From: Larry Watson Speaking of hex editing, I haven't looked at this for a while, but here's a quick run down on what I (don't) know regarding the Monza trackfile: (offsets decimal) 0 - 4095 Background. Always 4096 long, since it's straight data. 4096 - 4109 Various offsets, with unknowns at 4098, 4100, 4102, and 4106 4110 - 4111 Number of objects 4112 - 4207 Offsets to objects 4208 - 10717 Object data - I have no idea how this is laid out. Hopefully someone with experience with GP2 object editing knows or can work it out... 10718 - 12717 Unknown. The possible offset at 4100 would point to 11298, perhaps this is relevant. This section is exactly 2000 bytes long, but I think it's coincidence - I've checked a couple of the other files, and this section was a different length. Oh, and the apparent offset at 4106 points to 10718, so unfortunately it probably is important data... :) 12718 - 12745 Track header data. 12746 - 14345 Track segment data 14346 - 14661 CC line data 14662 - 14699 CC setup data, with a number of unknowns. Not sure on exact placement of things like air resistance and track grip, but judging by the actual values present I'm pretty confident they are there. The simple stuff like wing settings and gear ratios were easy to find. 14700 - 14955 Pitlane segment data 14956 - 15015 All unknown at this stage, by me anyway. Only exception is 15010, which appears to be the number of laps for a full-length race. Perhaps pit-stop percentages are in this section? 15016 - 15019 Checksum Something I did find that intrigued me, was a variation in the track header section, which is normally 28 bytes long. After writing a small program to step through the track data (as a way of checking command lengths), I found that two tracks actually had a header that was 30 bytes long. It wasn't hard to allow for, since every single track starts with a 0xAC command, but obviously it would be nice to know why they had a different length header, or perhaps some sort of special code at the start of the track segment data. (the values that are at the end of the header, which are most likely the kerb colours [2 confirmed: "Top dark colour" and "Bottom dark colour", with 2 spare values in front of them, probably the 'light' colours, at a guess) are in the normal places, so the header data itself is completely normal) Regarding the CC-line, I have a theory on how the CC-line is dependent on the track layout to some extent, but even if my thinking is right it won't be any good without knowing how to match the CC-line values to the track segment values, and although I've gone close I haven't yet come up with anything that works well. I saw someone say something about this being worked out for gp2 or gp3, hopefully that was correct, because it would help immensely. Larry.