What’s the difference between high quality model vs. normal quality model with downsizing disabled?
Hello!
Mostly in internally used settings that can slightly affect on precision/details, but mostly on speed. High required more memory optimisatons than lower.
Hi Eric Gingras
What’s the difference between high quality model vs. normal quality model with downsizing disabled?
If you set Normal and downsample of 1x ( from default 2 ) then the results are absolutely the same, the difference is only the image downsample factor, nothing else.
REALLY??? :shock:
That is a new!
When I compare the times from normal-1 to high, I come to a different conclusion…
I thought there are also different optimization algorythms?
Different optimisation. Cluster sizes, etc.
Hi Götz Echtenacher
Some examples ?
Only in my head! 
I’m not saying I’m right!
But it is to be read in plenty of posts - also not saying those are right (links only on explicit request 
)
But what is the purpose of being able to change image downscale on normal then?
For Preview and High it isn’t possible, I think…
Hi Götz Echtenacher
Preview -> only detected and used features are used for creating mesh
Normal -> Image downscale 2 is default, in many cases 3 or 4 are used for various reasons ( 50-250Mpix images ) so you can play a bit here.
High -> Image downscale 1x ( so NO downscale at all is used ) so if want use 2x 3x 4x then the NORMAL mode is here for changing settings…
Hi Wishgranter,
ok, that works for me!
I’ve been using normal-1 because I was afraid of calculation times in High… :lol:
So what is used if in preview the setting Use Sparse Cloud is set to False?
More features?
Or depth maps after all?