Ludwig Milde's tenth concert study for bassoon is a cantankerous, twisty little piece in C# minor. It is awkward and uncomfortable to play, and it is difficult to make the notes speak with the required rapidity. It is, in other words, Not Fun to practice.
I mention this specific étude not because it is unique in Milde's output for its difficulty, but because it's the one I happened to be working on when I was doing college visits my junior year of high school. On one such visit, I played for George Sakakeeny at Oberlin Conservatory. After working on some techniques specific to various problem spots, he asked why Milde hadn't written the thing a half-step lower in C minor. It would make everything much easier to play, and would probably sound better given the natural resonances of the bassoon after all. The answer he was looking for, and the only answer I find satisfactory, is that he wrote it where he did because it's hard.
Other musicians will, I am sure, be able to substitute in equivalent works for their instrument of choice. Whatever is uncomfortable, difficult, awkward, or borderline impossible — whatever it is that your instrument really does not want to do — you can guarantee that there's an étude out there dedicated to finding it, exploiting it, and making you do it over and over for pages and pages of sadistic exercises. Some of these pieces are, miraculously, a genuine pleasure to play, but most of the time, we work on these things not for musical satisfaction, but to hone our technical mastery, lest we fall flat on our faces when some fool composer actually writes something like that in one of our parts.
I think about that little exchange at Oberlin every time I'm in a class with composers that covers orchestration. These classes are, invariably, filled with very good advice. Those impossible trills? You really should not write them. Fast noodly passagework in the extreme upper register? This will probably end in tears. And for every one of these injunctions for bassoon, I can think of a work in the standard repertoire that flagrantly violates them. (Again, other instrumentalists should feel free to substitute their own instrument-specific struggles.)
It is absolutely reasonable to expect composers to be aware of the intricacies of the instruments they're writing for, with the goal of producing idiomatic parts that lie well under the fingers. On the other hand, it is also absolutely true that players are used to receiving horribly unidiomatic parts and making them sound effortless and musical regardless.
This tension — the tension between "I want players to have a good time playing my music, and that means being very aware of what is and isn't easy for them to do" and "tough luck, this is how the music has to go, and if it means the first violinist has to spend a few extra hours on this in a practice room, well, that's what they signed up for when they went into performing" — is always lurking at the back of my mind as I write. Because I perform so actively, I'm keenly aware of both my resentment towards composers who clearly didn't understand that what they wrote is needlessly difficult and my dedication to putting in the time to get it right, no matter how hard it is, because that's what the composer wrote. I don't want to be resented — or to come across like I don't know what I'm doing — but I also know I can get away with an awful lot, especially if the end product is good.
So as a composer, I try to strike a middle ground. There are times I've pulled back or re-written a passage specifically to make it more playable. There are other times where I've been too convinced by a first draft to change anything, even tho I know it's hard to play. It's only a rough compromise, and there's certainly no systematic evaluation of it, but it seems to be working out pretty well so far.
And as a performer? Well, I just keep practicing Milde.
Because it really is very hard.