The lower bound on disc's maximum speed is inferable
Consider these two pieces of text from the spell description:
If you move more than 20 feet away from it, the disk follows you so that it remains within 20 feet of you.
If you move more than 100 feet from the disk (typically because it can't move around an obstacle to follow you), the spell ends.
That last condition is either the straight-line distance or the total-movement distance.
Regardless, on an open field if you are 100ft away we know the disc can always cover enough distance to keep up with you in whatever a timeframe the magic considers suitable. Otherwise it could not keep up, and we'd need to be told about it's speed to know how close it gets.
To explain why this is, imagine a combat situation where the cast has a Disc following them. They instantaneously teleport so that they are exactly, but not over 100ft from the disc. Before the turn is over, the disc must follow the caster until it is within 20ft, covering the 80ft distance in less than the time it takes for a turn, 6 seconds. Any further in one go, and the spell ends. It's possibly faster (perhaps the caster is a part time monk who breaks up their movement), but this is the upper bound on how long it can take to cover that minimum distance. That is, it can't take longer that 6 seconds to cover at least 80 ft.
If it took longer it would need to break the movement up over multiple turns, and how many turns would need to be stated in the spell text.
From this reasoning it seems like we can assume the disc can travel at least 80ft in 6 seconds, which is roughly 14km/h or 9mph. This assumes the 100ft distance limit is because it couldn't move close the gap to 20ft 'in time' to satisfy the 'all movement is in one round' assumption we make.
Unfortunately this interpretation relies on the total-movement interpretation, i.e. that lower bound on speed adds a movement constraint. It means if the spiral corridor is more than 80ft in total length, the spell potentially ends as soon at the bridge goes away.
To see why, consider with the disc will not move if it's path is blocked, but with gate opened the disc now has an unblocked, and direct route to follow. This route takes more than 80ft of movement, placing it more than 100ft away on that path.
Either the spell ends dumping the curious onlookers on the ground, or the disc remains stationary seeing the 100ft path as 'blocked' for being too long, either way the passengers are likely wanting a refund.
If we ignore the total-movement interpretation, and instead use straight-line distance the disc must still remain within 100ft and follows you to be within 20ft of you. Now two things could happen, depending on the mood the disc is in (i.e., the DMs ruling):
- the disc attempts to follow you along the bridge, which it can't so remains stationary regardless of the existence of a path you didn't take yourself. This seems fair as you did not trace out the 6000ft spiral yourself, so traversing that ground would not be following you.
- no longer constrained by the concept of 'path length', not seeing the 100ft+ path as an obstacle, and not caring about following you exactly, the disc rockets around the spiral covering 6000ft in 6 seconds so as not to lag behind (using our previous argument of not being told how the movement is split up). Previously on the open field, the path and straight line were the same, so we never encountered this issue. However this means the disc is traveling at ~0.9 times the speed of sound, or 1090kmh/680mph!
Great you think, those are some dead excited onlookers (or possibly just dead), and you have their money. Win-Win.
But the issue with interpretation 2) is that as well as this not being the absolute maximum speed (there might be an arbitrary shape you could construct with longer distances, say a tightly coiled helix with just enough room for the disc to travel around and up but not over 10ft in elevation.) If the disc can travel more than 80ft in 6 seconds, then the 100ft maximum range in the spell text is not a movement constraint but a distance constraint. Plainly, we cannot argue the disc must make its movement (whether that be 80ft or 6000ft) in a fixed amount of time (despite having not way of knowing how if at all to split up the movement), and our previous reasoning about how to calculate the speed (the most it can move in a round) falls apart.
So, by my reading of the rules tex, either the disc moves at 80ft a round, or refuses to take a path more than 80ft in length if can't follow you directly. Either way limits the speed, and we've explained why the disc can't travel arbitrarily fast (if it could, the 100ft limitation becomes unusable in the calculation, rendering any method of calculating a speed moot.). Anything outside of this conclusion, and I would consider it a personal ruling for a GM.