j4nu wrote:
I wouldn't get too excited about AF. In my experiencee VXD motors are as fast as the latest Sony XD linear motors when it comes to "normal" usage, but fall behind when tracking (action, sports, erratic movement, esp. towards and close to the camera).
In my (non-pro) experience with the 70-180 on the A7IV, the VXD motors are excellent. I've been shooting swim events over the summer with that combo and the AF tracking can very quickly lock onto a swimmer's eye for the split second that they come up for a breath. This gives me a lot of confidence that the 50-400mm will be up to the task for most shooters and situations (backed up by most of the early reviews), even if it is not quite up to the level of the latest Sony lenses.
Well, I'm not saying it's bad - I've shot my kids sledging down a hill with my 35-150 and a lot of shots were properly focused. I'm talking about the difference you would see if you shot with your tamron against 70-200GM II some runners coming towards you, especially close. It's the most demanding scenario I think, but the difference between Sony native lenses and all the rest is there in that case.
In the JPG I can see it well. When I loaded RAW via DNG conversion to LR 5.7, at default settings it was almost invisible. It become visible when I increased exposure or did things like +100 shadows. So it will depend on processing. I sometimes do things like that. I sometimes shoot into Sun where high contrast is common (example).
Enabling chromatic aberration correction did not help at all though color of PF shifted a bit to pink.
Enabling Defringe and picking PF worked really well. There was a weak greenish-yellowish tint remaining on contrasting edges going other direction which was also correctable by picking it. Is it a PF at all? It behaves more like LaCA in that it gets stronger outside of center and there is greenish tint on edge going in other direction. I wonder if UV filter could reduce it as I have read on Wikipedia that PF is sometimes caused by UV image with different magnification.
My experience with LR's Defringe tool is often bad as it removes color where it should be kept (e.g. lips, leaves). It did work well here at first sight, but when I looked closer, I could see some areas where Defringe damaged colors (foliage above person and background behind rails). Some of it could be improved by reducing amount of correction.
Is newer LR (classic) any better here (compared to non-subscription LR 5.7 I use)?
I wonder if I should get the Tamron 50-400 to replace Sigma 100-400 Canon-mount on MC-11 which I use on A7Riv for landscapes (few landscapes at 400mm: 123) and Moon-with-landscapes (see my guide).
For wider stuff I carry most often 20/1.8 28/2 55/1.8 and 85/1.8 together with that 100-400.
I do not use 24-105/4 much as mine is not great at some FLs (measured at f/8; though two brand new 24-105/4 I tested head-to-head were worse overall - am I too picky? I feel like that since getting A7Riv).
I do not use 16-35/4 ZA much as it is quite weak (corners everywhere and whole image at 35mm) while the 20/1.8 is excellent - though I may replace it with the newer 16-35/4 G PZ.
I shoot landscapes on tripod almost all the time. With 50-400 I could leave the 85 at home, maybe 55 too for scenes where PF is not an issue which would be nice weight saving. I would also get better AF (Sigma on MC-11 has sometimes problems focusing accurately at 400mm).
Rest assured, the new 16-35 PZ is much better than the ZA. Don't believe some reviewers who say that they are "both very good, and if you have the ZA you needn't to upgrade." It is telling that Ebay is shock full of ZAs...
In the JPG I can see it well. When I loaded RAW via DNG conversion to LR 5.7, at default settings it was almost invisible. It become visible when I increased exposure or did things like +100 shadows. So it will depend on processing. I sometimes do things like that. I sometimes shoot into Sun where high contrast is common (example).
Enabling chromatic aberration correction did not help at all though color of PF shifted a bit to pink.
Enabling Defringe and picking PF worked really well. There was a weak greenish-yellowish tint remaining on contrasting edges going other direction which was also correctable by picking it. Is it a PF at all? It behaves more like LaCA in that it gets stronger outside of center and there is greenish tint on edge going in other direction. I wonder if UV filter could reduce it as I have read on Wikipedia that PF is sometimes caused by UV image with different magnification.
My experience with LR's Defringe tool is often bad as it removes color where it should be kept (e.g. lips, leaves). It did work well here at first sight, but when I looked closer, I could see some areas where Defringe damaged colors (foliage above person and background behind rails). Some of it could be improved by reducing amount of correction.
Is newer LR (classic) any better here (compared to non-subscription LR 5.7 I use)?
