News for March 2005

My to-do list for release 2.3 is now looking pretty short, so I think we might see the full release in April sometime.

The automatic design-map generation algorithm is being overhauled to improve the terrain realism and to expand the range of overlay effects. In particular, the design map will itself be eroded and shaped to provide more realistic mountains and plains:

Preliminary results, rendered in VTP Enviro.

Also, the website now includes a gallery page for 3rd party renderers (VTP Enviro, Terragen).

Hurrah! The last of the serious mosaic errors have been fixed. To celebrate, I've uploaded a beta-version of L3DT 2.3, the full version of which will follow in a week or two.

The main change from 2.2 to 2.3 is the re-implementation of heightfield mosaic maps, which were last seen in L3DT 2.1 (Nov 2003). As a result of this, you can now generate heightfields up to 32k × 32k pixels in size. This version also includes various bug-fixes, optimisations, and user-interface goodies. Also, users of VTP Enviro may be interested in the new 'Export to Enviro' option.

L3DTVi2 is incompatible with this beta-version. This is because I've slightly changed the mosaic file structure. An updated L3DTVi2 will arrive with the full release of L3DT 2.3. Aside from fixing L3DTVi2, the to-do list includes a few minor jobs plus some rigorous testing.

Feedback is most welcome at this stage.

Well, here we are in the middle of March and I still haven't found the water-flooding bug. This is getting a tad annoying.

Also, I've revised the file formats page to make it a bit easier to read.

I had a bug in the save function for the Virtual Terrain Project binary terrain format (.bt) that mucked up the horizontal coordinates. This has been fixed for the next release.

L3DT map in VTP Enviro

…it's not finished. The water table modelling is mosaic-friendly now, but the sea and lake flooding routines still seem a bit bung. Hrumpfh! Anyway, I have absolutely no idea how much longer it will take to finish L3DT 2.3 - it all depends on when I find the bug.

On a completely unrelated topic, does the old maxim of blue and green should never be seen also hold true for orange and purple¹⁾? As a result of that colour scheme I very nearly had an unpleasant Porygon episode, so I may have to tone-down the disco-paced flashing colours somewhat.

Let me assure you that this eye-candy is extremely useful for optimisation and debugging. Sure it is…