There's no guarantee that software 20 years from now will be compatible with our current RAW files at all. But assuming they are, I would put money on lens profiles and distortion and CA correction continuing forward. For instance, I know Canon now has lens profiles on their newer cameras to reduce vignetting. It isn't just m4/3 doing it, you know. Lens corrections in RAW software are quickly becoming an industry standard feature.
Do you think that I don't know how to perform distortion correction even without a camera profile?
Well, I agree this introduces a grey area. But mainly what we'll do, as I've always done in lens reviews, is show test data without corrections applied, along with examples of corrected images.
It shouldn't be all that difficult to work out that, if you can correct distortion and CA in your raw converter, you don't need to worry abut those measurements on your assessment of the lens. And while correcting distortion will soften the corners a little, it's unlikely to make a huge difference in the grand scheme of things - even with highly distorting lenses MTF50s don't drop by a huge amount after correction -at the very worst, maybe 10% in the extreme corners with highly barrel-distorted lense.
You say 'we', but shouldn't that really be singular rather then plural? Or have you been appointed to speak for everyone?
I'd disagree with that - they tell you a lot. It may require a little more effort on your part to work re-assess the data in a fashion that's most meaningful for your personal workflow, but such is life. Sometimes you need to make a little bit of effort of your own.
In your post starting the thread, you said the dark corners get "pushed out" in the process of applying correction for barrel distortion.Perhaps not a defect. A defect implies an unforeseen situation. The lens projects an image circle that is smaller than the sensor. That is a design flaw.Nobody is attacking you for pointing out black corners, only correcting you for calling them defects
I am surprised that you have come to this conclusion after saying that you do not want to buy a lens that requires software correction to work properly. As many have explained, to optically correct aberrations, a lot of glass is needed. In order to keep the whole system compact, m4/3 in a non-apologetic way designs their lens to be corrected by software. Such correction is so well integrated that what you see the corrected images in the VF, JPEG, and RAW. In fact, it is impossible for a m4.3 user to tell you whether black corners can be seen with the 9-18 lens. Considering this is a very wide lens and that it is so small and light, it must be heavily corrected by software. I thus suggest that you think twice before buying.
Ironically, despite your remarks, it just happens to be an option Olympus has exercised with the 12mm F2, which has a rather high level of native barrel distortion.
Are you serious. If this platform will be taken seriously then that is a ridiculous idea. RAW is the platform for serious shooters, and any converter should be applicable. Looks like bad lense design to me.
Why wouldn't serious shooters use raw converters that are fully compatible with the files created by their equipment? (Apart from special situations, such as when the person who posted this thread wants the barrel distortion in the uncorrected version.)
jagge, you've got it backwards. In this instance, the RAW converter is being manipulated to not apply software correction. If you manually defeat software correction or go out of your way to use a RAW converter that doesn't support your hardware, that's operator error , not bad design. Another car analogy: If you dump diesel into a gasoline engine because you know diesel has a higher energy density and want better fuel economy, whose fault is it when the engine sputters to a halt? Most of us would scratch our heads if someone did that and then complained about the end result, much as is being done over software corrected lenses.
Highly barrel-distorted lenses pretty much describes all wide-angle m4/3 lenses.
Speaking for myself and the dozen or so m4/3 enthusiasts who I correspond with and meet regularly.
But that's just it. If you use Nikon Capture NX or the camera's own JPEGs, your output by default will be corrected for CA, vignetting and in most cases now distortion.
I'm quite used to making that effort. Given the somewhat slow pace of reviews around here one almost has to look elsewhere to get a good sense of how the more current lenses perform.
Good to know.
Used on my GH1, and compared side by side on my E-30, the 4/3 9-18 had distinctly softer edges and corners at similar apertures. Same with the excellent 12-60mm at wider apertures. But significantly sharper in the center with the GH1. I figured this had something to do with the sensor microlens array and the closer rear element with m4/3 lenses. I sold the 9-18 and pony'd up for the excellent 7-14.
The good news: the 12mm f2 review is in progress
Do we know that yet in the case with the new 25mm f1.4? I would expect the maco lens to be very well corrected for distortion whether it was Leica-branded or not.
Not really. Sharp is sharp. They do not correct in software for sharpness (just waiting for the expected nebulous answer from one or two "experts" on this one). If the lens is sharp and it turns out software is needed to correct for distortions, well....this is the micro four thirds system, isn't it?
Well, bring it on. I' d be curious to see how much software correction Oly might inflict us for 800 S.
OR....just go buy something else and find something new to worry about.
