The EEPROM has arrived and been trying to burn the software on it, with not much success. I’ve sort of gave it up and asked a friend to do the EEPROM burning for me. I did a test to try the machine anyway. The ‘NOP test’ consists of hooking up a few wires and resistors to emulate a ROM with the NOP instruction. This particular instruction means ‘do nothing’, move to the next address. Since the processor reads NOP from the NOP generator circuit (as if it was a ROM full of this instruction), it jumps to the next instruction found in the next address. Which is a NOP, therefor the processor reads the next byte at the next address… and so on. I checked the address lines during the test with an oscilloscope and indeed, the processor was furiously sweeping through the entire memory (well, the NOP generator). As I moved the probe from address line 0 to 14, the frequency decreased. Why? The ROM, just like the RAM stores bytes which all have an address. To access a particular byte, we need to set the state of address lines individually (0/1 or low/high). This ROM can store 32Kbyte data, in other words 32768 bytes. To be able to access all of them, we need a 15 digit address. For instance, the first address will be ‘000000000000000’, the second one is ‘000000000000001’ while the last one is ‘111111111111111’ . The last bit of the address is called the ‘least significant bit’ or LSB since changing it will only alter the binary number with 1. If we count from zero to 32768, the least significant bit will flip every time the a number changes from even to odd (or vice versa). The first bit is called (drum roll!!!) the ‘most significant bit’ or MSB since changing it from zero to 1 will add 16385 to the number. In other words, when we count up the LSB will flip 16385 times, the MSB only once. The test (sort of) verfied it which means the computer circuit works, I just need to get the bloody software on the ROM. Maybe in a few weeks.

Advertisements