I wonder if I should get the Tamron 50-400 to replace Sigma 100-400 Canon-mount on MC-11 which I use on A7Riv for landscapes (few landscapes at 400mm: 123) and Moon-with-landscapes (see my guide).
For wider stuff I carry most often 20/1.8 28/2 55/1.8 and 85/1.8 together with that 100-400.
I do not use 24-105/4 much as mine is not great at some FLs (measured at f/8; though two brand new 24-105/4 I tested head-to-head were worse overall - am I too picky? I feel like that since getting A7Riv).
I do not use 16-35/4 ZA much as it is quite weak (corners everywhere and whole image at 35mm) while the 20/1.8 is excellent - though I may replace it with the newer 16-35/4 G PZ.
I shoot landscapes on tripod almost all the time. With 50-400 I could leave the 85 at home, maybe 55 too for scenes where PF is not an issue which would be nice weight saving. I would also get better AF (Sigma on MC-11 has sometimes problems focusing accurately at 400mm). ...Show more →
Yeah, it looks like LaCA to me too. I wonder if a proper lens profile (Adobe, DxO, ...) can reduce/eliminate it completely...
hasenbein wrote:
Rest assured, the new 16-35 PZ is much better than the ZA. Don't believe some reviewers who say that they are "both very good, and if you have the ZA you needn't to upgrade." It is telling that Ebay is shock full of ZAs...
I have compared my copy of ZA with a display copy of PZ in a shop last week on A7Riv, on my tripod, outdoor view, at f/8, in series of shots with 16, 20, 24 and 35mm. I was surprised that ZA was actually not all that bad. There was a difference in image sides in all focal lengths, but not a strong one. In mid and central area, the there was no meaningful difference. At 35mm the difference was big and in the whole image. If I would not have 20/1.8, I would buy PZ instantly, but 20/1.8 is so good that I do not feel pressure to do so (I shoot rarely wider than 20mm) and I will wait for bigger discounts.
Anyway, I decided to buy 50-400. It is available in local shops now. I will be able to compare it with my Sigma 100-400 Canon-mount and maybe friend's Sony 100-400.
jtra wrote:
So I have it. I will get tripod collar tomorrow. Over weekend I hope I will be able to compare the three lenses I mentioned for landscape use.
I have briefly tried it and I can confirm that mine was also super hard to mount.
Any questions?
Side-by-side against the Sony GM (especially at 400mm) would be appreciated.
Sharpness as well as bokeh comparison if possible
Also, it would be good to know whether the purple fringing issue at 50mm is really prominent or not an issue at all.
jtra wrote:
So I have it. I will get tripod collar tomorrow. Over weekend I hope I will be able to compare the three lenses I mentioned for landscape use.
I have briefly tried it and I can confirm that mine was also super hard to mount.
Any questions?
Eagerly awaiting to hear your thoughts .
I also managed to get a couple shots off and actually my results in some of the focal range are quite different from those reported in all the reviews, so maybe I tried a dud.
1st is new Tamron 50-400/4.5-6.3 Di III VC VXD
2nd is Sigma 100-400/5-6.3 DG HSM Contemporary Canon-mount on MC-11
3rd is Sony 100-400/4.5-5.6 GM
All three are standing on tripod collar foots. Sigma uses this one: https://stativschellen.de/for-sigma-100-400 (I can recommend it. It works well for me. It might be most stable of all three as the foot is supported by two collars).
I did a bit of indoor testing with test target displayed on computer monitor (other lights off). This works well at wide focal lengths. At longer focal lengths with the room I had it was too magnified showing subpixels so it was not ideal there.
At wide end the Tamron has stronger chromatic aberration in corners than my other lenses (compared at f/8 with 55/1.8, 85/1.8, 24-105/4 at 85mm). There is no/low chromatic aberration in center. So it is more like LaCA than purple fringing. Well, it may depend on dynamic range, DR on monitor is big, but not as big as you may find outdoors. I have tested bottom right and top left corners. The bottom right had more CA than top left while sharpness was about same which is interesting.
I compared the Tamron with the other two 100-400 at 100, 200 and 400mm. The CA was slightly stronger on Tamron. Sharpness was similar (though it was harder to evaluate it at longer focal lengths since monitor pixels were too magnified), but at 400mm there was significant field curvature (center out of focus when focusing on corners and corners were out of focus when focusing on center; focus distance about 3.3m). The other two 100-400 had flat field there. So I wonder how will that turn out at bigger focus distances.
