I'm going off the top of my head here since I don't have a copy in front of me, but I remember some very persuasive arguments and citations in the (terribly titled but otherwise quite good) book Extreme Productivity by Bob Pozen.

Basically, Pozen's cited studies found the steady approach pays off on basically every dimension you'd care about (including quality and quantity of the work, efficiency, and decreased various badness). I found it pretty persuasive and switched from working in intense bursts to a more methodical way when writing, for the next few years, and it worked well for me. I got the time it took me to write a 6000 word essay down from ~40 hours to the 12-18 hour range, quality was better, and it was less stressful.

Doesn't necessarily generalize, and I'd speculate it maybe generalizes least for things that benefit from being at some critical mass threshold for a short period of time (say, like, an auction). That part is just speculation thought.

I think this depends a very great deal on what your personal definition of a 'short burst' is. I know some people who will overwork themselves for weeks, then burn out and be unable to leave bed for weeks; this seems bad. 

On the other hand, when I hyperfocus on something for several hours and then take the rest of the day off, I know I get significantly more and better work done than when I sort of idle away for sixteen hours, half-working and half-scrolling-Twitter.

I'm not sure why cycles on the scale of weeks seem much worse to me than cycles on the scale of hours, but one hypothesis I have is that it's about avoiding the lows going below a certain threshold. If I work very hard and am tired and hungry afterwards, that's fine; I'll rest and recover. If I ever reach a state where I'm too tired and hungry to be able to cook a good meal and go through some bedtime rituals, then I'll stop eating/sleeping properly. Once you hit a local minimum, you can be trapped there in a vicious cycle where you don't have the energy to take care of yourself properly, and you don't have any energy because you aren't taking care of yourself properly. Big highs & big lows are fine so long as you can recover from the big low and get another big high, but above a certain threshold you can't dig yourself out of certain holes without help. 

If longer stretches of peak productivity produce worse burnout, then perhaps the key is keeping those stretches short enough that the burnout doesn't cross that threshold?

I tend toward the short burst category, but see huge gains in productivity when I manage to slow myself down. So that's a data point.

Then there are certain situations when you have to act in a short window of time to not let the opportunity slip - say, Xerox PARC burning up with creativity during three years when the stars aligned to make deep research on computers possible, which probably accelerated the development of the PC substantially. One could argue that SpaceX is in a similar position now, where its worth burning out to advance before the window closes on a society capable of going multiplanetary.

Definitely depends on the type of work for me.  I can do 8-12 hours of repetitive manual labor vs 1-2 hours max of something cognitively intense (like computer programming).   

FWIW I sometimes do steady work averaging 4-5 hours solid, focussed work per weekday (excluding breaks and semi-work like admin), which I think is the most many people achieve, as so much of their 'work' time is spent on chatting, breaks, pottering about, admin etc.

And sometimes I do intense bursts of many more hours per day for a few days or weeks, which tend in practice to be followed by recuperation periods in which I do much less than normal.

As it happens I've been recording complete data on my work & other hours for the last 9 years, so should do an analysis of which of the above methods works out more productive. That said it's affected by the fact that the intense bursts will be on more motivating projects.

In my experience, I don't usually get to choose. I am ineffective and distractable when I am unmotivated, so the vast majority of worthwhile work occurs when I am motivated. Over time this has led me to "ride the tide" of motivation when it is present, and not to force it when it is not. For externally imposed work, pushing work off to deadlines has caused much less frustration than attempting to start early and work steadily. If you are not constrained by motivation, it seems like working slowly and steadily would be preferable for large projects, because it would not be as likely to burn you out. It would also be preferable for jobs, because employers and customers prefer steady, predictable progress. For small projects, and projects that benefit more from "inspiration", it is possible that short bursts would be preferable instead, because there is less risk of losing the "spark" before finishing.