4416 DRAM pinout

[Tim Shoppa]

I'm trying to determine if an INMOS IMS2620P-10 is pin compatible
with the good 'ol 4416 DRAM. I know (thanks to the GIICM) that
the 4464 puts CAS on pin 18 and A7 on pin 10. What are these same
pins used for on a 4416?

Where can I find a 100 ns version of the 4416? (Other than the
no-longer-available 2620P-10 from INMOS, that is!)

Tim. (sho...@triumf.ca)

[Tim Shoppa]

I'm trying to determine if an INMOS IMS2620P-10 is pin compatible
with the good 'ol 4416 DRAM. I know (thanks to the GIICM) that

the 4464 puts CAS on pin 16 and A7 on pin 10. What are these same

[Kevin Horton]

In article <5alpfh$i6$1...@nntp.ucs.ubc.ca>,

sho...@alph02.triumf.ca (Tim Shoppa) wrote:
>I'm trying to determine if an INMOS IMS2620P-10 is pin compatible
>with the good 'ol 4416 DRAM. I know (thanks to the GIICM) that

>the 4464 puts CAS on pin 18 and A7 on pin 10. What are these same

>pins used for on a 4416?

OK, I just had to play around with replacements for the 4416. I couldn't find
any in my arrary of chips, and I really wanted to get this thing fixed (Atari
5200 videogame system), so I improvised. Since I strip tons of old computer
boards, I had about 300 41256 256*1 DRAM chips. I checked, and these are
pin-for-pin compatible with the 4416 and 4464. The only diff. between the
'256 and the '16 is the addition of two extra address lines. I tied these to
a logic level (I think I tied A8 hi and A7 low, since I could take one piece
of wire and tie all the pins together, since the chips were in a row; tho
just so long as you keep them at the same logic level, you'll be fine.).
I turned it on (after replacing a few more bad things) and it worked!

So, to replace a '16 with a '256: Tie pin 1 to a logic level; e.g. high,
and pin 9 also to a logice level; e.g. low. That's it. You can still use
'16s, since pin 1 and pin 9 are NC's on the '16. '64s will also work, since
pin 1 is NC and pin 9 is A7 (which has been disabled)

Of course, if you use a '256 in place of a '16, you'll only use 1/16 of the
chip, but it'll still work fine. :-)

Note that since you're only refreshing one part of the memory, the others will
not get refreshed. This is not a problem, since you'll only be using the
part that's getting refreshed. It'd be a problem, however, if you wanted to
bank the memory into 16 16K banks; you cannot do this without some extra
refresh circuitry.

[Tim Shoppa]

In article <5ame6c$9e4...@ind-0008-23.iquest.net>,

Kevin Horton <kho...@iquest.net> wrote:
>In article <5alpfh$i6$1...@nntp.ucs.ubc.ca>,
> sho...@alph02.triumf.ca (Tim Shoppa) wrote:
>>I'm trying to determine if an INMOS IMS2620P-10 is pin compatible
>>with the good 'ol 4416 DRAM. I know (thanks to the GIICM) that
>>the 4464 puts CAS on pin 18 and A7 on pin 10. What are these same
>>pins used for on a 4416?
>
>OK, I just had to play around with replacements for the 4416. I couldn't find
>any in my arrary of chips, and I really wanted to get this thing fixed (Atari
>5200 videogame system), so I improvised. Since I strip tons of old computer
>boards, I had about 300 41256 256*1 DRAM chips. I checked, and these are
>pin-for-pin compatible with the 4416 and 4464. The only diff. between the
>'256 and the '16 is the addition of two extra address lines.

You lost me there, Kevin. The 4416 is 16k*4. How do I replace it with
a 256k*1 part? There's more to it than just two extra address lines!

> I tied these to
>a logic level (I think I tied A8 hi and A7 low, since I could take one piece
>of wire and tie all the pins together, since the chips were in a row; tho
>just so long as you keep them at the same logic level, you'll be fine.).
>I turned it on (after replacing a few more bad things) and it worked!

But where did you derive the three extra data lines from?

Tim. (sho...@triumf.ca)