Battery backed up 8K to 16K area

[PokeMon]

77 hex = 119 dec = 01110111 bin

Yes, additionally we have octal system which would be 167 in octal.
Octal is for example used in unix filesystem for rights management to files/directories.

@crankorgan
Yes, $ means hexadezimal, quite old manner of writing I think.
There are several methods for coding hex:

$77 (old way, not often used now but I like it )
0x77 (which is used by microsoft / intel)
77h (some c-compiler or assembler needs a leading zero for maybe ffh => 0ffh)

The ZX81 Original ROM listing was printed this way (coding hex with $)
http://www.wearmouth.demon.co.uk/zx81.htm

[crankorgan]

Pokeman,
I got good with basic. A buddy told me to buy the Quick Basic compiler so I could turn my BASIC files into programs. If you took out the bogus library files you got a really fast lean program. When I worked for TRW the school would teach us site parameter documents. These were areas of the main program you could change. I did mostly banking equipment repair to the chip level. So programming is my weak link! Other than changing the bank name or how to switch the order the data was polled from the machines, I learned nothing! At the hardware level building has been nonstop since 1964.

John

[crankorgan]

Live and learn! I added a second latch off the first. Both have the same clock. It seems the first latch is sending the second latch all zeros. By eye it looks like there is data because the first set of leds are lit. I just love this stuff! Never assume anything when it comes to timing!!!!

[PokeMon]

This was only a hint. It was not my intent to joke about you.
Well timing is interesting stuff.
If you have two latches with the same clock than you have something like a FIFO memory with about one byte capacity.
On the second clock the second latch will get the data of the first latch if it's only clock triggered.
But has also latches which are transparent during clock high or low and save data on the other state.
First effect is caused through delay from input to output which has nearly every IC.
I don't know if your last post was a question about it.

[crankorgan]

Pokemon,
This thread is just a report of what I have done and my experiences. The timing thing is always there! A guy I worked with built a frequency counter circuit where everything happened at once. When it did not work he brought it over to my house. I explained to him the three steps he needed to get his circuit to work. The guy had no real concept of time itself!!! When I build a circuit I can usually figure out where I went wrong. Being 60 years old it takes me longer. To see the signal that inhibits the ROM in the 8K to 16K only happen during a PEEK and POKE blows my mind. I am into logical things. I would expect to see a signal every time the computer's address goes past that spot not just a read or write. My take on the latch is I am catching it while it is loading. My eye says there is data there because of the LEDs. I now know better! My proto board is really worn out! Thirty five years of circuits. Some holes have no grip on the wire!


John

[crankorgan]

I swapped out the 74HC273 with 74HC373 chips. The only difference is pin 1 on the 74HC273 is a reset and pin 1 on the 74HC373 is an output enable. Pin 1 was at +5 and now it is at ground. Guess What? The output of the first chip is feeding the second. Same positive going clock and both chips are working!!!!

[crankorgan]

I slowed the program way down. Both latches are displaying the same content. So feeding one latch into an other does not work either. At least this chip calls it latch enable and not clock. Next Chip!

[crankorgan]

Today I will experiment with a 74HC138 and latches. The other idea had a big flaw. Even if I used 74HC74 flip flops lights would only move in one direction. Using a 74HC138 to create multiple addresses and latches lets me have lights with more patterns.

[crankorgan]

If I add the data lines of more than two 74HC373 together things quit!!! My buffer off the ZX81 is a 74HC245. What happened to the days of fanouts of 10? I tried a 10K pullup!

[crankorgan]

If I don't use the supply on the proto board and I bring over the ZX81 power it all works. Before I was using a common ground between the two units but I let the proto board supply the +5 to the chips on the proto board. I also swapped the 74HC245 with a 74LS245. Nope! Well at least I got three latches working together. Now for the address decoding.