TESPort-Discussion

[KuKulzA]

Originally posted by Mormacil
mutliple personality disorder? In that case Dagoth Ur is alos joining us?

no, cause in our Main quest ideas we have plans for him/her to come back to Vvardenfell because s/he has disturbing dreams of turmoil... but s/he is too late to stop the short bloody war, but in time to stop something else...
afterall, though he go to Akavir, its always possible s/he comes back to Vvardenfell for a bit, I mean thats where s/he is tied to, Akavir is nice, but s/he is the protector and champion of Morrowind (Vvardenfell specifically)

as for Dagoth ur, he's dead... but you might see 'him' or hear 'him' sorta...

[The Grey Wizard]

I'm a little torn, I have consulted with Vality, and he has really got me thing about TES port as a possiblity. What do you think KuKu? Would you ask for permission, or would you rather me make a 6x6 quad Vvardefell from scratch? It's up to you, though doing it by hand, I won't get an exact replica, just damn close

[KuKulzA]

Originally posted by The Grey Wizard
I'm a little torn, I have consulted with Vality, and he has really got me thing about TES port as a possiblity. What do you think KuKu? Would you ask for permission, or would you rather me make a 6x6 quad Vvardefell from scratch? It's up to you, though doing it by hand, I won't get an exact replica, just damn close

I know. It IS very tempting (I must admit to Vality7 )
but, how about this Grey Wizard...
I know you are experienced with heightmaps so i want you to judge...

Vvardenfell, we want a bigger/taller Vvardenfell and a taller Red Mountain, a flatter Grazelands and a West Gash and Molag Amur with more ravines and canyons and plateaus, and a slightly bigger Bitter Coast and Azura's Coast...
otherwise everything is the same...

so do YOU think it would be easier using TESPort and then editing or making the heightmap then thats that?

[Lightwave]

Originally posted by KuKulzA
Vvardenfell, we want a bigger/taller Vvardenfell and a taller Red Mountain, a flatter Grazelands and a West Gash and Molag Amur with more ravines and canyons and plateaus, and a slightly bigger Bitter Coast and Azura's Coast...
otherwise everything is the same... so do YOU think it would be easier using TESPort and then editing or making the heightmap then thats that?

Scratches head. :confused:
TESPort doesn't (currently) directly help with scaling up a landmass, unless of course you wanted to then export it from the TES4 TESCS heightmapper, scale-up the RAW files in an image editor and re-import them, which would probably work. The TESCS heightmap exports seem to lose a lot of detail though (they're only 1024x1024 and 16-bit heights), but it'd be a base.

Alternatively it should not be terribly hard for me to scale up a perfect replica of Vvardenfell by 2x2 I think by just copying each gradient point twice and doubling the height offset for each cell. Each point on the landmass would also be twice as high, but the mountain gradients would the same (they'd just run twice as long), so it'd be correctly scaled in all 3 dimensions. Since it's much much easier for me to do this with the TES3 spec first, it could then be imported straight in to TES4 with TESPort for any other specific modifications.
I just hope a 2x2 conversion wouldn't look too big. It'd certainly be interesting ...

And after all this, well, would you really need permission if it's been this heavily modified from the original?

Lightwave

p.s. It's not impossible that I could just double the X,Y,Z co-ordinates for each NPC too so they'd be in their correct relative places in the town and outback on the scaled up landmass.

[KuKulzA]

so Grey Wizard... what do you think...
i mean what Lightwave says is very interesting...
several options here, and if each one can work out well, then there are 2-3 good choices...

as a good heightmapper...
what do you feel would be best quality?

editing it then export to Oblivion?
or heightmapping it from scratch?
or import it then edit in the changes?

it's your call, whatever YOU feel will get the job done best, cause I know you're a good guy and can do the heightmap anyways very well if you wanted to, so you choose. However, doing all three and comparing, if we have the time and will to do that, isn't out of the question... hrug: only if ya want to.... all I want is to end up with the best Vvardenfell I can get, and I'm sure thats what you want as well

[Lightwave]

Originally posted by KuKulzA
as a good heightmapper...
what do you feel would be best quality?

editing it then export to Oblivion?
or heightmapping it from scratch?
or import it then edit in the changes?

Heightmapping it from scratch and then still comparing it with the other two solutions really wouldn't be a good use of time, it'd hardly be fair on the Grey Wizard.

