Vrscans at glazing angles / blur issues

[MANUEL_MOUSIOL]

At glazing angles, the fabric material I am using is loosing its detail. A different sample method or less blur would probably help, but there is no such option.
Can anything be done about this, please? I need these to be as sharp as possible.
displacement would probably help too, but there is no way to apply the material or any maps (as there arent any) to a displacement mod...

[Ivaylo Katev]

the material should have internal displacement and bump , check in the advanced panel the values that control them

[MANUEL_MOUSIOL]

Right, thanks.
I see no displacement details except for the edges though.
But what about filtering and blur? How are the materials handled?

[Ivaylo Katev]

the image sampler takes care about this, btw is this the same material (ft0020) ?

[MANUEL_MOUSIOL]

mmmh, with the default settings the angled view is a little mushy, dont you think?
it is the one you gave me, yes.

[Ivaylo Katev]

the attached image is too low res to say, but i noticed that by default the material has no bump . you can enable the bump from the user interface

[MANUEL_MOUSIOL]

But even with your resolution you can see the difference between the front and the top. Bump is overin helping but why is the diffuse becoming blurry? I would like to do something about it

[Ivaylo Katev]

i just uploaded a new eexperimental verion of the same matetrial, with the suffix "reduced", check it out, i think it behaves better under low observation angles

[MANUEL_MOUSIOL]

thanks! lookes better.
still not perfect as you can see at the most distant point to the right, but better.
do you think it can still be improved?

[Ivaylo Katev]

"do you think it can still be improved?"
Well, without the sample i really doubt. So far all the improvements i made were result of the bigger experience and the better building tools that we have, but the scanned data are the same..Today we have a better approach to scan fabrics, but it requires different scan type, that is not possible without the physical sample.

[MANUEL_MOUSIOL]

ok.
what I dont get is that the glazing angles loose detail - does that mean that in the scan process these angles are already scanned with less detail? Why is this happening?

[Ivaylo Katev]

No, the problem appears because under shallow view, a small change in the angle of observation leads to mutch bigger change in the appearance than in the front views. This property of the shallow angles makes impossible the smooth transition between the captured images, if you mix them the result is not a plausible intermediate view, because the details do not overlap properly. As i said, we have solution of this issue, and the new materials prepared for upload to the library does not demonstrate it, but this material is very old, and can't be improved further without significant efforts.
I will ask my coworkers if they can find this piece of fabric.

[MANUEL_MOUSIOL]

that would be great. thank you.
so, are you planning to rescan all of them with the new technology?

[Ivaylo Katev]

no, they are too big in count, but if someone have complains, like you, and we still have the sample (some samples are returned to the customers who agreed to use their scans for the library) , we can rescan it

[MANUEL_MOUSIOL]

ok, too bad. I would think that you would strive for the highest quality in all the materials you offer.
Maybe it is a good idea to put a little info box about the scanning technique. If troubles like this occur, it is faster to figure out if something could be potentially be done about it. From my point of view at least. I wouldnt have to go back and forth to ask about it at least.