Nice point. I've been thinking about posting a thread to this effect, as I have been recording often. My thread title was going to be something to the effect, how do you drum knowing your sound is going to get processed like involving compression.
Well to a great degree, you don't know what's going to happen downstream in the production process. You might have an idea... and you might be involved through the whole process... but you very likely won't be. And that even holds true for recordings where you are a member of a band... it all varies. But again to a great degree, you don't know.
As for how can you play... IMO just like always, you make decisions and play based on what you are hearing and what is being communicated to you by the artist, the producer, other players... you just do your best with what's in front of you.
And then for me, after that... I pretty much let it go. (Again, unless I'm engineering or producing, as those "hats" of course keep me involved all the way through.) Making music is typically a collaborative effort - and making recordings are even more so.
I always just focus on what my contribution is supposed to be - do the best I can - and then wait and hope, I'm pleased with the end result. And usually I have been.
The thing about compression IMO is it compensating for the fact that the drums are going to get squashed. The effect of cutting the cymbal off using compression is basically preventing the wash of the cymbal from overwhelming a weakly reproduced bass drum, which isn't an issue live, as you have plenty of headroom, the wash of the cymbal may not even be heard.
I don't know that (like in this case) compensating for a weak anything is the point. I think it was more more creating punch. The beginning of a cymbal crash is already a lot louder than the after ring - compressing for punch in the attack can make this even more so. In order for the crash to make its statement, then get out of the way, so other parts of the music can be better heard.
That being said a compressed sound is going to bring out certain characteristics of the drums that you might not be aware of. For example, the decay on the snare is going to be much longer, you may get more detail out of the hihats, and they may sound completely different depending on the sound engineers interpretation, there may even be additional rhythmic complexity there due to compression timings.
What you are describe is one way to use a compressor - and that is to even out a sound - which will lower the volume of the attacks to create a more consistent, more constant sound. This will bring out the sustain - at the expense of the attack. But the end result can make a sound seem more present - more in-your-face.
To get this effect, the compressor is set with a very fast attack time - this makes the compressor clamp down quickly on the peaks of the sound - and making them closer to the same volume of the average sound. This reduces the overall dynamic range - which allows the whole sound to be turned up (without running out of headroom).
But there's another way to use a compressor that is almost opposite of that - and that is with a slower attack setting. This means at the beginning of that loud note - crash, snare hit, bass drum hit - nothing happens, the full loud attack is heard. Then depending on the attack time setting - very shortly after that initial attack, the compressor kicks in and turns the sound way down. Do this to extremes - and you can reduce a snare drum to a strap TACK with nearly zero decay or after ring. Of course, extreme settings aren't the only choice - so just a little nudging of the attack is possible as well.
The reality is people have been using and usually combining both of these approaches on recordings for decades.
It might seem that the two techniques would cancel each other out - and if done exactly the same by the same amount that might be the case. But it very common to use different processing for different purposing and then blend them together to create end results that are quite spectacular.
It is a very very deep art. One that as an engineer-by-necessity, I struggle with and do the best I can. All the more reason for me, when I'm involved in a project as just a drummer, I focus pretty much entirely on that - producing good sounds at the kit and then musically figuring out and executing the best parts I can come up with. And try to literally not even think about what the engineer is doing.
