Following up on a recent FM thread on use of Live View and MLU:
http://www.fredmiranda.com/forum/topic/904829a post by photoguy referenced this entry where a researcher using a Canon 50D was trying to optimize Live View use for photo microscopy
http://krebsmicro.com/Canon_EFSC/index.htmlHe shows examples and communications with Chuck Westfall of Canon describing the sequence of events when using Live View Silent Mode. On the 5DII you get to this via Menu, Live View/Movie func. set, then Silent Shoot with a choice of disable, mode 1 and mode 2. Chuck's description is more detailed, but to summarize:
In Silent Shooting, mode disabled, when taking a picture the mirror stays up, but the shutter closes then opens to initiate the exposure. It closes again to end the shot and reopens to reinitiate Live View.
In the Silent Shooting mode 1, the mirror also stays up the whole time, but the shutter does not close nor reopen to initate the exposure. Rather the shutter stays open and the exposure is initiated electronically, with a soft click. The shutter then closes to end the exposure, then immediately reopens to reinitiate Live View. Continuous shooting is slowed down.
Silent Shooting mode 2 is similar to mode 1, but at the end of the exposure the shutter closes, but does not reopen until your finger pressure is off the shutter button. Continuous is not possible.
Since I have 4 manual focus lenses I use Live View very frequently for landscape shooting on a tripod. I ran a test to see if Live View Silent Shooting Mode 1 or 2 would make a improvement by reducing vibration as both modes eliminate closing of the shutter to initiate the exposure.
Thus I set up a situation to amplify visual detection of MLU type vibration. I took the longest lens on hand, Canon 400 f5.6, added a 1.4X teleconverter and then slowed down the shutter speed with ISO 50, a circular polarizer and a 2 stop ND filter. I focused with Live View 5-10X on a house across the valley from inside my house, more than 1/2 mile away. Manfrotto 055CXPRO4 Carbon fiber tripod was used with RRS BH40 head. I used a remote in addition to a 2 sec delay to make sure no vibration was accidentally added by some other mechanism. At f8 I was at 1/15 sec and at f16 I was at 1/4 sec, bracketing the shutter range most susceptible to MLU type vibrations, 1/30 to 1/2 sec. All shots were in RAW, with some sharpening added during conversion, all at the same level of course.
Here is the full frame shot reduced in size for posting:
http://www.fototime.com/37B819B5F4F52C5/standard.jpgWhile there are some nearby OOF branches framing the shot, I will evaluate only 100% crops of the center deck of the house.
Here is the crop of Live View Silent Shooting Disabled:
http://www.fototime.com/027A7AE1813B6EC/orig.jpgNotice the bricks of the chimney, the cable running next to it, the wood crack in the upper right and the dark leaves of the potted plants on the right. In particular notice the wire tomato cage (funnel shaped).
Next is the same shot, same exposure, same focus, seconds later using Live View Silent Shooting Mode 1.
http://www.fototime.com/CE27F4FE75A364B/orig.jpg Sharper. The tomato cage is much more detailed, and this is of a 2mm wire from >1/2 mi away.
Here is the same exposure, same focus, using Live View Silent Shooting Mode 2.
http://www.fototime.com/DB71D4332E5E6F4/orig.jpgthe chimney and the plant leaves look pretty similar to Mode 1 but the wire tomato cage appears clearer in mode 2.
Here is the same shot at f16, 1/4 sec. Some manual refocusing between these next 3 shots and the previous 3, so don't make that comparison. Live View Silent Shooting Disabled:
http://www.fototime.com/5C6E9C45651A235/orig.jpgAnd the same exposure and focus seconds later using Live View Silent Shooting Mode 1
http://www.fototime.com/A40532DE42085E9/orig.jpgAnd again the same exposure and focus seconds later using Live View Silent Shooting Mode 2
http://www.fototime.com/B96E9231E48C4C9/orig.jpgIn this sequence we can easily make out the Live View Silent Shooting Disabled is not as crisp as Mode 1. Mode 2 is very close to the detail of Mode 1; the tomato cage is just a bit sharper in Mode 2.
In summary, the Silent Shooting Mode 1 or 2 provides an small but consistent improvement in image sharpness over Live View Silent Shooting Disabled. This result is also consistent with the photo microscopy paper referred to above. Thanks to Charles Krebs for the write up and communication with C. Westfall. This result is also quite logical in that Silent Shooting Mode 1 & 2 use electronic shutter for first curtain, and avoid the additional mechanical closing and reopening of the shutter that is present in the Live View Silent Shooting Disabled. Traditional MLU operation is a bit in between in that it does not require closing of the shutter to reopen it for the exposure as in Live View Silent Shooting Disabled, but does involve mechanically opening the shutter as opposed to electronically initiating exposure. The Krebs paper does support the Live View Silent Shooting Modes as being sharper than traditional MLU sequence, but I did not confirm that. Cameras with Live View Silent Shooting modes, currently 40D, 50D, 5DII, and 7D, would probably be better off not turning off Live View to return to MLU sequence. Unfortunately this feature is not enabled on the current 1D series bodies. My understanding is that on Nikon bodies, for Live View implementation the mirror momentarily swings down for an exposure reading immediately prior to exposure. For landscape photos, MLU certainly would be advisable with Nikon Live View as the mirror slap would be much greater than the shutter induced vibration shown here.
The results presented suggest that for Canon Live View Silent Shooting Mode 2 may be a bit better than Mode 1, but the difference is small, if any. There is no obvious reason for it to be different as the Modes 1 & 2 differ in what happens after the shutter closes to end the exposure.
It would seem that Canon has eliminated the need for a dedicated MLU button as Live View Silent Shooting mode 1 or 2 is better at reducing shutter opening induced vibrations over MLU shutter sequence. In addition, Live View 5-10X is the more accurate method of focus adjustment for tripod shooting over viewfinder use, even with optional focusing screens.
Mike K
