P51 Model Bug

[Ali Fish]

i would like to show a model bug and general inconsistancy of the model materials over the model concerning the rear section mainly. please note for this i use plain textures with no detail. all u see is my custom normal map. and the issue between the front and back parts of the plane. issues with reflection and all light transition over that area around the join between the 2 parts. there are no texturing issues that cause the issue.

 

THE UVW mapping is not perfect and the error is "not too" noticable with textures on. i see it everytime though given what my prpjects are. a mis alignment with the uvw. hope to see this fixed, if i could see the UVw layout i could point out the error. edit, found some more errors now. verticaly down the model.

 

[ATTACH]67620[/ATTACH]

 

more images if required.

 

also the under side of the rudder on the normal map is incorrect. possibly another uvw issue.

Edited June 28, 2012 by Ali Fish

[Ali Fish]

quiet round here these days, they must all be at duxford or something. lucky gits.

 

is it possible anyone else can confirm the issues i see with the models UVW mapping?

[-Rudel-]

The miss align I can concur with, but it's an easy fix on your skin, by just raising or lowering your layers by 1 or 2 pixels.

 

There are stretched areas near the windscreen as well.

 

There's a lot more issues than that Ali. On my venture to create a more accurate template, I find that things are in the wrong places, areas are too short or too long. The rear half is giving me a hell of a time..I have a strong belief the rear end is to long. I'm using blue-prints to check my work, along with real photos.

 

I need to add my other errors, but please see my thread from a long time ago.

http://forums.eagle.ru/showthread.php?t=88287

Edited June 30, 2012 by -Rudel-

[Ali Fish]

The miss align I can concur with, but it's an easy fix on your skin, by just raising or lowering your layers by 1 or 2 pixels.

 

There are stretched areas near the windscreen as well.

 

There's a lot more issues than that Ali. On my venture to create a more accurate template, I find that things are in the wrong places, areas are too short or too long. The rear half is giving me a hell of a time..I have a strong belief the rear end is to long. I'm using blue-prints to check my work, along with real photos.

 

I need to add my other errors, but please see my thread from a long time ago.

http://forums.eagle.ru/showthread.php?t=88287

 

hi rudel thanks for commenting. what you mean by raising the pixels ? brightness ? no matter what i do i cant rid of it, every trick tried. uvw is the problem my modellor person tells me. ive done all the brightening to the front or back to match them but its as i say the light normals are off because the reflection doesnt transition properly over the model. but i know what you mean about geometry seeming off, cant confirm myself but it definetly needs some precise work, the detail on the default textures masks the problem significantly.

 

some nice info on your thread there, missed that thread initially i did. i may have modelled this plane like in those sectional views then joined it all up and then removed unnecessary polys maybe even after laying the uvw to later determine whats off and whats right. too much stretching. and not enough consistant scale applied to the poly layout

 

hopefully not an issue, but the original artists style masks all of the problems, i hope that doesnt deem it with lack of neccesity to fix, the metallics i been doin are way beyond what the original artist intended and even consulting the style used for the a10, whilst relfection is there and can be used dramaticly, its deliberatly not payed any attention and under used in my opinion.

 

it would be great to have the main shaders tuned to accomodate the effect more and it would be great to have an alpha channel used in the shader construct that can inhibit the amount of colour being used via the reflections. with this we could go near a photographic quality staged arena. very important actually. and very possible.

 

hope the artists are reading because their conceptualisation of light regards the construct of the materials could be different which for me could equate to better, more dynamic. but subject to opinion. for my stuff all i did was apply my own conceptualisation beyond what the original textures where doing for the p51. incidentally ive tried applying the same methods to the a-10 to find at the core via the models materials everything is set up completely different in terms of powering the shaders, id like to see some conformity for all the aircraft in the engine having the same opportunities via the texturing for future ed aircraft.

Edited June 30, 2012 by Ali Fish

[-Rudel-]

It's not a huge deal to me, but if DCS wants to maintain accurate models, I feel they should fix what I have pointed out.

 

I just fount a huge dent near the bottom rudder joint.

[Ali Fish]

It's not a huge deal to me, but if DCS wants to maintain accurate models, I feel they should fix what I have pointed out.

 

I just fount a huge dent near the bottom rudder joint.

