From: Gideon Zweijtzer (gideonz_at_dds.nl)
Date: 2003-06-13 22:07:41
>
> > RAM is build from Flip-Flops (FF), which have two states: 1
> and 0 (or:
> > on and off). Well, this is not really true with the C64's
> RAM, because
> > it is DRAM (and not SRAM), but the effect is the same.
>
> DRAM is made of capacitors, and SRAM from flip-flops.
No. SRAM is not made of flip-flops. This would take up a huge amount of
silicon area, because a flip-flop is made of two latches, which are each
made up out of two gates (and a 2-input CMOS gate is at least 4
transistors). Traditional SRAM cells are made up of only 6 transistors,
but I have heard that newer SRAM cells are made out of 2 or 4
transistors only. Well, still a lot more than DRAM, that is why SRAM is
always smaller in memory-size than DRAM in the same die-size. (You can
buy 32 Mbit SRAMs now I think... but the biggest DRAM I heard of is 1
Gbit in one chip!)
> SRAM is
> more stable even when you cut the power (unless you are in
> .18-micron with high transistor leakage currents).
This is nonsense. The contents of true SRAM is lost as soon as the power
switches off. The contents of SRAM is stored in an active transistor
loop. If the power of the loop is turned off, the information is truly
lost. [However, SRAMs are capable of holding its contents at low
voltages, such as 1.0-1.5V (depending on the fet's threshold voltage).
When the power does not drop under this level, the contents can be read
back after raising the power back to its nominal value.] SRAM is very
low-power though, because the active loop is static and does not consume
power, other than some leakage. Leakage is very small, because when the
level of the SRAM cell does not change, there is just some charge kept
on the insulated gate of the transistors.
So DRAMs are more stable. The cells are specified to hold the
information at least 64 ms (standard refresh time). This is a guaranteed
time. But in reality, the cells are a lot better than that, and can hold
the information for some seconds even (especially at lower
temperatures). That is why you can find your program still in memory
after quickly switching your Commodore off and on.
> This is because the neighboring cells usually have different
> polarities. The cell itself 'resets' to the same state, but
> the column-read logic inverts every other read line.
I would love to see some proof of this :)
Gideon
Message was sent through the cbm-hackers mailing list
Archive generated by hypermail pre-2.1.8.