know if this is possible, or is there some reason that only EPROMs work with the 2600?

Edited December 2, 2008 by Sdw

[roland p]

I don't see why this could not work. Ofcourse you have to be sure the eeproms are pin compatible with the atari roms.

 

I found another (old) thread about this subject: http://www.atariage.com/forums/index.php?showtopic=76719

 

Switching games may not be as easy as it seems, because you might have to deal with different bankswitching scheme's for each game. Or you can just buy every kind of PCB (4k, 8k, 16k etc.) and put 4k games on the 4k pcb and 8k games on the 8k pcb etc.?

[Sdw]

Thanks roland! That link seems to verify that at least it is possible!

 

Regarding the 4k/8k thing, I was thinking of going for the 8k PCB, and then do create 8kb files of the 4kb ROMs aswell by doubling them up, as I heard that that might work.

[Albert]

EEPROMs should work fine, I've used them before.

 

..Al

[+batari]

I've been looking into buying a socketed PCB and some EPROMs from the store, and swap the EROMs back and forth for different games/demos.

 

Then I talked to a friend who's rather knowledgeable about electronics, but not familiar with 2600.

 

He suggested that it could be possible to use an EEPROM (note the extra E) giving the ability to erase/reprogram with using an U/V-eraser.

Also since the EEPROM would be larger than 4 or 8kb he suggested hooking up some switches to the extra address pins and thus getting a crude sort of multi-cart, where you by manually setting the address pins access different regions of the EEPROM.

 

Anyone know if this is possible, or is there some reason that only EPROMs work with the 2600?

An EPROM is already UV-erasable. The difference is that an EEPROM is Electrically erasable - i.e. the programming device will erase it instantly instead of having to place it under UV light for ~20 minutes.

 

But yes, EEPROMs work fine and the 28-series EEPROMs may be directly substituted for the 27-series EPROMs.

[Sdw]

An EPROM is already UV-erasable. The difference is that an EEPROM is Electrically erasable - i.e. the programming

 

Ah yes, I see that I wrote that sentence incorrectly, I meant to write 'withOUT using an U/V-eraser' (since I don't have access to one!). I have corrected it now!

Edited December 2, 2008 by Sdw

[roland p]

When I compare the 27C256 Eprom to the 28C256 EEProm, it seems that pin 1 and 27 are exchanged...

 

27C256: http://www.us.st.com/stonline/books/pdf/docs/2384.pdf

28C256: http://www.atmel.com/dyn/resources/prod_do...nts/doc0006.pdf

[+batari]

When I compare the 27C256 Eprom to the 28C256 EEProm, it seems that pin 1 and 27 are exchanged...

 

27C256: http://www.us.st.com/stonline/books/pdf/docs/2384.pdf

28C256: http://www.atmel.com/dyn/resources/prod_do...nts/doc0006.pdf

You are absolutely right. It must have been the 29-series chips I was thinking of.

[Shawn]

When I compare the 27C256 Eprom to the 28C256 EEProm, it seems that pin 1 and 27 are exchanged...

 

27C256: http://www.us.st.com/stonline/books/pdf/docs/2384.pdf

28C256: http://www.atmel.com/dyn/resources/prod_do...nts/doc0006.pdf

You are absolutely right. It must have been the 29-series chips I was thinking of.

 

 

yep, you where thinking of 29 series.

[supercat]

Before designing for any particular EEPROM or flash chip, check to ensure that there is no address setup time requirement for read cycles. On a typical EPROM, if the address changes while the chip is selected for reading, the behavior of the device outputs will be unspecified for next 50-200ns or so (the access time, or TACC) but by the end of that time the device will start outputting the data from the new address. To put it another way, the device doesn't care when the address is or becomes valid, provided the address is valid at least TACC before the data is needed (the device doesn't really care when the data is needed either, but if another device tries to use the data less than TACC after an address change, the data from the memory chip may be incorrect).

 

On some flash and EEPROM devices, however, the address lines are only sampled once at the start of a memory cycle. In a system that strobes the memory once per cycle, this is fine. On the 2600, however, it would cause severe problems in the absence of a controller that could infer the timing of the 2600's clock (the 4A50 cart does that). Some chips are fine, but some aren't--check data sheets to be sure.