Quote:
Originally Posted by Ledgem
The Nokton, huh? I've read the reviews on it and I think you're going to have a lot of fun.
Some would say that shallow DoF is cheap and overused... personally, I like it. I don't care to have the background blown into one big color, but I like it when it looks almost like an undetailed painting. You can always close the aperture a bit, but you can't open it beyond the maximum! Some day I'd like to get a Nikon D700 or D800 for the DoF potential...
|
Well to be honest there's quite a bit of pure bokehporn out there. Bokeh for the sake of the bokeh isn't really something I'm all that into. Though I must admit I'm guilty of coming up with some of it myself. Okay quite a bit of it. Then again shallow DoF is quite useful in separating the subject from the background or mending a crappy background a bit. I've found that especially in street photography a fast lens really helps since controlling background otherwise is near impossible. So does being partial to both ways make me impartial, bipartisan or just schizophrenic... Go figure. Anyway it has it's time and place so gearing up for the occasion is in my opinion a good enough idea.
D800 indeed seems like one hell of a camera. Even as a bit of a canonist in rehab I do have to say, assuming I had the budget which I don't, that would be the one I'd go for. Just the overall expense of body, necessary accessories and a fair selection of lenses that in terms of quality do justice to that monster of a body being somewhere closer to 10k€ puts me off quite a bit. No... Not a bit. A lot!
Quote:
There are methods out there, it's just a matter of implementing them. The best implementation IMO was patented by Fuji and is used in Nikon's V system: the phase detection sensors are placed directly on the imaging sensor. You give up a few pixels (but not a notable amount - nothing that would affect the marketed pixel count), but you don't need to add anything to the camera body itself. Then there's Sony's implementation with their SLT line of cameras, using a non-moving mirror to redirect a bit of light to the phase detect sensors. The downside is that the body still needs to be relatively thick and there is minor light loss to the sensor. Lastly, Panasonic had patented (but never implemented) a phase-detect solution that would exist in a lens adapter.
It's also possible that they'll just ignore and abandon their 4/3 lenses, reproduce them in µ4/3, and forget about phase detection all together. I don't know that such a thing would make financial sense, but it's up to them. I would think that the ideal system would utilize both PD-AF and CD-AF, relying on PD-AF for the initial fast calculation and then handing off to CD-AF near the end of the focus sequence. That would give the speed of PD-AF and the accuracy of CD-AF. CD-AF has become pretty fast (with certain lenses), but it seems to me that it begins to slow down dramatically as the lighting gets worse. PD-AF loses a bit of speed in poor light, too, but the difference isn't as apparent.
|
I'm still quite puzzled how Olympus has managed to make lenses designed for PD-AF work so badly on CD-AF cameras. Actually tried the 25mm pancake on a E-P3 at work today just to know what I'm talking about.
I suppose I would have learned to live with it but still it was painfully slow. Okay the 25mm pancake isn't a fast focusing lens to begin with. I've been in the impression that there isn't such a huge difference in technical implementation of the actual motor between µ4/3 and 4/3. To my knowledge all µ4/3 lenses use micro motor so by applying some logic they should be just as slow as mm 4/3 lenses and slower than SWDs.
So either the camera's AF commands are different and translated into actual motor commands in the lens itself (which would make sense since the motors can be quite different) and the implementation of 4/3 commands on µ4/3 bodies has been made by a retarded chimpanzee or Olympus have purposefully crippled the AF performance of the 4/3 lenses. Since the focusing on a µ4/3 lens would by my logic be done just as incrementally as with 4/3 lens I'm actually starting to suspect the latter. Oh well... It's always good business to get your customers to buy more stuff that they wouldn't in ideal world really need by being a Dick Dastardly and not actually getting caught doing it. Those b******s
I'm not quite convinced that going back to phase detection is actually any solution. The only reason I can see why phase detection has been used for so long is that CD-AD hasn't really worked all that well and now it very very nearly does. Okay the low light performance isn't quite on par with PD-AF but it's getting quite close. PD-AF systems are actually quite pricey and the accuracy leaves quite a bit to be desired compared to modern CD-AF systems.
And don't even get me started on those semi transparent mirror abominations that Sony's nowadays pushing. Suffices to say in my opinion one of the most retarded ideas ever. They didn't even manage to reduce shutter lag to any measurable degree so the only benefit is the reduced shock, which is only a plus if you don't know what you're doing, and slightly quieter operation at the cost of 30% reduction of transferred light to the sensor. Not to mention that because the they needed to maximize the light transfer to the actual sensor the AF-system ends up making due with quite a bit less light than in a traditional PD-AF DSLR. None of which a trades I'd be willing to make. Okay Live View PD-AF is neat but only if the CD-AF sucks as much as it does in many DSLRs and following/continuous focus should still have a slight advantage over CD-AF. But still trying desperately to figure out why not to use CD-AF instead of trying to fix its shortcomings seems a bit moronic. Simpler is almost invariably better and CD-AF mechanically is simpler and not by a small margin.
It's nice to talk shop at asuki again.
_____
Anyway got through the rest of my Japan photos quite a bit sooner than I expected.
Japan Batch 2