Mike,

The setup that you described was the following:

RAM connected to Port 0, LEDs connected to Port 1.
You are writing to the RAM on Port 0 and then to the LEDs on Port 1. That is what I deduced from your software snippet. So the LEDs are unaffected by what is going on with the RAM. The RAM could be removed and based on the software that you have written, the LEDs would still continue to operate.

Erik is correct the 50 mSec delay is for the LEDs, not the RAM. If the Chip Select, ALE, RD, WR are connected properly and the access time is 200 nSec, you should be able to write and then read on the next uProcessor cycle.

I like the serial port method of debugging because:
1. I don't need special hardware.
2. The software runs at real-time speed.

Keep plugging at this problem, consider the advice given on this forum and eventually, you will solve your problem and learn some new insights. I know that I have learned some new tricks from this forum, even though I have been working with uProcessors for some 20 years.

Regards,

Jacob