I decided to take test shots with 3 of my lenses as I was concerned the TSE 24 II wasnt performing as it did when I had a Canon body. The lenses tested were the TSE 24 II, Samyang 24 and EF 16-35 @ 24mm. All shots were taken from a tripod with IBIS off, uncompressed RAW, ISO 100 aperture priority. The focus point was the building in the centre of the image.
When I view the shots in lightroom and view at 100% all look ok with the Samyang looking the sharpest followed by the 16-35 followed by the TSE 24. However when I zoom to 200% the samyang and the EF 16-35 hold up ok but the TSE lens looks all pixelated and mushy.
Now I know you will be saying you dont view photos at 200% anyway but thats not the point here. I want to understand why my TSE looks like crap at this magnification and the other two dont?
Could it be that the camera is out resolving the TSE lens? (I always thought this out resolving chat was a load of rubbish but its the only thing I can think off)
Any ideas?
Images shot at f/8 from the 3 lenses can be found HERE
Uhm this is what I explained a while ago in another thread. But I got good results without movements for most of the image area. 90mm is sharp edge to edge.
No worries because it wasn't my thread anyways. I can post few examples to this thread for comparison.
BTW, zero tilt might not mean zero tilt. This 42mp sensor 5dsr are very sensitive to slight tilt movements. I found that middle position was not exactly zero in any of my lenses.
Did you get to a conclusion as to what was going on? At the moment it seems to me the 24 does not perform well on the A7RII (compared to the other two lenses I mentioned)
This is taken from unprocessed shot which had no movements. ISO100, f/8 no filters. Focused to upper branches using liveview. Remember, this is A7RII + MB4-T and 24mm TS-E MK2. Total IQ failure here...
Here is another example with flat LR conversions, this time with ~8mm shift upwards. First the whole scene from my balcony. Focused to upper part of the trees.
Nope. Totally pristine lenses. As I explained before, 24mm TS-E mk2 works very well with Canon 5Ds(r). Something does not add up in this combination. 24mm variant is the most affected, 17mm works up to 6-7mm shift and minor tilt. 90mm is almost flawless with full movements and stopped down.
I'm thinking these lenses are super sensitive to sensor glass thickness as is the case with A7x and Leica-M glass.
My 17mm tse works very well in the extreme shifted coners. I have two metabones 3 adapters, I filed, then sandpapered and painted one. That seems to get me the best performance in the extreme corners as opposed to flocking. I can't shift to the extreme corners with flocked adapter.
My 17mm TSE actually is weaker at 8mm shift and then gets better at 12mm shift. I'm not sure if it's some sort of field curvature.
I actually have two 24mm tse's as I got one to replace the one I dropped. Extreme corner shifted sharpness has been poor on both of them. I've found adding 0.5 tilt (half way to one line) in the direction of the shift sharpens up the corners. So this is my work around, as I mostly use the lens as a shift lens for panoramas. I focus stack when I shift, but if I tilt, I don't usually do a panorama.
But yes, when I shot canon 5d3 the 24mm tse was sharper in the corners. The 24mm TSE seems to be the biggest problem shift lens for me with Sony. 17mm TSE, is wonderful on Sony! Pentax 645 lenses are also great with more shift than with the Canon TSE's. That new Pentax 645 HD 35mm is something to dream about... One day!
A more definitive answer would be if the same shots where then also made with the TSE attached the your 5Ds(r) for comparison. Regardless, something does seems a bit off in the results of the a7rII+mb4+tse.
Hell yes! This is a good thread. Nice to hear that others have experienced the same issue as well. Personally very important to me as I originally purchased A7RII to be used with Canon T/S lenses. Cannot afford MFD technical camera systems.
Vertical shifts are the equivalent to 6mm horizontal shifts, when I've said I've had problems with my A7r2 +metabones 3 (flocked) + 24mm TSE II I've not had any problems with 6mm shifts. My vertical panos are tack sharp corner to corner (and nothing like the photos posted here). 12mm horizontal shifts can be improved by adding a tiny amount tilt in the same direction, from my testing.
That is weird...the same 24mm lens which was problematic with a7r2 workwd okay with 5dsr. And the same a7r2 gives balanced results with 90mm and 17mm. So what takes? Is it really the flocked adapter? You said you painted and filed yours. What paint did you use?
I'm really pissed off if the default flocking in mb4-t is causing all this.
My issues are with unshifted frames. Im starting to think it is indeed my lens but I have also checked 5D3 shots taken with the same lens and they are fine (before I bought the A7RII) which is odd
For those who have tested the 24 on the a7rII, did the mode of the MB make any difference (Green versus Advanced)? Also, was SteadyShot tested both on and off?
One more thing... could shooting the lens at f/4 or 5.6 improve the results? Maybe diffraction is making things worse than they necessarily have to be? I know with Rodenstock's UWA lenses (the latest 28, 32, and 35) they recommend shooting wide open and using camera movements to maximize DoF because diffraction sets in so early on high resolution MF backs. Could things be improved if we took the same approach with the TS-E 24? I assume there would need to be more work done on vignetting in post.
The flocking material appears to be the sames as the Edmunds flocking. I would be surprised if the T version caused an issue. Wouldn't the flocking affect all lenses and we'd have heard by now?
MJKoski wrote:
That is weird...the same 24mm lens which was problematic with a7r2 workwd okay with 5dsr. And the same a7r2 gives balanced results with 90mm and 17mm. So what takes? Is it really the flocked adapter? You said you painted and filed yours. What paint did you use?
I'm really pissed off if the default flocking in mb4-t is causing all this.
