Yesterday I received my mini M2 Pro with 32 GB of RAM and 2 TB of disk. Of course, first thing I did after it finished to do their things was install last DT version. And… well, this thing fly.
I’ve done some fast dirty tests between this mini and my MacBook Pro 16 M1 Pro, 16 GB RAM, 1 TB disk and:
Cold start time after reboot from completely closed are similar
Hot start time from completely closed are about 1 second faster
UI performance is the the same
First search on first open takes more or less the same time
However:
Live search even with partial words flies
Next searches are almost immediate
(In MacBook Pro they take some seconds each).
DT contains 7 databases with 17 626 814 unique words, 1 421 681 284 total words, 42 830 items, and about 560 GB in disk discounting Global Inbox.
(In both cases, DT memory footprint is about 20 GB, then the MacBook Pro has to compress memory and/or disk swap).
Apart of DT, in general, mini performance is a little bit higher but it is not worth replace one by other… if you don’t have half terabyte of databases in DT.
RAM is indeed a very important factor (especially as it’s not upgradable anymore), otherwise the difference between the two machines would be hardly noticeable.
I, too, yesterday upgraded a late 2019 16" MacBook Pro to an M2 Pro MacBook Pro with 32GB of RAM and…well, let’s just say, rather a lot of disk space.
It’s certainly true that DEVONthink is wonderful with the extra power (my previous MacBook Pro also had 32GB of RAM).
I post only to mention one small thing that may help others. If you upgrade for the first time to Apple Silicon from an Intel Mac do check DEVONthink → Install Add-Ons…. You’ll almost certainly be offered an update to ABBY FineReader OCR which, once installed, will resolve an oddity otherwise present in DEVONthink → Settings → OCR where you may otherwise find the PDF Resolution set to 0.
