RC needs auto 100% coverage of texture space in UV Unwarp, no metter of texture resolution (> 2048).
You recommend to use 2048x2048 in Unwarp settings, but it is actually resolution of texture that will be on output.
So you recommend to use low resolution…
You know, 2048x2048 It is very low for 3d printing.
We loose quality of texture because of it, not all of us use mobile phones to take photos for photogrammetry.
So what is your request? 
wrote:
So what is your request?
My request if it is no obviously in first sentence.
“RC needs auto 100% coverage of texture space in UV Unwarp, no metter of texture resolution.”
Ah, ok, I guess I missed that. Sorry.
You want a button for that?
Or just that RC does it properly right away?
The thing is that I guess if it was easily done, it would already exist.
I can only tell you from my experience, that I usually get 100% without doing anything special.
Nevermind that I agree with you (like in the other thread) that the settings and mechanism could be better explained…
I do remember issues from before I knew how to handle alignment in RC properly though.
So I think that there is potential for improvement there - I gather that you more or less just started with it?
wrote:
Ah, ok, I guess I missed that. Sorry.
You want a button for that?
Or just that RC does it properly right away?The thing is that I guess if it was easily done, it would already exist.
I can only tell you from my experience, that I usually get 100% without doing anything special.
Nevermind that I agree with you (like in the other thread) that the settings and mechanism could be better explained…
I do remember issues from before I knew how to handle alignment in RC properly though.
So I think that there is potential for improvement there - I gather that you more or less just started with it?
Just about a week, but previously I was using Agiosoft PhotoScan.
I think texture size + unwrap improvement should be listed as high priority feature.
With good unwrap you can easily use and modify the model in a VFX pipeline. Just look how good Unfold3D (unwrap software) works, it would be awesome to have that kind of nice unwrap in RC!
Can you try to set the “RECONSTRUCTION \ Settings \ Advanced \ Use legacy unwrap algorithms” to false and try to unwrap again?
There is a new algorithm that should work better…
Hi Michal!
If it works better why isn’t it set to “false” as a default value?
Can you explain the difference between the 2? Thanks 
You can consider it as experimental. We implemented it but had to swith to another tasks so it is not yet approved. It is needed to be tested properly, the tool needs to pass internal reviews etc. it is a process. But we released it in this form so that we can get a feedback meanwhile. So any comments on the new unwrap are welcome.
The difference is that it should produce better utilization.
Have you tried it? Do you like the results?
I’m going to try it out tomorrow, I’ll give you my thought on it
Thinking of it… there is a feature that would be really cool and less time consuming: the possibility to see the unwrap without exporting the texture. Everytime I had to bake the texture and export it to see how it looks. There is no way to just see the unwrap in RC, right? Like in a “2D” view or something? If not I think I’m going to create a thread about it, because it is really needed IMO.
Hi Julian,
if you abort texturing after the unwrap, you will see a checkered pattern in black an white.
At least that happened to me once by accident.
An orthophoto I created from that also showed the pattern, so I assume it’s a temporary texture file and I guess it could be exported.
Is that what you need?
Hey there, here’s my results!
It has been tested on a simplified mesh (16M (~20parts) -> 5M (1part)). I don’t think this changes something (?), just giving the more info that I can.
i** n blue: parameters that I changed**
in green: best results (between the 2)
in red: worst results (between the 2)
best utilization: 71%
So far the new unwrap seems to work better in “maximal count texture” method and with the “adaptive Texel size” set as default.
But if we try to get more details by decreasing “the minimal Texel size”, the old unwrap method gets better results but not as good as what the new unwrap can come up with. Decreasing “the maximal Texel size” doesn’t get big changes.
Maybe I should have tried with the maximal Texel even more lower to see if legacy unwrap could compete with the new one? I may try it when I get some spare time.
Götz: I just need a UV viewer, that’s all
by the way the resolution 8688*5792px is per photo ofc.
Hey Julian,
I thought you wanted to see a neutral unwrap. How would you be able to use a UV viewer if you don’t export anything?
Nice that you uploaded all the screenshot. I’m sure there will be many people who appreciate that and can learn from it (including me of course).
What do you mean when you say that one method is better than another. Do you just mean texture utilization or also texture quality or max resolution or… ?
Why would you have to export to see an unwrap? lol. If it is made in RC, it can be easily shown directly in RC. It’s a waste of time to export everytime.
(btw your idea about having a simple checker texture as a display possibility seems to be a good idea as well. +1).
For the comparison, you can see in green and red the parameters I use to determine what is best. Is there any parameters else to check out texture quality that I’m not aware of?
Ah, you still meant you need a viewer INSIDE RC? Thats different. I thought you wanted to use something external. Well, if there is no such viewer in RC (which I doubt), then there is no way around an export, right?
Just out of curiosity, what’s so difficult about exporting it? Is it just this one step more that we all wish we would’nt have to make ourselves? 
Been there many times… lol
About “better”: its all very subjective. What is better for you might be worse for somebody else. Hence my question. So you mean by better a higher texture utilization? I think it means how much of the file is used by the unwrap, right? In my case, that couldn’t matter less, which is why I never change much there. But I understand that for other purposes, like loading it into a game engine, it can be quite important.
How much the texture utilization changes might also be highly dependent on the object and how the images were taken, I imagine. Your example is very straightforward, basically aerial photography with all cameras at almost the same distance to the object. But if the object is more complex and the surfaces are at very different depths in one image, the results might vary much more since there will be areas with much higher possible resolution than others. And if you then use adaptive texel size I recon that the texture utilization will be higher since some areas will have to be bigger due to a higher resolution.
Texture quality is a whole different story. I never quite understood the principles behind that. I think it puts the resolution of the image in relation to the achieved texture resolution. But I suspect that it is somewhat arbitrary for the same reasons as I mentioned in the last paragraph. I had a very simple object similar to yours (a wall) where I got 100% texture quality without doing anything special. Then I had a very complex roof space where I paid much more attention to alignment quality and even used RAWs instead of in camera JPGs (in the former project) but only got an abysmally bad texture quality of 35%! The resulting texture however is as good or even “better” than in the project with 100% texture quality. So I guess that the value can only be a guideline.
By the way, how is your alignment report? Because that can also influence the texture.