Larry Colen lrc at red4est.com Wed Sep 19 15:31:43 EDT 2012 > On Sep 19, 2012, at 11:17 AM, Paul Stenquist wrote: > > > But improved autofocus will be even nicer. If it's truly a step up, > > the K5 IIs will take the lead spot ahead of my K5, and the current > > backup, a K7, will go on ebay. > > > I recognize that I tend to take photos in conditions that are a bit > more challenging than most people face. I've spent the past couple of > days going through my photos from last weekend. I'd say that I lost > the most shots through mis-focus. Unfortunately I can't easily tell > in lightroom which were manual and which were auto focus. The biggest > camera change to improve my keeper ratio would probably be more focus > points, each one restricted to a narrower range. I lost several > photos because the camera focused on someone closer to the camera that > was just barely in the frame. Then, of course, there is the ongoing > problem with focusing on the microphone.
Often suggested technique to deal with those cases is to set the AF to use the central point. That can be done for centered objects or for static ones even when they are off center (AF, hold, reframe). That can work for shooting single musicians who are more steady, and/or typically are in the middle of the frame. The problem with that when you are shooting dancing couples is that in many cases, the central AF point is right between the dancers. A possible work-around that is that you switch AF to a point that you choose off center, where you expect one of the two to be. This is not universal and requires switching (so it is not suitable if you are trying to shoot several couples within one song), but it is good, e.g. when you are trying to get a good shot of a particular couple and taking a series of their shots with simillar framing and camera orientation. I've done this with *istDS, and it seemingly worked. I haven't done it much with K7, as its frame-based AF engine improved drastically. But I am considering using that again at some events. > > Beyond that, faster accurate focus in stupid low light would really > help. Dancers just are not considerate enough to only move > perpendicular to the line of focus. I feel for you Larry! Those damn dancers! Shoot them! ;-) > > One feature that would really help me would be different exposure > metering modes, one where I could say to not meter on the bright back > lighting, another, where it would ignore dark backgrounds (though that > is more a lightroom issue). I mostly just shoot manual exposure > anyways, but sometimes when people are moving around from light to > dark areas, I don't have that luxury and have to go to TAv. When there is a large variation of the light throughout the frame, and the camera either cannot cover the full dynamic range, or you don't care for the extreme parts of it (direct light, dark background shadows), you can try a very similar technique as described in relation to AF above: use the central point metering. I haven't tried it myself yet, but in principle you can couple the metering and the AF points (in the menu), - and use the same approach for dancers who are off-center. You can play with that when not "on assignment". :-) > > There were also a bunch of pictures that weren't "lost" per se, but > were rougher than I'd like because in order to have a fast enough > shutter speed for lindy hop, I was shooting in the ISO 12,800-25,600 > range. An 8MP sensor with the latest technology would do me more good > than a 24MP sensor. I think you mean that the noise level from a larger pixel is lower. I didn't think about it in full depth, but I would assume that averaging the signal, say, 3 adjacent pixels while downsampling from 24 MP to 8 MP, if done correctly from the RAW format, should produce close, if not equivalent result. It might, actually, produce even a better result under certain conditions (that's my guess only). But then the question is whether the software you are using (e.g. LR) is doing a good job in averaging those pixels in a right way in the process of downsampling. After all, - you see a similar noise-smoothing effect on a 900pxx600px image, where the noise is not that pronounced. > > Well done mirrorless technology could alleviate a lot of my problems. > Without the mirrorbox and needing retrofocus lenses, I could get > faster wide lenses. When focus is an issue, a sharp 30/1.4 on APS > would do me more good than a FF 50/1.4. Likewise, for manually > focusing in the dark, liveview, and I imagine focus peaking , would be > an immense aid. > I am not sure what you mean by "retrofocus". Are you talking about the crop-factor of the lenses on FF and APS? If that guess is correct, it's just a question of the sensor size on the mirroless camera, not the fact that it is mirrorless. I suspect I am just not understanding what you are trying to say here about the mirrorless. Best, Igor -- PDML Pentax-Discuss Mail List PDML@pdml.net http://pdml.net/mailman/listinfo/pdml_pdml.net to UNSUBSCRIBE from the PDML, please visit the link directly above and follow the directions.
