EB-1 wrote:
I have no idea how you guys do acceptance testing, but the first part is to test a flat target under controlled conditions. That will show the general MTF, distortion, light fallof, and you can infer field curvature by changing focus. There are other common tests for the rest of Seidels and CAs.
Photographing a random 3D subject is typically a field test. That is important to demonstrate the borkah and AF. But if you only have a few random images, it's easy to overlook critical image defects like MTF degradation, astigmatism, LaCA and field curvature that produce typically worse IQ with increasing radius. Sometimes I'm so sick of mediocre lenses I could scream.
I still have some good hopes for the 100-400/4.5mm, but need to see some test results, not a Googles U-Tube.
Boy, are you funny or just OCD? The vast majority of 'us guys' get feedback before buying a lens from friends, other photographers, and photo reviewers that we trust.
When we get the lens we take it out and use it in the real world, examine the resulting images, and determine if it meets our requirements. I would bet that 99% of photographers buying lenses do NOT do any flat target testing under controlled conditions. If you use your lenses in the field, you test them in the field.
I'm trying to find some useful data, but have not yet. Maybe Lenstip or TDP will get one. The costs in time and travel to locations is very high for me, so I'm not about to squander thousands on uncontrolled tests of those kind of zooms. I imagine those with free time and living near the subjects have far more opportunities. I really need to know how it does at r=15 and r=18mm.
Jemini wrote:
Thanks for the initial impressions Geoff. What do you think about the AF with 1.4x TC compared to 200-600? These days most lenses have enough IQ. I'm not much worried about it. I'm more curious about performance of the the quadruple AF motors.
TBD....I will test AF more today on swallows IF. Yesterday I shot them with 200-600 and 300/2x before the 100-400 arrived. I will return to that same spot and see how it goes with the new lens.
Imagemaster wrote:
Boy, are you funny or just OCD? The vast majority of 'us guys' get feedback before buying a lens from friends, other photographers, and photo reviewers that we trust.
When we get the lens we take it out and use it in the real world, examine the resulting images, and determine if it meets our requirements. I would bet that 99% of photographers buying lenses do NOT do any flat target testing under controlled conditions. If you use your lenses in the field, you test them in the field.
I am sure you are right that many do that but I wouldn't be surprised if that just as many do so on advice from Youtubers you wouldn't trust. The only thing that matters is, as you say, whether the images you take are good enough for you. I am sure again that it wouldn't make any difference to most of us whether or not our lenses were decentered etc as long as we liked. what we got. For someone as experienced and as competent as you, you can test it in the field and know how good it is. But EB-1's approach is also valid and one which the more scientific enjoy. For me, I'll just take Geoff's advice and then do testing on my own copy.
I just took my standard test shots (sheet music) from across the room and compared to the 200-600, I am pleased with the sharpness and contrast.
I am a nerd and appreciate the time people take to do all of the scientific measurements, but good enough is good enough for me. This lens was an interesting mix of compromises and the kicker was that it is way less than what I was mentally prepared for, for an f4 lens, so I just bought it. Looking forward to my first soccer match with the new lens (unfortunately not until August).
Anyone interested in ultimate photo quality is carrying 2 bodies with a 400 2.8 or switching systems, no? I always find it weird that people are so demanding of a lens that no one is forcing anyone to buy.
Steve Perry and Dustin Abbott have already flat field testing, so we know what's that's like. Time for 3D outdoor subjects at a range of distances to test IQ and how bokeh evolves with subject-to-background distance, AF speed with TC's, performance near mfd, OS especially for video, where the 200-600 is useless IMO.
FWIW I received mine today and, being a pixel peeper, did a number of brick wall tests (at enough of a distance and with enough adjustments and repetition to make an initial conclusion) as well as shooting up to an angle to a slanted target (a roof) to see what was happening on the Z-axis. I used the A7R VI on a tripod and remote shutter, electronic shutter, and when there was wind increased my shutter speed to around 1/500s as needed. I found a few things:
-When shooting the brick wall, the lens was sharp wide open across the frame at 200-400mm. Stopping down to f/6.3 made a very small improvement in the center and midframe, but the corners were actually better in some cases at the maximum aperture. Basically, it was fully sharp wide open.
-At 100mm, the corners were softer but also on my copy the left edge was soft and didn't become convincingly sharp until f/10. The right side was better, but overall I could see that peak resolution was better from 200-400mm across the frame. If Sony decided to optimize for the long end then that would seem to be a good design choice, at least for my uses.
-I shot the slanted target, the roof, and found that sharp detail ran along the red line as shown below (this example at 200mm):
This tilt in the plane of focus seems to be present across 100-400mm, but it is more pronounced at 100mm and eases at 400mm, but it is still there to some degree. I think for those shooting sports or birds or those who shoot where having a flat plane is not critical, you will be happy, because even the flat target looked good. I am not sure yet because I want to also shoot landscape and I haven't determined if this is acceptable for my uses. I have other Sony telephoto zooms including the 50-150 and 70-200 II and they don't have this behavior, so I am wondering if another copy might do better.