From my perspective whatever I do won't all be a waste of time. They have generic uses, so tailoring and applying them to Vvardenfell isn't an all-or-nothing exercise.

I feel that the logical thing is to start with a replica of Vvardenfell to the scale you want and then let the Grey Wizard be individually creative afterwards with the changes you'd like. There's nothing to stop landscapers who take certain areas from making their own modifications either (personally I find landscaping is the most fun part of any mod).

One thing I can do that might help the decision in the days to come is to scale up the TES3 Morrowind.esm using TESfaith; I'll modify it to make quadruples of each cell and quad out the gradient data so you'll have a 4x larger Vvardenfell (twice as wide, twice as long). The beauty of it being that the code for moving all the placed objects (and doors/scripts for that matter) is already in TESfaith and you can visit a real large working Vvardenfel before trying the same in TES4. You could try this modified version in TES3, walk around, see if you still feel a 2x2 scale is OK before commiting to it. Obviously buildings will be further placed apart (e.g. Balmora will be 4x as large) so there'll be gaps between platforms, but that's fine for testing purposes since you wouldn't be using these meshes for TES4 anyway.
For a laugh I'll probably produce a second copy too (since it'll be little extra effort), just setting the scale on each placed object to 2. All platforms should still join correctly then, but every object (tree, rock, building, door etc.) will be twice as high and wide - it'll really give the idea of being in a giant's landscape.

I'm also spending my time getting the best fit land textures to work between TES3->TES4 and TES4->TES3 (the latter is for TEStroi and will mainly help Vality7's Cyrodiil project - most the landscape will come ready textured) so am keeping myself plenty busy atm.

Lightwave

[KuKulzA]

I appreciate it Lightwave :goodjob:
I leave the decision up to Grey Wizard, but in any case I hold to the my asking that the quality in the end be just as good, I don't want to have the easy way taken and then end up with bit less than what is asked for, I want efficient and good. And I know Grey Wizard can make a good heightmap fast, but its up to him, his call on this

[The Grey Wizard]

Sorry, I have had some pressing issues lately, and I have been stranded and unable to access a computer, so please excuse my three day absence. Anyway, I think making from scatch would be fun, but it would get tedious, using lightwaves method will be lower quality, but nothing that can't be fixed after it's imported. Honestly, my maps are low quality the first swipe, then they get smothed out and made realistic. I think we should take every aprouch, and just take it slow, and then see what is the best, and use that. That leaves the door open. It will take longer, but if you want the best quality, time is hardly even worth considering. In fact a team member of mine spent a week just making one tent for Hobbition for my mod because he wanted top end qualtiy. Unfortunatly this cost us alot of time, but the fruit of time and dedication is all in balance with how fast you want a production. All in all, I can make a height map fast, but not as fast as lightwaves utility, and in the end I think both will be high quality, if we do both, we'll see which is better.

[KuKulzA]

then let us do both...
is a month's time good enough? say by the end of January?

[Lightwave]

Originally posted by KuKulzA
then let us do both...
is a month's time good enough? say by the end of January?

I just started looking at writing this earlier this afternoon. So far I've got all the cells and land blown up 4 times (2x2), just some join problems to sort out, it's a bit like doing TESPort over again. :eek:

But I have a hitch with the conversion approach I'm taking. Doubling each raw land data point in both X&Y dimensions is fine for the X-plane since the heightmap data in each LAND record is actually the gradient of each X point relative to the previous (western) X-point. But doubling each data-point in the Y-plane data causes a kind of 'ribbed' or 'banded' effect, mainly noticable around any elevation because neither TES3 or TES4 store Y gradient values.

From north or south it looks fine, perfectly doubled in all dimensions by, from east/west a normally steep but smooth slope looks stepped, something like this: http://projectmanager.f2s.com/morrowind/...Effect.jpg
The only solution is to start interpolating the Y gradient data in one direction, which is not fun and the heightmap will cease to be a 'perfect' replica of the original. I'll probably still release a scaled up morrowind.esm with the band effect in a few days anyway just so one can get an idea of size and then I can decide what approach to take to fix the bands afterwards. These are just my interim musings ...

One thing I'm confused about. The Grey Wizard is talking about a 6x6 heightmap. Is this a 6x6 scale of the size of Vvardenfell? (Because I'm only modifying the code for a 2x2 scale atm).

Hope you all have a nice christmas anyway, or Nadolig Llawen as we say over here.

Lightwave