Parts I've made

x1108.dat Duplo Brick 2 x 6 x 2 Double Concave
Submitted to the Ldraw Parts Tracker 4/4 2004 (originally called x250)
A duplo part I needed to make one of my fathers old dinosaurs. Good and simple for a first project. Nicked a lot from 31110.dat Duplo Brick 2x2x2 by Tony Hafner, 3659.dat Arch 1x4 by Unknown, and after it entered the PT a lot of help from Steffen in making the cylinders and explaining "matrix all zero errors" so I could fix them.

44661.dat Tail 2 x 3 x 2 small finlike
Submitted to the Ldraw Parts Tracker 5/4 2004 (originally called x260)
A very usable little fin, that I didn't really need, but it was an easy project to work with when I got stuck with the x1108 because I lacked the 48/2-4ndish.dat to finish it and hadn't time to backup my computer so I could safely try to install Primitive Generator by Paul N. Easter and make it myself. The part is primarily based on the 3069B.dat Tile 1x2 with groove by James Jessiman.
After Lego Digital Designer came out the part were given it's correct number 44661 26/12 2004.

33303.dat Fence 1 x 4 x 2 Paling
Submitted to the Ldraw Parts Tracker 28/7 2004
Just something I found in a Pick'A'Brick, probably some Fabuland part. It seemed pretty easy, and a good and simple part to learn using subparts with. The base of the part is probably copied from the 3710.dat Plate 1x4 by James Jessiman. On the 20th I finally had backed up my computer so I could use the Prim_Gen to make the 2-4disc's I needed and was therefore able to upload it seven days later.

After it had stayed in the PT for a while, the reviewers agreed that using high resolution primitives for the rounded tops were overkill, so I was forced to downgrade it which happened on the 25/3 2005. It can be hard to kill your darlings sometimes...

4266.dat Wheel Technic 20 x 30
Submitted to the Ldraw Parts Tracker 4/8 2004 (orignally called x363)
I'd needed this part as engine for my model 37 Escort fighter for a very long time: It was actually submitted only one day short of the one year anniversial of the recreation in MLCad! Unlike the previous one I had huge difficulties with converting the pretty odd dimensionsto LDraw (it turns out to be exactly 2x3 cm's). Parallel to the previous one, I was delayed for a long time by a massive need for new primitives.

5/11 I discovered that the three parts I'd given special x-numbers had already been numbered on Peeron.com and therefore had to be moved to the correct numbers. And if that wasn't bad enough, Guy Vivan made me aware that somebody else (Damien Duqunnoy) had already modeled and uploaded the part a half year before me making all my work unnecessary. Bad day.
After a closer inspection, I decided that, although mine had the best use of primitives (and much more smooth), the central part of Damien's version was much better modeled and only used one submodel compared to four in mine. So, remembering Steven Bliss' complaints that the 4 submodels in 41670.dat was too many, I prepared to surrender and ask an admin to delete my child.

However busy with my course, I didn't get around to do it, and 2/12 2004 I suddenly learned that the admins had found my version as most precise, and  because the error in Damien's version would take a lot of work to correct, decided to overwrite it with my version before I had time to prepare it.

41670.dat Technic Bionicle Leg 4 x 4 with ball joints
Submitted to the Ldraw Parts Tracker 5/8 2004
Another terrible part made parallel to the previous two ones due to lack of primitives, and huge difficulties with "stretching" and scaling the curved parts into position, and "painting" edges onto the the whole afterwards. Originally I used the s/32482s01.dat ~Technic Balljoint with Axle made by Robert Sexton for his 32482.dat Technic Liftarm 1 x 7 bent with 2 Ball Joints, but found the file too big and unprecise so instead I made my own s/41670s04.dat ~Technic Balljoint with 21 LDU Axle based on a new primitive p/ballj1.dat 1-4 Technic Balljoint with Axle edges

However the part ended up using 4 submodels like the last one: Balljoints (04), 2 flanges (02&03), and a subpart 01 basically comprising of half of the main part. Far too much according to Steve Bliss so from 20/11 I started working on reducing the number of subparts by cutting the 01 in half and merging it with the flanges, ditching the 04 and using the ballj1 directly in the main file. But as the flanges suddenly should be imbedded in the subfile, there wasn't any excuse that the slanting ones should stick into the interior, so I decided to cut them off.
But I couldn't just do this by hand because the ringsections I'd used for the flanges had to be changed to thinner versions (which took huge amounts of time to figure out how to calculate), and because the top quad immediately caused non-coplanar error, I had to download and use Paul Easter's Ldraw calculator to calculate the correct coordinates, which also took it's time delaying the completion to 4/12.

On November 15th 2004, sbliss suggested that I reduced the number of primitives drastically and used low res for the balljoints. I reduced the number of subparts in the spring of 2005, but couldn't get myself to throw away all the work I'd had making precise edgelines for the high res balls, which got me stuck with the part for a long time until I decided to upload it anyway on August 19th.

And I haven't even had the chance to use the file yet! But if nobody starts making those bionicle parts, they're never going to be used...

6032 Brick 2 x 3 x 1 & 2/3 Octagonal offset
In progress...
After a couple of pretty nasty parts, I decided that the next one should be an easy one: This one had been in the back of my mind since I had searched for it and found it missing in connection with my 50 Hoverbus project. It turned out that it wasn't the best choice for the job and nowhere easy to make: The internal tubing is simply horrible!

But that was partly my own fault because I made it truly octagonal unlike the other ones... Now I have to change them too :-(