Focus on Tamron (AF-S with light conditions for proper exposure f/8, 1/3s, ISO 100) felt slower than on Sony 100-400, but it was accurate too. Focus on Sigma 100-400 on MC-11 was not working at all in these conditions (it did some movements then it gave up so I focused it manually).
I plan to process my test shots into a review.
I will need to do outdoor test to decide on the lens. I did not buy it to shoot test charts, but it did not fare great there.
At this time includes 55 full resolution outdoor pictures at f/8 using four lenses (Tamron 50-400 obviously, Sony 100-400 GM, Sigma 100-400 older Canon-mount non-DN, Sony 24-105/4)
After doing some test chart tests (as I wrote above), I felt it was not great performer, however on actual landscapes it has worked reasonably well. Usually it delivered sharper sides than other two long zooms while being a bit weaker in central sharpness. On the wide end it was a match or slightly better than 24-105. Scenes were mostly low contrast and CA/PF was not a problem on them.
This is interesting as the copies I checked, actually behave differently (wide open though). At 200 and 300 mm, my GM is a bit sharper in the center, while at 400mm the Tamron get a small advantage (but looks like the GM is better in the rest of the frame).
Unfortunately, we had some cold air arrive here which dropped the temps to around ~15 °C and somehow this made my attempts at longish distances futile (heat distoriton in most shots), so I was only able to test reliably on close distance targets.
One thing to note, not related specifically to IQ, is that OIS/VC/VR effect on viewfinder stabilization is significantly worse on the Tamron. In pictures, I think the Sony might still a bit better but the difference is nowhere near as close as when lookiing through EVF.
I also get the impression that at close to MFD on the long end (300-400), the Tamron looses quite a bit of sharpness, as usually zooms do (but the GM excels here).
This is interesting as the copies I checked, actually behave differently (wide open though). At 200 and 300 mm, my GM is a bit sharper in the center, while at 400mm the Tamron get a small advantage (but looks like the GM is better in the rest of the frame).
Unfortunately, we had some cold air arrive here which dropped the temps to around ~15 °C and somehow this made my attempts at longish distances futile (heat distoriton in most shots), so I was only able to test reliably on close distance targets.
One thing to note, not related specifically to IQ, is that OIS/VC/VR effect on viewfinder stabilization is significantly worse on the Tamron. In pictures, I think the Sony might still a bit better but the difference is nowhere near as close as when lookiing through EVF.
I also get the impression that at close to MFD on the long end (300-400), the Tamron looses quite a bit of sharpness, as usually zooms do (but the GM excels here).
There is quite a difference in colors / white balance (Tamron is yellow'ish, reminds me of the first version of their 28-75mm). Is this due to changing conditions or the lens? The colors in the photo from the Sony appear to be more accurate...
keepcoding wrote:
There is quite a difference in colors / white balance (Tamron is yellow'ish, reminds me of the first version of their 28-75mm). Is this due to changing conditions or the lens? The colors in the photo from the Sony appear to be more accurate...
Yes, the difference in colors SooC is pretty big. I guess Sony is a bit closer to what my eyes see ...
The conditions in the 3rd photo were roughly the same.
I have always suspected it’s the coatings that Tamron uses being yellow… this has been the case even 15-20 years ago with Tamron’s very first 28-75 on EF mount
Regardless the Tamron yellow is a real thing and getting it out of skin tones is annoying to say the least
Maybe blue light is being suppressed somehow
j4nu wrote:
Yes, the difference in colors SooC is pretty big. I guess Sony is a bit closer to what my eyes see ...
The conditions in the 3rd photo were roughly the same.
At this time includes 55 full resolution outdoor pictures at f/8 using four lenses (Tamron 50-400 obviously, Sony 100-400 GM, Sigma 100-400 older Canon-mount non-DN, Sony 24-105/4)
After doing some test chart tests (as I wrote above), I felt it was not great performer, however on actual landscapes it has worked reasonably well. Usually it delivered sharper sides than other two long zooms while being a bit weaker in central sharpness. On the wide end it was a match or slightly better than 24-105. Scenes were mostly low contrast and CA/PF was not a problem on them....Show more →
Theoretically diffraction starts to become visible around f/5.7!
As such I think some shots with the 50-400 wide open (f/6.3) are needed, and compared against the other lenses both at f/6.3 for an apples to apples test and at f/5.6, because the f/5.6 shot is theoretically sharper than the one at f/6.3 on a lens with zero coma and astigmatism!
.