Multi-View Deconvolution always quitting

Dear all,

I am trying to use the Multi-View Reconstruction plugin to register, fuse, and deconvolve a SPIM dataset. I am very new at this.

I manage to go through with registration and fusion. However, whenever I try to perform the deconvolution step I get a message like the following:

Blocksize in dimension 0 (1024) is smaller than the kernel (1029) which results in an negative effective size: -4. Quitting.

I have tried increasing the block size with no success. Every time I increase the blocksize, the kernel size increases to a bigger number and it quits.

Am I missing something obvious?
Can anyone guess if this is a bug or a simple user error?

The full log is below:

Using spimdata version: 0.9-revision
Using spimreconstruction version: reconstruction-0.1.3
angles selected: 169, 349
channels selected: 488
illuminations selected: 0
tiles selected: Tile0, Tile1
Timepoints selected: 0
Wed Mar 21 10:41:30 CET 2018: Computing maximal PSF size ... 
Wed Mar 21 10:41:31 CET 2018: Maximal transformed PSF size = (731, 153, 2181)
Wed Mar 21 10:42:38 CET 2018: Selected (MultiView)Deconvolution Parameters: 
Bounding Box: Bounding Box 'My Bounding Box' (-1923, -7631, 63) >>> (-1011, -6355, 763)
Downsampling: None
Downsampled Bounding Box: Bounding Box 'My Bounding Box' (-1923, -7631, 63) >>> (-1011, -6355, 763)
Input Image Cache Type: Cached
Weight Cache Type: Cached
PSF Type: Efficient Bayesian - Optimization II (very fast, imprecise)
Psi Init: Blurred, fused image (suggested, higher compute effort)
OSEMSpeedup: 1.0
Num Iterations: 10
Debug Mode: false
use Tikhonov: true
Tikhonov Lambda: 0.006
Compute block size: (1024, 1024, 1024)
Test for empty blocks: true
Cache block size: 32
Cache max num blocks: 10000
Deconvolved/Copy block size: 64
Compute on: CPU (Java)
ComputeBlockThread Factory: ComputeBlockThreadCPUFactory: CPU based
Blending range: 12.0
Blending border: -8.0
Additional smooth blending: false
Group tiles: true
Group illums: true
Split by: Each timepoint & channel
Image Export: Display using ImageJ
ImgLoader.isVirtual(): true
ImgLoader.isMultiResolution(): true
(Wed Mar 21 10:42:38 CET 2018): Deconvolving group 1/1 (group=0-0 >-> 0-1)
(Wed Mar 21 10:42:38 CET 2018): This group contains the following 'virtual views':
(Wed Mar 21 10:42:38 CET 2018): 0-0
(Wed Mar 21 10:42:38 CET 2018): 0-1
(Wed Mar 21 10:42:38 CET 2018): Remaining groups: 2
(Wed Mar 21 10:42:38 CET 2018): Fusion of 'virtual views' 
(Wed Mar 21 10:42:38 CET 2018): Transforming group 1 of 2 (group=0-0)
(Wed Mar 21 10:42:38 CET 2018): Transforming group 2 of 2 (group=0-1)
(Wed Mar 21 10:42:38 CET 2018): Normalizing weights ... 
(Wed Mar 21 10:42:38 CET 2018): Caching fused input images ... 
(Wed Mar 21 10:42:38 CET 2018): Caching weight images ... 
(Wed Mar 21 10:42:38 CET 2018): Grouping, and transforming PSF's 
(Wed Mar 21 10:43:29 CET 2018): Setting up blocks for deconvolution and testing for empty ones that can be dropped.
Blocksize in dimension 0 (1024) is smaller than the kernel (1029) which results in an negative effective size: -4. Quitting.

Best,

Rafael