Latest esm and esp updates

[sandor]

Originally posted by Ibsen's Ghost

I liked this one. It's a good composition for the mountains. However, I noticed that the following textures made an appearance whereas they don't appear at all in the preset cells:

TerrainHDRockLichen01
TerrainHDRockLichen02
TerrainHDEvergreenNeedles01

Is this because the region generation process often reaches outside of the preset cells?

No, these are part of the GFSR01 preset, the cells you selected are on the border of two regions (blended) hence they didn't show up or in a lesser percentage than normal.

Originally posted by Ibsen's Ghost
Strangely, whilst testing, I replaced TerrainHDForestGrassDirtRocks01SU (00027363) with TerrainHDForestMossGrassDirt01SUNoGrass (00089005) and the whole cell went black! I'm not sure what happened but it also happened in W1 when I tried to change TerrainHDRocksDirtMoss01SU (0001C7B4) with ManiaSmallRocks01b (000135BF). I'm not sure what's going on there.... :eek:

Using the Gecko?

Originally posted by Ibsen's Ghost
W2

My second favourite. Again, HydrangeaBlueSU made an unexpected appearance.... and quite frequently as well. I'd have to select something to replace it with or just request for it to be deleted altogether if that's possible....?

Deleting stuff is quite easy.
Keep in mind it's the same preset as W3.

Also the size of the generated area determines the final outcome.

Originally posted by Ibsen's Ghost
However, this was where I noticed a big discrepancy with the textures. I'm aware that SACarrow has only just introduced the option to re-texture cells using the Gecko so I wondered if it was Gecko-related.

I only used the CS so it can't be Gecko related.

Originally posted by Ibsen's Ghost
Conclusion

I had no idea things would be so tricky. I think my best option is to go for W1 but I'll need to get back to you on a new set of replacement suggestions for some of the models and textures as, even with W1, the texture composition comes out all wrong. In the meantime, if you can confirm for me that the following cells were indeed created using this preset (I'm sure they are) then please feel free to continue to generate using "W1" (GFSR01 and NBVSR01 blended):

-14,26 and -13,26
-14,25 and -13,25

I can then get back to you with details on which textures and models to replace. Thanks for all your efforts! :bow:

Yes these cells belong to the blended area.
You said you didn't like area W3 but blending in this case affects the areas where you have a transition around a height of 4096.

There are a few large rocks (GFSR01) that will be generated for the lower areas as well, so it will change the dispersion of objects.

I'll generate a new and larger test area. :check:

[sandor]

I attached a new test file.

Area 1, GFSR01 and NBVSR01 blended (using both presets).

coc "W0"
coc "W1"
coc "W2"

For this area the land below the water has been generated as well.


Area 2.

coc "W3"
This is near the same location as before now using both the presets (see above).


What do you think?

TIA.


EDIT
I deleted (April 25th) the esp for reasons of board efficiency.

[Ibsen's Ghost]

It's difficult to tell what's missing and what's there in some cases but it certainly appears to severely restrict the flora it puts out to Bergamot01, Monkshood and Lavendar. However, I can add flora as a part of the landscaping, I guess. The problem that remains is that it's still coming across as too much, particularly with the rocks (there are thousands of them!). As a corrective measure, would you be able to answer any of the following as part of the generation procedure:

• Is there any way to reduce the total amount of rocks (as part of the CS generation or the search-and-replace facilities offered by the Gecko)?
  
• Is there any way to get the procedure to produce a slightly wider variety of tree types and flora e.g. so that it's roughly equivalent to the composition of the preset cells? Just a little more variety (not content) would be good.
  

The last one isn't quite so important....I guess I could do it by hand...it'd just be a lot less time-consuming if it was done automatically.

I'm still suffering from the problem of the land textures going black. I wasn't using the Gecko, just the quads in the CS. So I haven't been able to judge very well how it will look but I'm getting impatient with the whole process and I just want to get it done now tbh.

W0

The lake just north of W0 seems to have been textured with BW content and doesn't match any of its surroundings. I think it's probably best just to use normal textures for the watery areas and then I'll replace the riverbeds with ManiaSand02 and ManiaKelpSand which seems to merge okay with the rocks. In fact, I could probably do this by hand as well but, again, it would be easier if it was automatic.

Other than this, everything should be fine and looks ready to go with the object replacement stage. I like the new W3 the most because there's space for me to add stuff without having to switch too much around....but the big problem remains the vast amount of rocks still....

I'll send you an update about the object replacements, taking the new arrivals (including the HydrangeaBlueSU) of textures and stuff into account. Not much has changed but these texture compositions seem to be extremely unpredictable so I'll change those.

Thanks again. It's much appreciated.

[sandor]

Originally posted by Ibsen's Ghost

1. Is there any way to reduce the total amount of rocks (as part of the CS generation or the search-and-replace facilities offered by the Gecko)?
   
   
2. Is there any way to get the procedure to produce a slightly wider variety of tree types and flora e.g. so that it's roughly equivalent to the composition of the preset cells? Just a little more variety (not content) would be good.
   