Interesting observation about the 100/400 f4: it renders like the 300 f2.8. Just like older Minolta lenses, Sony seems to be developing a style. I love it.
Interesting observation about the 100/400 f4: it renders like the 300 f2.8. Just like older Minolta lenses, Sony seems to be developing a style. I love it.
ketang wrote:
FWIW I received mine today and, being a pixel peeper, did a number of brick wall tests (at enough of a distance and with enough adjustments and repetition to make an initial conclusion) as well as shooting up to an angle to a slanted target (a roof) to see what was happening on the Z-axis. I used the A7R VI on a tripod and remote shutter, electronic shutter, and when there was wind increased my shutter speed to around 1/500s as needed. I found a few things:
-When shooting the brick wall, the lens was sharp wide open across the frame at 200-400mm. Stopping down to f/6.3 made a very small improvement in the center and midframe, but the corners were actually better in some cases at the maximum aperture. Basically, it was fully sharp wide open.
-At 100mm, the corners were softer but also on my copy the left edge was soft and didn't become convincingly sharp until f/10. The right side was better, but overall I could see that peak resolution was better from 200-400mm across the frame. If Sony decided to optimize for the long end then that would seem to be a good design choice, at least for my uses.
-I shot the slanted target, the roof, and found that sharp detail ran along the red line as shown below (this example at 200mm):
This tilt in the plane of focus seems to be present across 100-400mm, but it is more pronounced at 100mm and eases at 400mm, but it is still there to some degree. I think for those shooting sports or birds or those who shoot where having a flat plane is not critical, you will be happy, because even the flat target looked good. I am not sure yet because I want to also shoot landscape and I haven't determined if this is acceptable for my uses. I have other Sony telephoto zooms including the 50-150 and 70-200 II and they don't have this behavior, so I am wondering if another copy might do better....Show more →
This might be a silly question, but did you lock out the OSS for the test? I have run into similar problems on Canon, where the IS was tilting the sensor...
But, but, but, those shots are meaningless unless you guys do acceptance testing, but the first part is to test a flat target under controlled conditions. That will show the general MTF, distortion, light fallof, and you can infer field curvature by changing focus.
Jemini wrote:
Thanks for sharing these Geoff. Looks like AF works great even with 2X. Would you say it's on par with 300/2.8 with 2X for AF?
If that's the case I can have the 1.4X glued to the lens all the time. Your thoughts on comparison to 200-600?
I feel the 300GM/2xTC is better AF hit rate than the 100-400/2xTC. Just that extra 1.33 stops of light will give it the advantage.
I think the 100-400/1.4TC is better AF than the 300GM/2xTC. But I think the IQ is fairly close.
For just IQ I think the 100-400/1.4TC is about equal to the 200-600 for perched birds. But as soon as you start shooting action I find the 100-400/1.4TC to be way, way better. I was really impressed with the hit rate on the Purple Martins with the 1.4TC. The handling of the 100-400/1.4TC is also way nicer than the 200-600. Only thing I'd give the nod to the 200-600 would be the smoothness of the zoom ring.
The bare lens is really, really good also. I'm not sure how it would compare to the 300GM/1.4TC but I think they are fairly close for IQ and AF speed. I think bare 300GM is still the sharpest and fastest AF I've ever used.
arbitrage wrote:
I feel the 300GM/2xTC is better AF hit rate than the 100-400/2xTC. Just that extra 1.33 stops of light will give it the advantage.
I think the 100-400/1.4TC is better AF than the 300GM/2xTC. But I think the IQ is fairly close.
For just IQ I think the 100-400/1.4TC is about equal to the 200-600 for perched birds. But as soon as you start shooting action I find the 100-400/1.4TC to be way, way better. I was really impressed with the hit rate on the Purple Martins with the 1.4TC. The handling of the 100-400/1.4TC is also way nicer than the 200-600. Only thing I'd give the nod to the 200-600 would be the smoothness of the zoom ring.
The bare lens is really, really good also. I'm not sure how it would compare to the 300GM/1.4TC but I think they are fairly close for IQ and AF speed. I think bare 300GM is still the sharpest and fastest AF I've ever used....Show more →
Can you post links to full resolution uncropped files or the RAW files. So far I am not blown away but it's hard to judge anything without seeing the full resolution originals.
A couple of reference images - taken with the old smaller, cheaper, lighter 100-400 - and the A1. Be sure to zoom right in on the eyes - you might have to save the images as a local copy first because FM doesn't seem to allow zooming in on the images.
I doubt any lens can improve much on these but the ergonomics on the new 100-400 with the new zoom ring might make it easier to use.
