Now that I've had the chance to try out my new-to-me 12-100 f4, I'm a bit disappointed. I was expecting it to be much better based on the unanimous rave reviews, but it doesn't seem to be nearly as sharp as my Panasonic 12-60mm f2.8-4. I guess I was expecting too much for a lens with such a wide focal length range.
My guess is that your lackluster experience is due to sample variation. I have tried two samples. The first was very good - about equal to the Olympus 12-45mm f4 "Pro" at all overlapping focal lengths. The second is truly excellent, better than the first sample by a lot at 12mm, and much better centered at various focal lengths, so that there is no discernible tilt or weak corners at any focal length. Truly exceptional for any mid-range zoom, and even more so for an 8x zoom. The first was brand new, the second is from lens rentals (which I decided to buy). By comparison, it's more consistent across its focal range and probably sharper at the pixel level than either of my two Sony 24-105mm f4 zooms that I've used on the Sony A7RIV, which I consider to be an excellent mid-range zoom.
Thats interesting! I've had the cheaper Lumix 12-60mm. I found it plenty sharp and wasn't that very impressed by the supposed modest improvements for the leica versoin which was supposed to be just as sharp.
Either both Panasonic lenses are really sharp, or I have noticed sharpness in the field. Do you have A/B tests? I'd be curious to see.
Hi - No AB tests with the 24-105mm. A lot of AB tests with the two samples of the 12-100mm and 12-100mm vs. 12-45mm (which was also the better of two samples I tried of the 12-45mm, one rental and one new).
Sorry, I very much enjoy the forum, but I don't have the time to post samples.
I do a lot of testing before I buy a lens or after I buy a new lens. I always rent one to compare it to if I'm going to buy and then do various decentering/tilt tests based on a combination of Fred's and Jim Kasson's published methods. Combining the two methods works best for me. I know it works because it's repeatable. I test lenses in 9 quadrants (like Jim Kasson) vs. 4 (like Fred), but use an infinity landscape target (like Fred) vs. a test chart (like Jim Kasson). Corners can be sharp but there is still quite frequently tilt from side to side or top to bottom and tilt from side to side or top to bottom, in my experience, is more likely to adversely affect an image than one weaker corner (most lenses have a weaker corner when viewed at 200% though few have a weaker corner at 100%, and zooms vary throughout their focal range; it's always a matter of degree in judging a good vs. mediocre lens). Then I always take a lot of real world AB comparisons to confirm what I'm seeing with the 9 quadrant tests. Alot of effort, but then I have a lens that I know performs well for my purposes and I stop thinking about it.
The comparison with pixel level sharpness of the 12-100mm vs. 24-105mm (on different cameras obviously) is just my impression from looking at a lot of images at 200% on my 27" LG Ultra Fine Display for my Mac. I've used the Sony A7RIV for about 1 1/2 years now (and the A7RIII for about a year before that), and the OMD EMIII for only about two weeks (although I had EM Mk I's for about five years before the Sony A7Rs).
httivals wrote:
I do a lot of testing before I buy a lens or after I buy a new lens. I always rent one to compare it to if I'm going to buy and then do various decentering/tilt tests based on a combination of Fred's and Jim Kasson's published methods. Combining the two methods works best for me. I know it works because it's repeatable. I test lenses in 9 quadrants (like Jim Kasson) vs. 4 (like Fred), but use an infinity landscape target (like Fred) vs. a test chart (like Jim Kasson). Corners can be sharp but there is still quite frequently tilt from side to side or top to bottom and tilt from side to side or top to bottom, in my experience, is more likely to adversely affect an image than one weaker corner (most lenses have a weaker corner when viewed at 200% though few have a weaker corner at 100%, and zooms vary throughout their focal range; it's always a matter of degree in judging a good vs. mediocre lens). Then I always take a lot of real world AB comparisons to confirm what I'm seeing with the 9 quadrant tests. Alot of effort, but then I have a lens that I know performs well for my purposes and I stop thinking about it.
These ad-hoc testing methods are pretty much a waste of time, as the inherent margin of error is just too great to provide any meaningful conclusions. I suspect these testing methods are the reason so many people think most lenses are decentered, and not because the lenses are actually bad.
The reason I believe a lot of zoom lenses in particular are decentered is because: (1) I've seen it based on thousands of images I've shot on tripods, etc.; (2) I've read about it in the lens rental blogs that tests lenses on highly specialized equipment; and (3) I've then applied a modified version of Jim Kasson's test method that has worked for me over the past year or so and I've been able to replicate the same behavior for one sample of a lens against another in A/B tests using two identical bodies and two identical lenses (different samples).
That you don't accept my tests as meaningful is fully understandable and I wouldn't either if I were you. I haven't shown samples, you haven't seen my methodology, you don't know me. If you want to determine whether the lens you are using are good copies of lenses (not significantly decentered or tilted) and given that photography is as much art as science, you have to decide if it's worth it to you to spend hundreds of hours testing and reviewing lens samples. For me it is, because I find the results satisfying and I enjoy knowing that my equipment is performing optimally.