 

noticed that too. my complaints are a big deal - to me, maybe not to you. thats your perogative. and please dont suggest that they should fix what you suggest whilst in another persons bug reporting post, i did have a question for you but you failed to respnd and stated you dont care much for the issue.

 

i want these metallics to look as good as possible. given that they are possible and well for money im not accepting anything less than approaching perfect. so yes contrary to a beliefthats its not that important, for me it seems to be. please fix the uvw issues.

Edited July 3, 2012 by Ali Fish

[-Rudel-]

We're clearly on the same page here, but because we're not face-to-face, my reply before hand was misunderstood.

 

I'm not here to ruine your thread, I'm sharing my opinion about your topic title "...Model Bug". The more we/others talk about it, hopefully the more it gets attention.

 

I did not intentionally skip your question either, as "I don't care much for the issue". You say there is a mis-align, all you need to do is adjust, or rather, move your template layers to correct an alignment issue....

 

You want your metallics the best and I want the shape the best so my lines and rivets closely mimic the real deal.

 

So, have you messed with the Normal Map as in NMp files? Not the SPEC or DIF. Word on the street is that 3DS Max's Normal Map (NMp) rendering is not really the best to use in games.

 

Perhaps creating an area on the NMp file can rid of your dark corner seams. I'm not saying it will work, but its worth a try.

Edited July 3, 2012 by -Rudel-
PM Received...

[Ali Fish]

We're clearly on the same page here, but because we're not face-to-face, my reply before hand was misunderstood.

 

I'm not here to ruine your thread, I'm sharing my opinion about your topic title "...Model Bug". The more we/others talk about it, hopefully the more it gets attention.

 

I did not intentionally skip your question either, as "I don't care much for the issue". You say there is a mis-align, all you need to do is adjust, or rather, move your template layers to correct an alignment issue....

 

You want your metallics the best and I want the shape the best so my lines and rivets closely mimic the real deal.

 

So, have you messed with the Normal Map as in NMp files? Not the SPEC or DIF. Word on the street is that 3DS Max's Normal Map (NMp) rendering is not really the best to use in games.

 

Perhaps creating an area on the NMp file can rid of your dark corner seams. I'm not saying it will work, but its worth a try.

 

this was the last display of that new nmp file. http://forums.eagle.ru/showpost.php?p=1488790&postcount=5556

 

 

yes thats an idea ive missed, whilst using completely blank textures for testing, i may find the issue via a completely blank nmp file. good thinking man ! sorry bout what i was saying earlier, but yeh were at such precise levels of discussing this bug as they are practically invisible to most. we get our wires crossed a bit, no harm intended.

 

btw have you joined the front template to the back template ? i believe this is how this bird had its uv layout done, and naturally the templace creation. and due to that could be some errors via splicing it into 2 seperate templates !!!

 

edit : problems still there with zero detail textures and nmp files.

Edited July 3, 2012 by Ali Fish

[-Rudel-]

I saw that picture before. It's nearly perfect. Nice job.

 

I joined the front and back to help me ensure my details line up. It's interesting to know the top area behind the canopy, right half fuselage and bottom oil cooler outlet shrouds wireframes do not line up by 1 or 2 pixels either vertically or horizontally.

 

You'll need detail on the dark areas in the NMp. I'd imagine a very light shade of pink.

[Ali Fish]

 

You'll need detail on the dark areas in the NMp. I'd imagine a very light shade of pink.

 

its just a grey scale heightmap passed through the dds normal map filter and converted to bmp. it needs bmp as dds isnt to sharp at these levels and artifacts etc. but i wont do that to fudge a fix.

Edited July 3, 2012 by Ali Fish

[Ali Fish]

just had another thought about these issues. the bort serial number layering system could be affecting.

[-Rudel-]

I doubt it, have a look on the belly near the tail section.

[Ali Fish]

I doubt it, have a look on the belly near the tail section.

 

dodgy polys indeed. ive been working on a new panels system graphically one that can aged & weathered dynamicly at various stages, sadly ive found even more little issues. doh! what i must say is im looking for better consistancy over the 4 templates that conclude the final effect where fine details are concerned. pixel perfect transitions please, more accuracy when splitting the initial uvw to the 4 texture plates.

[-Rudel-]

That wouuld be nice. Hoping for a fix too :)

[Ali Fish]

we can remove this thread now i believe.