The company I currently work for uses many surface mount chips in their products. We also make extensive use of FPGA's. Since we use this stuff so much, it seems like the "obvious solution" for getting the LCD's running.

I admit that my current work may bias my opinion.

However, I will throw out some ideas on how I'd approach the problem from my experience, using the tools I'm used to.

I've already talked about "get something running by hook or by crook" just to have a sample circuit that can be used as a base. From this example and a logic analyzer, the working circuit can be throughly documented. From the specs I've seen so far, I'm still am bit confused, and I don't really trust 'em.

A working circuit can also be used to see if the LCD's are decent quality, etc. (I had this horrible thought that none of the LCD's even worked... they're rejects for a reason!).

Once we know what to do for sure, using a Altera FPGA create a circuit that is something like this:

What this shows is the Altera part in the middle, with all the pieces glued around it.

I choose the Altera part because it's a SRAM based part. Using a two wire interface, you can program the part by wiggling a couple pins.

The draw back is that you have to use the Altera development kit, that costs money. I have access to that kit here, that's not a problem for me, but would be for other people.

The Altera parts are < $80, and you can just solder them onto the board. In fact, you'd just solder the entire thing together and not worry about the FPGA code because you in-circuit program it.

From my (maybe limited) world view, I'd go for a PCB first time out. Since everything is tied to the FPGA, and all the pins on the FPGA are completely programmable, the FPGA becomes a "virtual PCB".

The "Interface of some sort" could be anything, I even had thoughts of making it a USB serial interface. I'm more interested in having a pure digital interface than VGA inputs... thus the circuit proposed. This interface would at minimum be a second port access to the display ram.

The display ram would be ordinary cheap SRAM, multiplexed inside the FPGA to give it the look and feel of two port ram. I'd be easy to time multiplex the external interface and the LCD requests to the SRAM.

I was thinking of driving the LCD with high voltage open collector drivers such as the 7406, or 7407 (I can't remember which one I choose right now). A simple resistor array can be used to build a D to A converter.

The problem with providing the "DC balanced signal" that is done with the driver chips we've discussed could be accomplished by inverting the signal every other line. That'd be simple to do with the FPGA.

This summarizes my thoughts on the project. Unfortunately, I work a full day, and am starting my own business. I really wish I had time to help out. If somebody is really interested in laying a board out and it's close to what I'm thinking about, I may help out by putting some $$ where my mouth is and help finance a PCB, FPGA, etc. I may have even more incentive if there were people who would buy some of the boards (usually you get more than one made even when doing proto's).

Any PCB would have to be compatible with the layout tools I have here also.

Thanx, Tom C.