I was only responding to your comment that you're disappointed with the one 12-100mm sample you tested. If I were you I would test another sample before giving up on it, or deciding the lens isn't all that good. But obviously you should make your own decisions.
Just to emphasize my point - that you completely discount my tests based on the way I've presented them is perfectly reasonable and appropriate, and like I said, I would also if I were you and our positions were reversed and I hadn't done the tests myself.
BTW for the 12-100mm Ienses I've tested, I changed my methodology and tested them handheld using IS which adds another variable. I did it because I expect to virtually never use the lens on a tripod because that's not what I bought it for. This obviously adds two other variables (at least) - one the performance of the IS on the particular lens and body combo, and two how consistently I am hand-holding the lens between A/B tests. But if I am able to replicate clear results that I can see, then it works, for me.
molson wrote:
These ad-hoc testing methods are pretty much a waste of time, as the inherent margin of error is just too great to provide any meaningful conclusions. I suspect these testing methods are the reason so many people think most lenses are decentered, and not because the lenses are actually bad.
httivals wrote:
My guess is that your lackluster experience is due to sample variation. I have tried two samples. The first was very good - about equal to the Olympus 12-45mm f4 "Pro" at all overlapping focal lengths. The second is truly excellent, better than the first sample by a lot at 12mm, and much better centered at various focal lengths, so that there is no discernible tilt or weak corners at any focal length. Truly exceptional for any mid-range zoom, and even more so for an 8x zoom. The first was brand new, the second is from lens rentals (which I decided to buy). By comparison, it's more consistent across its focal range and probably sharper at the pixel level than either of my two Sony 24-105mm f4 zooms that I've used on the Sony A7RIV, which I consider to be an excellent mid-range zoom....Show more →
I don't think my copy is defective, it's just a general lack of sharpness, most noticeably at longer distances, similar to my Fuji XF 16-80mm f4 and many Nikon lenses I've used over the years. Based on all the hype, I was just expecting a bit more... one more reason not to put too much faith in other people's "reviews", I guess.
My copy was purchased used, and it seems fine for closer subjects, so I don't know if it's worth the cost and hassle of buying a new copy or not. Maybe I should replace it with the 12-40mm f2.8, which is an excellent lens, with the trade-off being the considerably less versatile zoom range.
I have primarily tested the two copies I have used/tested of the 12-100mm f4 at infinity. As I've said, the copy I kept exceeds my expectations. I also remember the 12-40mm f2.8 as being an excellent lens when I used it on the 16 megapixel sensors. I thought a bit about it as well, but it doesn't have the built in IS which, along with the EM1 Mark III is magical, and it's not nearly as useful of a range.
. . . One thing I've learned from using the Sony A7Rs, and that I've seen some indications of in my brief use of the OMD EM1 Mark III is that infinity focus is not always accurate when just using single spot AF. To try to minimize the risk of focusing errors for infinity/landscapes, I have magnification assigned to one button and use the magnification at 10x to focus for inifnity when I want infinity focus.
I'll be on look out for 12-100 here on FM if I can get it for a decent price.
I am fine with the size. I used their 2.8 zooms earlier and am ok. The sizes are COMPARITIVELY smaller to other manufacturers IMHO.
I was considering this lens myself but opted for the Tamron 28-200 f2.8-5.6 for my Sony A7III instead. The Tamron is about the same size, maybe a bit smaller, but covers full frame and is faster up to about 70mm and can achieve slightly shallower depth of field at all focal lengths. Plus you get the extra pixels and lower noise levels of FF and it is several hundred dollars cheaper. That said, it is missing the 24mm field of view and the IBIS is not as good as Olympus.
People seem very happy with the Oly 12-100 f4. But I'm not sure it can be regarded as providing a smaller option than other systems now.
Unfortunately for Olympus, as someone with a foot in both camps I am finding more and more of my money going towards the Sony system.
roofdweller49 wrote:
How is the IS with a Panasonic body, like a gh4?
I can get reasonably sharp results down to about 2 seconds at the 100mm end with my G9; with the E-M1 III my limit is about 3 seconds, so not a huge difference.
molson wrote:
Now that I've had the chance to try out my new-to-me 12-100 f4, I'm a bit disappointed. I was expecting it to be much better based on the unanimous rave reviews, but it doesn't seem to be nearly as sharp as my Panasonic 12-60mm f2.8-4. I guess I was expecting too much for a lens with such a wide focal length range.
That's unfortunate and a bit of a surprise. I've owned the 12-60 Leica and while I didn't have any complaints, I would take the 12-100 any day over it. My two copies have been incredible but it sounds like that's not the case with yours.
I do ALWAYS wonder when I purchase a used copy of any lens and it doesn't meet my expectations or live up to the hype if the lens had been dropped. That can really mess things up. Obviously.
