The computer specs you need is highly dependent upon the type of data being processed (scale), speed at which you need the processing done (throughput) and the type of processing being done.
For example, if I am processing time-lapse stem cell images where each timepoint is 20k x 20k pixels and the entire movie is 500GB, that is a different problem than single FOV time-lapse studies.
Factors to consider:
- size of the input data: how much data do you need to hold in memory at any given time to perform the required processing?
- runtime requirements: Fiji has some plugins which utilize multithreaded implementations, but not all do, so this requirement is harder to specify. Check with the individual plugins/operations you need to run.
- Disk access speed: this one is often overlooked. If you are reading and writing to disk a lot, a m.2 ssd can be worth its price tag. If you have larger datasets (larger than can fit on the ssd) it might be useful to setup a m.2 ssd for the boot drive and processing swap space, but store the full datasets on a 3.5" hard drive. This can give you the benefit of faster access time, while preserving the raw capacity a HD can provide.
Overall, the most important factor is the amount of data that you need to hold in memory concurrently. If you have too little memory you cannot process the data at all. However, if you have an under-powered CPU the processing just takes longer.