[ome-users] czi multi-file images

[Maier, Lorenz]

Hi Melissa,

I've compiled the changes that you proposed for the CZI reader a few days ago from your git repository and replaced the jars/bioformats/formats-*.jar files of Fiji with the new ones.

The changes look quite promising! By now, the setId() of the 300gb multi-file czi dataset I've sent you takes only 2.5h instead of 5h when they are loaded directly from our network folder. 

How much effort would it be to allow loading the individual 1gb files of the dataset?

I assume that you have implemented reading the headers from all of the files because the CZI specs do not specify which data is stored where. The actual implementation in Zen blue, however, only has the option to split the dataset by the series.

So I know that the images I would like to load are stored e.g. in file 123, but Fiji spends 2.5h reading headers from all 300 files before even asking me which series I want to display.


Best regards,
Lorenz