I think I would need more information:
1. How big is X in relation to the blinds. If X=SB - then 3 to 1 is probably too high (because the "big" stack here is still inconsequential to the rest of the table and to the blinds - playing any hand is all in). In this case the closer to 1 to 1 the fairer you would get in my opinion. On the other side of the equation, as X becomes larger and larger (say 50+ BBs), the skill level becomes more and more important. Since you've already stated that the players are equally skilled - this seems like the case where 3 to 1 is the better route.
2. On another side of the coin is the value of X vs 3x. It could only take 2 hands for the situation to go from X to 3x all the way to 4x to nothing. Although that situation is unlikely barring some cooler type hands - it is still a factor to consider. After one double up - the big stack has just evened out entirely.
3. Position on the table and the other players around. If the short stack is followed by a couple of weak players - it would stand to reason he'd be able to increase his stack relatively easily; by contrast, if the shortie is followed by the big stack - that decreases his odds as well. Stronger other players in position related to the others is a pretty big factor.
All of that said - I would probably say smaller stakes are more reasonable here.because the variables involved are relatively unknown and unpredictable. In this particular case, as an observer, I would think 2 to 1 is a good and solid agreement; as the short stack - I would be happy to take 3 to 1