Monthly Archives: April 2006

NES tomfoolery


I have started playing with NES again and have thought about making a re-programmable cartridge since it could be possible to put in EEPROMs instead of the PRG and CHR chips. I will have to look into the ROM file mapping to the chips and modify my parallel EEPROM programmer program accordingly. Below are a pair of files which outline the chip pin-outs and the mapper which are originally from by way of More on this later when I get the circuit together.

rom.txt NES_ROM_Pinouts.txt

How to make your Firefox and Thunderbird experience better

Every once in awhile, internist I’ll say to myself that I should update or something. Yeah, pilule I forgot about this thing. Life happened.

Anyway, this t-shirt (hat-tip to Hide Your Arms):


Two helpful extensions that I have found lately are Adblock plus and domain mismatch remember. The adblock is self-explanatory and allows you to selectively block images as well as flash adverts. As for the thunderbird extension, pharmacy
it makes thunderbird remember that you don’t care about the domain and ssl cert mismatch on your server.

How to abuse JTAG specs

My roommate, surgeon
Russell, was having some difficulty programming a JTAG board. The board was designed some time ago when the JTAG specifications were gratuitous and claimed that an infinite number of devices could be chained together and programmed in serial. Over the years the JTAG spec has shrank from 20 to 10 to 5-6 devices on a single chain, so most of JTAG programmers probably only allow enough fan-out to support the spec, save a few pennies, right? As a consequence, these large boards with 15+ devices sometimes could not be programmed properly with modern programmers which sometimes resulted in setting the on-board chips in such a condition that cause permanent damage and required high pin-count device replacement. We discussed the problem at length over a few drinks and concluded that it was indeed a fan-out problem which could be solved by placing a pull-up resistor on the output of the JTAG programmer and then a voltage-follower that can supply enough output current to drive the chips. After building a prototype, it seems that the solution worked. Not a bad way to spend a few dollars to save a few hours of pin-lifting and chip replacement.