1. Deleting rocks is quite easy with the Gecko --> All of the rocks with that Form ID will be deleted. It's possible to reduce the amount of rocks within the RE preset but it will affect the outcome of the generation. I need to know which rocks and then I'll generate a new test area.
   
   
2. We have to change the OB preset, adding new trees and change the settings for the other objects/textures. This will take a lot of time which I don't have.

Originally posted by Ibsen's Ghost
I'm still suffering from the problem of the land textures going black. I wasn't using the Gecko, just the quads in the CS. So I haven't been able to judge very well how it will look but I'm getting impatient with the whole process and I just want to get it done now tbh.

Save the esp, and reload it, this often fixes the problem.



I'll have a look a the "below the water" (sea and rivers) preset and I'll try to use ManiaSand02 and ManiaKelpSand.


Let me know how to proceed.

TIA.

[sandor]

I altered the underwater preset (adding ManiaSand02 and ManiaKelpSand).

Please check cell W4.


EDIT
I deleted (April 28th) the esp for reasons of board efficiency.

[Ibsen's Ghost]

Originally posted by sandor

• Deleting rocks is quite easy with the Gecko --> All of the rocks with that Form ID will be deleted. It's possible to reduce the amount of rocks within the RE preset but it will affect the outcome of the generation. I need to know which rocks and then I'll generate a new test area.
  
  
• We have to change the OB preset, adding new trees and change the settings for the other objects/textures. This will take a lot of time which I don't have.

I'll have a look a the "below the water" (sea and rivers) preset and I'll try to use ManiaSand02 and ManiaKelpSand.


Let me know how to proceed.

TIA.

I took your advice about the black textures. That's fine now.

With the rocks, do you mean I would have to select a particular set of size models of the GreatRocks to be deleted so that they would not be generated at all? I'd hoped that it would be possible to reduce the general number of rocks per cell significantly rather than compromise on the variety of the models but I can see where this might prove to be impractical in real terms.

So, using W3 from your previous test esp, if I were choosing set models to delete then I would go for these: RockGreatForest095Moss, RockGreatForest195Moss and RockGreatForest255Moss. Would deleting these models also reduce the number of 045 rocks that tend to surround them? If so then that would be a bonus.

Is deleting the rocks likely to change the general composition of other elements such as trees, flora and textures? Would I then be needing to perhaps create another set of replacement directions?

It's okay about the second point above because, as long as there isn't too much within the cell then I can add stuff as I go so I leave it as it is and just do a landscaping sweep.

As for the underwater alterations then yes, that looks a lot better to me. Thanks :yes:.

[sandor]

Originally posted by Ibsen's Ghost
I'd hoped that it would be possible to reduce the general number of rocks per cell significantly rather than compromise on the variety of the models but I can see where this might prove to be impractical in real terms.

This has to be fixed manually.

Originally posted by Ibsen's Ghost
Would deleting these models also reduce the number of 045 rocks that tend to surround them?

Yes.

Originally posted by Ibsen's Ghost
Is deleting the rocks likely to change the general composition of other elements such as trees, flora and textures?

Yes.

Originally posted by Ibsen's Ghost
As for the underwater alterations then yes, that looks a lot better to me. Thanks :yes:.

Noted.

[Ibsen's Ghost]

Thanks for that, sandor. It might be best to do one more jungletest esp then and I'll just amend the replacement list where I see fit. :yes:

[sandor]

I reduced the RockGreatForest095Moss, RockGreatForest195Moss, RockGreatForest255Mos and the type 045 rocks.

Please check cell W3 (using all of the 3 presets GFSR01, NBVSR01 and the altered water preset--> Blended).

Cell W4 (GFSR01, NBVSR01, no water preset --> Blended).


EDIT
I deleted (May 11th) the esp for reasons of board efficiency.

[Ibsen's Ghost]

I'm just altering these item ID codes and I noticed that the numbers in the Cell View window are different from those of the Object Window even if they are the exact same model....? :eek: No wonder the original preset was such a mess!

Do you know which item codes I should go by?

Edit - I've checked these and they seem to be fine now. The textures are driving me absolutely INSANE though! The composition changes with every file.

For any generation process, please use W4 for this as the sand textures were only really intended to make an appearance on the riverbed. If it's possible to get them to do that (appear only on the riverbed) then, by all means, go for it (and the lichen rocks would probably go well with these too)! Otherwise, I'll just add the sand textures during the landscaping process.

So I now have three possible paths in order of preference:

1. If it's possible to continue with generating the cells using the SAME COMPOSITION as we see in W3 then I can submit my final asset replacement notes now.
   
2. Is it possible to separate the process of generating the region from the process of replacing the items? If so then I think it would be best to just generate the whole region using the final item replacement document attached to this post.
   
3. Otherwise, let me know and I'll quietly have a mental breakdown in the corner.
   [/list=1]
   
   The water textures aren't that much of a problem because I can just do it in a big sweep but if I have to delete tons of content or add loads of small flora that need their angles adjusting then it's going to kill me. Likewise, I'm really not keen on producing yet another set of replacement notes unless I really have to.