Pamnal

[Lionhardt]

Interesting hypothesis. Is that your conclusion from your tests? Might run some tests on that as well.

[Egon]

Interesting hypothesis. Is that your conclusion from your tests? Might run some tests on that as well.

no just a pretty wild guess, i bet someone with knowledge about this with will laugh about it and find it absurd because the AI doesnt use Marker that Way but dunno, might be worth a try anyway. I dont have the time atm to test this further. I need to get the map to Beta 3 tomorrow because of big LAN Week with my Friends, more or less off the Grid for 5 Days.

[Egon]

v0015 Tested in Lan and vs AI
Map should be playable in 8P North vs south and 4x2P Config now
increased ambient light and reduced bloom (forgot that ppl may play without bloom)
tuned ampient color to compensate for low bloom/lowered blue for warmer colortemp
added more decals and tuned LOD Cutoff
prevented players from building naval bases where they shouldnt be able to.
fixing player 4/9 had more buildspace than others
reworked 4/9 Base to makes sense for North vs South / 4x2P Teams (60% done)
moved mass points
many AI marker fixes
most likely last patch for 3v3 and 5v5 and droppping in Vavor of 10P and later 8P/12P(streched) Version.
further work on startbases (50% done)

[Egon]

iam back from vacation and got lots of ideas from my friends. here are the first results.

v0016 Tested in Lan and vs AI
added spawn scipt and dropping 3v3 and 5v5, big thx to speed2 and Apofenas.
Ressources now scale with the amount of Players in the Map.
Basically can be played now from 1vs1 to 5vs5,in 2 or 4 Teams, east vs west or north vs south.
increase plateau/dropzone size for better t3 airdrop.
opened new route for even number player
added new airdrop location.
fixed dead angle midplayer proxybase (wasnt dead enough).
fixed description.

v0017 Beta3 Milestone "SchreberGaertner"
release of 10P Version.
much work on startbases (66% done).
much work on hinterland (50% done).
relocated and added 5 mass thus resolved big balance issues.
generated all stratums according to hight and slope from hightmap, removed everything handrawn.
added new and rearranged all stratums: stone->dirt->grasslight->grassdark->sand->wetsand->snow.
changed lighting: high color lvl caused all sorts of issues with bloom and texture selection aso.
added more decals and removed some Placeholder.
final hightmap symetry fix.
much props balance fixing
consolidated markers
updated Pamnal mini.

Spoiler: show

[Lionhardt]

I'd go with some convexity selectors rather than height and slope selectors only. And if you are using world machine already, why not apply some erosion?

[Egon]

I'd go with some convexity selectors rather than height and slope selectors only. And if you are using world machine already, why not apply some erosion?

hi Lionhardt, ty for exceptional feedback as always. Erosion is big on my list from the beginning, however i face a big problem there. i can use the free version of Worldmachine only to modify 1024x800or sth. which boils down to 513x513 hightmaps of a 10km x 10km Map or use it for Stratum Masks which are half the Hightmap size. I could then strech the hightmap to 20x20 in Photoshop but would lose needed detail. ATM i work the other way, creating the 20x20 as a basis and using minimum brushsize of 8. Downsizing to 10x10 and keep nearly 100% detail. Maybe u know a better Solution?
another Problem i have is the loss of data during import/export or resizing the Hightmap (most likely PS). do u know sth lossless?
lastly, what would u say are the benefits of convexity selectors in detail, how would such a pipe look like and for which stratum do u use it?

[Lionhardt]

Erosion is big on my list from the beginning, however i face a big problem there. i can use the free version of Worldmachine only to modify 1024x800or sth. which boils down to 513x513 hightmaps of a 10km x 10km Map or use it for Stratum Masks which are half the Hightmap size. I could then strech the hightmap to 20x20 in Photoshop but would lose needed detail. ATM i work the other way, creating the 20x20 as a basis and using minimum brushsize of 8. Downsizing to 10x10 and keep nearly 100% detail. Maybe u know a better Solution?

Split the height map into k² parts, where k = l/512 where l is the edge length of your map (for non-square maps it's obviously k1*k2). Give those parts to world machien individually and piece them back together afterwards in PS. It will also require some touching up. I know... it sucks

another Problem i have is the loss of data during import/export or resizing the Hightmap (most likely PS). do u know sth lossless?

Hm, I never noticed data loss on im- and export, but always suspected there might be some. Re-sizing certainly loses data. Don't re-size.

lastly, what would u say are the benefits of convexity selectors in detail, how would such a pipe look like and for which stratum do u use it?

For example: Chose a dark rock stratum for the recesses and crevices of a rocky area. Then put a convexity selector on your map, export the selection mask, re-size and blurr in PS, import as a new rock stratum into editor, that lies on-top of the dark rock stratum.

Result looks like this:

download/file.php?id=7792

Or do the opposite of the whole process with concavity data. Could also be used for snow in recesses, as I have done here:

download/file.php?id=7796

[Egon]

Now that u say it. Ur Worldmachine results look amazing esp. the snow. 1 Split would be enough for a 20x20 map with the free limitations me thinks.
That filter and value stays the same for snow and dirtlook or do u invert something (maybe u could share them)? However i will defo look into it, expect some results with v19

[Lionhardt]

I used the same for both. But I always experiment around until I by accident find some configuration that looks good. It also depends on the stratums you use... just experiment a little.

[Egon]

Beginning to realize that applying erosion will be a pain. Am i correct that u went with classicWM+Power Preset as a Basis and do u apply it on the whole map or only selective with some Layout Generator as Shapemask?
I foresee a near 100% rework but unsure about when to do it, wouldnt it be wise to do it at the very end?