S1tbx vers 3.0 Phase unwrapping works differently?

Hi, I recently updated the s1tbx to vers 3.0, using my previous interferogram, I export it to snaphu format, with the option:
TOPO
MCF
tiles Row: 10 Column: 10
No. Processor 4
Row overlap 0
column overlap 0
tile cost threshold 500

When I tried to unwrap it in snaphu line command, the operation works differently. And the process lasted only a few minutes and completed which is different from previous version where it lasts for days.
And the output files:

coh_IW2_VV_18Sep2015_06Sep2015.snaphu.hdr
coh_IW2_VV_18Sep2015_06Sep2015.snaphu.img
?M}???
Phase_ifg_srp_IW2_VV_18Sep2015_06Sep2015.snaphu.hdr
Phase_ifg_srp_IW2_VV_18Sep2015_06Sep2015.snaphu.img
snaphu.conf
snaphu.log
snaphu_tiles_4328
UnwPhase_ifg_srp_IW2_VV_18Sep2015_06Sep2015.snaphu.hdr
UnwPhase_ifg_srp_IW2_VV_18Sep2015_06Sep2015.snaphu.img

Did I make any mistake in my selection? I knew that in this recent version there’s an added function on tiles selection, but I don’t know what to choose. My scene is subsetted to around 20km^2.
Please help! much appreciated

The tiling should help for people who have been running out of memory and it’s also faster. Sometimes there can be jumps between the tiles. If this happens you can set it to one row by one column to disable the tiling.

1 Like

Thanks Iveci it works fine now :slight_smile:

I’ve been having problems with tiling in snaphu phase unwrapping. Following the steps TOPSAR-coregistration->Interferogram and coherence formation->Goldstein Phase Filtering->Snaphu for two SLC files (Interferometric mode), when I try to make the unwrapping using Snaphu algorithm in Linux, an error message appears and says: “numbers of tile rows and columns must be positive”, when I checked the snaphu.conf file at Linux, I saw that the number of tile rows and columns is zero.
This didn’t happen with the previous version of SNAP (2.0), what I’m doing wrong?

Well, I think I located my problem. I made the coregistration, and in this step I select 3 bursts, the same bursts at the two images, the “problem” is that as the two images are not exactly the same, when the software made the coregistration, there are parts that are in one of the images and there are not in the other image, and viceversa, so after the proccess there are regions with no data, black regions. I didn´t make a deburst in my previous try, so maybe that black regions caused any type of problem in the snaphu export proccess.
Now I am doing the phase unwrapping in Linux, now I have a snaphu.conf file with 10 tile rows and columns. I hope all works good.