Crack has the reputation of being the gold standard, and short of some kind of big n study (which doesn't exist), I don't see a reason to disbelieve it. Most who use it seem to kill PAT. It's as hard/slightly harder than the actual DAT, which was comforting to me.
I don't know what the difference is, because I didn't pay for bootcamp. You seem dissatisfied with it, though, so not sure why you're defending it unless I misread you? Sunk cost fallacy is not your friend in DAT prep.
Your original post was very broad. Anyone who has put in significant work has a specific problem section or two. If you're bad at all/almost all of them, you just need to put the work in, i.e. researching strategies on the forums (search bar - tons of threads on this)/youtube/various PAT prep sites (working smarter) as well as just running through practice mini-tests (working harder).
In short, I used kaplan bluebook strategies along with my own shorthand (you need a quick/compact way to write the cube face numbers on a page) and tested it against crack dat. I gave myself permission to skip in a couple sections if the shape was more trouble than it was worth (keyholes and shape folding), and I set a hard time limit per question depending on the section after which I'd skip (very short for angles [long time staring at angles is pointless], longer for keyholes, etc). There's a way to figure out 95% of holepunch problems in seconds, which frees up everything else (youtube it), and again, write down all the block faces before you answer the questions.