imoi
but what's the point of such prep if on the exam this will kill me?
Great question, but there's an extremely important answer:
"Rome wasn't built in a day."
"Before a man can fly he must first learn to walk, and run. A man cannot fly into flight." -Nietzsche
Meaning - before you can solve a problem in ~2 minutes or less, you first have to understand how to identify concepts and recognize common trends, how to work through certain skills and processes, etc. You have to develop that muscle memory that comes from factoring common algebraic terms or leveraging important definitions within DS question stems or applying triangle side ratios. And that won't always come within 2 minutes right away - sometimes you have to struggle through a problem in 3:15 to finally hit that breakthrough "OHHHH, it was isosceles so I could have just..." or "Right, had I just left the multiplication/division until the end the ugly numbers would have factored out."
I'd actually go as far as to say that you SHOULDN'T use a timer in your initial phases of prep. Why rush the learning process? It's that process of struggling and actually thinking that will help you - doing a problem halfway in 2 minutes, then guessing, then reading the solution isn't nearly as helpful as forcing yourself to really think it through. The lessons you learn from having dissected a problem from multiple angles, those will serve you extremely well as you're seeing new problems on test day.
On test day you have to be quick, but quickness comes from having trained yourself to think critically, recognize clues, etc. 3:15 on a problem when you learn an important lesson or build those critical skills is time really, really well spent, especially when compared to 2:00, a guess, and a read of a technically-written solution.