How many use Live View or think they would?

I think it might be useful shooting macro with a little surge. I would just switch to manual focus and time the shot for the point of best focus.

See, the problem with this poll is that "LiveView" right now is not at all a level playing field. There are differing implementations, MOST of which either heat the main sensor, require an extra mirror flip thus add shutter delay, or have slower or even manual autofocus requirements. Who in their right mind would want to use that underwater? (Well, maybe on a wreck shot with a tripod and remote triggering, but otherwise...)

 

Olympus had a faster implementation in the E330 (with still some limitations - no real WB or compensation visible in the 'live view' even with A mode) but abandoned it so far. (Micro-4/3rds might bring it back to one degree or another). Sony has a faster implementation now with the A300 and A350. But otherwise, it just ain't a "real" feature yet.

I could imagine a situation where Live View could be used just like a viewfinder is now, with a hood around it etc.

I think it might be useful shooting macro with a little surge. I would just switch to manual focus and time the shot for the point of best focus.

How good is LiveView for seeing if something is perfectly in focus or not?

I could imagine a situation where Live View could be used just like a viewfinder is now, with a hood around it etc.

Yes, that's how I use it - even without a hood. I'm not too fond of the idea of putting blinders on underwater. But still, it's only usable as a framing aid equal to the VF if there's no added delay caused by using it instead of an OVF...or else you may as well be using a Point-n-shoot with a fairly good lens, and taking lots of pictures of fish-butt or having to anticipate way more than a DLSR user does. So again, its the inequal implementations that kill it for more use I think.

How good is LiveView for seeing if something is perfectly in focus or not?

Depends on several factors, including the screen (brightness, resolution), and again, how fast the shot is taken after you pull the trigger. You could ask the same question about an optical viewfinder, couldn't you? If the view is too small and/or dim (because of conditions, not implying anything about OVFs themselves), while shooting shallow DOF shots, it's hard to tell exactly where the super crisp focus point is. If your mask or your personal eyesight makes it hard to use a particular OVF, it can be difficult too. On a tripod, in a controlled setting, either would work great: it's easy to manually toggle in and out a tad and visually tell when your OVF view is focused where you want. On a LV, most all of them can flip the mirror out of the way and use the 'main' sensor, to permit manual focus with zoomed in views, which help you tell even if the screen isn't as "crisp" as the optical view. But again, neither of these applies terribly well underwater in most cases, since we don't often have tripods or otherwise the positional stability in ourselves or our subjects to suffer the extra mirror flip down for metering and back up for the shot.

I really think the critical factors (as gobiodon indicated we should identify) is a Live View implementation useful for UW shooting would be:

--no additional delay introduced vs. shooting with the viewfinder (applies for both autofocus speed as well as avoiding additional mirror flips or the like)

--no impairment to autofocus ability vs. shooting with the viewfinder

--adequately bright, clear, and "live" (as in frames-per-second or screen refresh interval) monitor for use as the "Live View" viewfinder

--LV screen correctly mimics a viewfinder in terms of showing depth of field and framing (e.g. not grossly under or over cropped)

--no adverse impact on the quality of the captured image in terms of added sensor noise caused by heating of the sensor, or whatnot

--no or (minimal enough so as not to be burdensome) impact on battery life

Of all of these the only implementations that can sort of deliver all the above are the sole, now orphaned, Olympus E330, and the Sony A300 / A350. With perhaps some caveats on screen brightness and clarity, and battery life, which could always be improved. Note that if there's no "difference" in LV vs. using an optical viewfinder, then other factors become more important overall such as the camera's autofocus performance, noise and image quality, etc etc. Many here would not buy an Oly E-330 even when it was new because of the whole small vs. FF sensor argument, or other factors relating to AF speed/zones, dynamic range, noise, etc. regardless of their doing LV "right". So assuming there was a standard definition of LV that everyone followed, then the discussion would again be just about other features and some would use it and some would not. Then it becomes a 'shooting style' question. Instead right now LV seems to be a buzzword to most manufacturers that's only useful on a tripod or in a studio, and really has no bearing outside of that. (I'd actually say it's worse - the halfarsed implementation seems to be actively giving people the impression that "LV" as a concept is poor or useless, vs. just "this LV implementation ain't great".)

Other bonuses that should be possible with an electronic viewfinder would include live white balance or other tonal effects (more for JPGs than RAW shots), live histogram, continuous live metering perhaps, etc. That's why I think Oly's Micro 4/3rds announcement might presage a return of an E-330-like future DSLR from Olympus, as research necessary to make the mirror-less design work just might trickle 'up' as well.