At 07:46 PM 6/2/04 -0700, you wrote: It's asking if you want to read the FIRST address in the ROM and if you want to read ALL the addresses. Normally you'd just enter 0 (the first address) and the size of the ROM (512) and read the entire thing. But it gives you the option to read only part of the ROM and you don't have to read starting at address 0. IIRC the M9312 ROM is 512 in length so enter 0 and 512 to read the entire thing. Actually I think you have to enter the address in hex and since 0 counts as an address, the last address would be 511 or 1FF hex. IIRC2 the M9312 ROM is only 4 bits wide and the programmer operates with 8 bit words so you're probably going to have to tell the programmer weather to use the upper or lower nybble of data when you start to burn a new ROM. Again it's giving you the option to load the dats somewhere in RAM besides at the beginning. This is so you can "patch" code together. For your purposes just load it at the beginning (address 0). These are common options and are found in just about any decent programmer. BTW Found my 29B manual if you have specific questions. The means that the amount of data exceeded the last address but I don't know if it means the last address in the ROM or the programmer's RAM. In other words you did something like telling it to read 513 bytes of data That's an error from the plug-in pak so you'll have to look it up in the pak's manual. I haven't been following this thread closely and I still don't know what pak you're using so I can't help here. But I suspect it's address related. Joe

the copy from device address is relative to the device. if you have a 2k device, then your answer would have to be between 0 and 2k-1 (or a copy would make no sense). you have to supply size if you don't supply 0 as an answer (to keep from reading off the end of the device.) the ram address allows you to skew the read to somewhere in ram other than aligned with the device address. again, you need to be aware of the size of ram to keep from writing off the end of the ram. if I recall the unit I had could do 2764's in one pass, so had 8k bytes of ram. if you had to do a 27128, you would do two passes. copy from device 0-8k to ram 0-8k program at 0 to 27128 copy from device 8k-16k to ram 0-8k program at 8k to 27128 a 27256 took 4 passes, and so forth. I had both 29a and 29b's and this sort of thing worked for both, whether with unipak or later version. jim pavl wrote: