FS7: How about this sony 18-110 lens

dipstik said:

The downsides of this lens:

1. Lots of moving parts. Will the internal electronics still work after 3-5 years of heavy use? Hard to say. I'd have much preferred a fully mechanical lens (the aperture and focus won't change unless powered by the camera) so for now I'm considering this a red flag.
Click to expand...

This is my main reason now to worry on this lens. To be fair my cra_p_y sonys 18-50, 18-200, 16pancake all still work and I take no care of them at all.

Overall I dont really trust electric lenses and also feel that the coatings can be very delicate on some modern lenses.

The EX1 had a rep for its lens going wonky (similar floating element tech?) .. again to be fair mine has been fine once I learned how to calibrate it in the camera menu.

--
Buy care for, sell on after 36months, buy another from a dentist rinse, repeat -less than £500pa 'rental'
 
If I really had to pick one of the lenses that Lenny compared in that video I guess it had to be the Sony. Just because it have the best focal range and works with the FS7. But at the same time it wouldn't bring me more jobs or allow me to take on more work. Because my set of Canon L lenses already full fills everything. I'd rather invest in a FS5.
 
morgan_moore said:
Overall I dont really trust electric lenses and also feel that the coatings can be very delicate on some modern lenses.

The EX1 had a rep for its lens going wonky (similar floating element tech?) .. again to be fair mine has been fine once I learned how to calibrate it in the camera menu.
Click to expand...
My EX1 once had those problems. I sent it to Sony in Munich, they replaced the whole optical unit and it went fine the next 3,4 years. Then again, the problems came back, it did not react as it should. Because of that i became very sensitive about those Fly-by-wire or whatever electronical stuff in a lens!
I don't trust it!
 
I just put one on order with B+H Friday. Everything I've seen tells me this is the glass I need for 85% of what I shoot. BH said they were out of stock Fri when I ordered but today looks like they got some in.
 
Thanks Klaus,
That's more like what I would expect. However you should always check par focal quality at the most open f stop in this case f4. That's where the problems show up.

If you really want to be ambitious and post another comparison between the 18-110 and the Sigma 18-35 at f4 that is focused on a scene with lots of detail in the distance that would lay to rest how sharp the 18-110 is because we all know how good the Sigma is.

I don't see the CA that Osslund is talking about but maybe my monitor isn't good enough to reveal it. I'm looking at an iMac 27" display full framing at 1080. Klaus probably has a better look at the original files. Another test that should reveal more substantial CA I would be to rack focus slowly just before and after those branches at different focal lengths. Virtually any sub $10K lens should have a fair amount of that though perhaps at f4 it won't be bad. Again though best to test wide open. By f 5.6 most any lens good.

Anyone else see much CA?

An interesting question for me is whether there is any way Sony has for calibrating the back focus on this lens. I seriously doubt whether every Sony E mt camera has exactly the same distance to the sensor and even tiny differences throw back focus off. That's why all PL zooms should have a back focus adjustment or the PL mounts should have shims available . The Sony F3 had a back focus adjustment for its FZ mt though I don't know if the F5/55 does. The Sony EX-1/3 got around this with an electronic back focus tool. I would have expected there would be one for this lens but haven't heard about it unless the lens actually communicates with the sensor itself. That would be pretty space age but who knows what Sony is up to on this lens.
If there is no adjustment compensating for different cameras I would expect that everyone's mileage would be subject to some slight variation . This is a potential problem on all the Canon and Sony amount zooms including the Fuji by the way.
 
Thanks Lenny for the explanations,
Ok, i wll go ahead on f4.
1. Sharpness.....on which cam distance to object shall focus be nailed ? ..... vocal lengh 18, 28, 35 ??
2. CA......can you explain such a right CA test more detailed becouse of my less experiance.

Klaus

Update:
to go ahead a friend offered me to use his
Zeiss CP2 21, 28 50, 85 mm + Speedbooster or Comlite
Sigma 50-100 f1.8
Canon 24-70 + Speedbooster or Comlite

which combination would be most interested ?
Let me know.
 
Last edited:
Klaus,
Wow, i don't know where to begin. For the test I think its easiest to nail focus on infinity so I would just pick a shot with lots of fine detail like trees and greenery, buildings, brick etc but far enough away so that you aren't as likely to get confused by near stuff being out of focus. The background of your shot in the first video would be fine. That tree with all the barren branches might be Ok but it would be easy to end up focused on slightly different sets of branches that would confuse a test. Also with those particular branches, CA could confuse a test of focus. Make sure you nail exposure also because overexposure always looks soft.

Re what lenses to compare. Well the more the merrier I guess . I like that you chose the 18-35 Sigma because I know that for a consumer zoom lens that's as sharp as they get. It That's enough for me but it would be interesting to compare to the Canon 24-70 as well because that's a typical lens that people currently use for run & gun on FS7/5. Its a pretty good lens also so if the 18-110 came out weaker that would be a red flag & vice versa. The primes especially at f4 should handily beat out the Sony but its all up to how much you want to bother with on tests.

Re the focal length. Again - how much time and energy do you have? I would imagine that the way you did it before at 18 and 35 would tell you how good the lens is generally. You could go nuts and compare longer focal lengths as well but it gets to be a lot of work. Same re speed booster.

There's basically 2 kinds of CA and it would be valuable to look it up on the web because I won't explain it well. But for a quick version there's Axial also called longitudinal CA which creates green and magenta edges around high contrast edges just in front of and behind the focus point all through the frame. Its most obvious when the f stop is very open ( f 1.4) and resolves as it gets more closed. Those barren tree branches can show it quite a bit especially with f1.4 primes and even more if the branches are dark. Its hard to get rid of and often you need to go to expensive cine glass to really get CA free. Also unfortunately the sharper the lens the more prone it is to axial CA. Usually you don't notice it but sometimes it can jump and out and burn you. I've often seen it as a green or magenta ring around a lamp shades on interiors in lowlight, also on the edges of window frames, or it can be very bad on barren tree branches against a bright blue or cloudy sky. The worst I've ever seen was on the Zeiss Distagon 35mm f1.4 DSLR primes. That turned a piece of bubble wrap on my table into a mass of vibrating green and magenta. Very sharp lens though.

Transverse or lateral CA shows as red and blue edges on the left and right sides of the frame also on high contrast edges. and when you are on dead focus. Its usually pretty well corrected these days so I rarely see it, but I have on Canon 24-105 lenses especially using speed booster. I got rid of that lens for that reason. I wouldn't expect to see it but if you do you might be worse at the long end of the lens and the tree branches would be a good test. It might disqualify the lens for me.

Hope that's all clear - again wikipedia has it all explained very well and there are plenty of other sites on the web also. There are other kinds of CA but those 2 are the most common and important for us.

Hope that helps. I can't thank you enough for doing these tests and posting them so nicely. I know its very time consuming. I reached my limit just writing this.

Lenny
 
Last edited:
One other test that might be useful is to check how parfocal the lens is at minimum focus distance. Think I read Sony's 28-135/4 wasn't parfocal at minimum focus distance, or close to it, but never tested.
 
I think you should have made the CA test with something high contrast without colour to really show CA. The red branches can hide CA so I would choose something more colour less.
 
Nice Klaus- Thanks again
Whiie there is something to what Ossland says I just don't see any CA is your tests at least. The color in the branches might hide Magenta but not green. Of course a chart could show CA off very clearly but we don't ant to use that word in polite company.
 
I stand corrected. Took a closer look at your first test video Klaus and at first glance it looked like CA but I guess what I saw was more the red colour of the branches. But if you shoot another CA test please choose something more mono colour and at high contrast and then we can at least write off CA being a problem.

I have to admit it looks pretty sweet for a video lens!
 
You must log in or register to reply here.
Back
Top