Home · Register · Software · Software · Join Upload & Sell

Moderated by: Fred Miranda
Username  

  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Sony Forum | Join Upload & Sell

1       2      
3
       4       5       end
  

DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom
  
 
DavidBM
Offline
• • • •
Upload & Sell: On
p.3 #1 · p.3 #1 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


Fred Miranda wrote:
By the way they messed up with the 24-70GM review, I've lost complete trust in their assessments. I just post it because it was Friday and we need something to talk about.

From Roger's tests, the Canon is noticeably superior to the Sony at center throughout the entire range and that's the test I trust.
I am also absolutely convinced that Roger's optical bench tests are non-bias and independent.


What was messed up with their 24-70GM review, Fred? I donít recall any special issue ....



Oct 14, 2017 at 01:12 AM
philip_pj
Online
• • • •
Upload & Sell: On
p.3 #2 · p.3 #2 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


chez, the general rule is that it costs five times as much to gain a new customer as compared with keeping an existing customer. It seems to be a lesson that is hard to learn, and may point to a lack of feedback from the retail/distribution stage to the maker.

Lens tests - it's a bizarre situation in such an advanced technological industry. It's one reason to buy Zeiss and Leica. Their MTF reports the facts, nothing but the facts, be it good news or bad. MTF has been with us since CZ guys transcribed results by hand for Contarex lenses, sixty years ago:

"The graphs to be published here are unique as they are absolutely original drawings from the Zeiss optical department. It is clear that the use of the MTF graphs is not a recent phenomenon. The lenses were designed around 1955 to 1965 and the graphs are calculated and drawn on graphing paper, again an indication that the method was well-known, but the computer could not handle the output."

http://www.imx.nl/photo/zeiss/zeiss/page67.html

That's progress for you.



Oct 14, 2017 at 01:56 AM
Fred Miranda
Offline
Admin
Upload & Sell: On
p.3 #3 · p.3 #3 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


DavidBM wrote:
What was messed up with their 24-70GM review, Fred? I donít recall any special issue ....


Ok, maybe they messed up for me only (), but I know the FE 24-70GM pretty well and its corners are pretty much amazing from 24 until about 35mm. That's not what they showed on their colorful, fancy graphs.

BTW: The 24-70GM corners at the 24-35mm range beats the 16-35GM but that's not what their test shows either...



Oct 14, 2017 at 02:06 AM
Messier77
Offline
• •
Upload & Sell: On
p.3 #4 · p.3 #4 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


Fred Miranda wrote:
Ok, maybe they messed up for me only (), but I know the FE 24-70GM pretty well and its corners are pretty much amazing from 24 until about 35mm. That's not what they showed on their colorful, fancy graphs.

BTW: The 24-70GM corners at the range beat the 16-35GM but that's not what their test shows either...


Perhaps the infamous Sony copy variation struck again and they tested a weak copy of the 24-70GM?



Oct 14, 2017 at 02:17 AM
DavidBM
Offline
• • • •
Upload & Sell: On
p.3 #5 · p.3 #5 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


Fred Miranda wrote:
Ok, maybe they messed up for me only (), but I know the FE 24-70GM pretty well and its corners are pretty much amazing from 24 until about 35mm. That's not what they showed on their colorful, fancy graphs.

BTW: The 24-70GM corners at the range beat the 16-35GM but that's not what their test shows either...


Interesting. When I said 'I don't recall any issue', that wasn't code for 'I don't think there was any issue' - I just didn't remember anything about it!

Two things spring to mind; one might be that it's an infinity versus closer focus thing, the other is that it's still a closer focus thing, but it's only that there's just a little bit of closer focus CoF that doesn't matter except for planar subjects like test charts... (My distagon 2/28 is like that; close enough to flat field to be fine across the field stopped down at infinity, but significant CoF at chart distance that doesn't go away, and rarely matters if you know about it...)

Of course the third option is that they stuffed up...



Oct 14, 2017 at 02:44 AM
Schlotkins
Offline
• • • •
Upload & Sell: On
p.3 #6 · p.3 #6 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


Fred Miranda wrote:
Ok, maybe they messed up for me only (), but I know the FE 24-70GM pretty well and its corners are pretty much amazing from 24 until about 35mm. That's not what they showed on their colorful, fancy graphs.

BTW: The 24-70GM corners at the range beat the 16-35GM but that's not what their test shows either...


Did you finally find another good copy of the GM 24-70 Fred?

Chris



Oct 14, 2017 at 04:11 AM
Fred Miranda
Offline
Admin
Upload & Sell: On
p.3 #7 · p.3 #7 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


Schlotkins wrote:
Did you finally find another good copy of the GM 24-70 Fred?

Chris


Chris,
My first copy was exceptional. (Just like the 16-35/4 ZA I sold you) but unfortunately I dropped it and it got severely damaged. I tried renting a couple more but didn't get lucky finding a golden copy. I'm thinking about trying again to complement my 12-24G.



Oct 14, 2017 at 04:23 AM
freaklikeme
Offline
• • • • •
Upload & Sell: On
p.3 #8 · p.3 #8 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


Don't ultra wides tend to get better at the long edges if you use the hyper-focal distance rather than "hard stop" infinity?


Oct 14, 2017 at 06:14 AM
Holger
Offline
• • •
Upload & Sell: Off
p.3 #9 · p.3 #9 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


mttran wrote:
Believe whatever you want but let the number of oof periodic threads from each forum conceal which milc/dslr system is a better platform base for engineering model to grow. There are good number of reasons why Sony engineers broke those mirror(s) - why don't you start from there


There are a good number of reasons. Sure. But that is not what we talk about. It is your exclamation that it is a literally "lottery" to get an in-focus shot. For people knowing what they are doing it certainly is not.


Edited on Oct 14, 2017 at 02:42 PM · View previous versions



Oct 14, 2017 at 07:41 AM
thenoilif
Offline

Upload & Sell: Off
p.3 #10 · p.3 #10 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


So, let me get this straight....a $2,000 + lens can be purchased with a noticeable defect in image quality and people are ok with that?



Oct 14, 2017 at 10:46 AM
 

Search in Used Dept. 



telyt
Offline
• • • •
Upload & Sell: Off
p.3 #11 · p.3 #11 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


thenoilif wrote:
So, let me get this straight....a $2,000 + lens can be purchased with a noticeable defect in image quality and people are ok with that?


Even much more costly lenses will have noticeable image quality 'defects' depending on what you're looking for. Some people will notice the OOF color fringes of spherochromatic aberration, others won't notice for example. The way to use a $2,000+ lens, or any lens for that matter, is to be aware of its strengths and weaknesses and take advantage of its strengths and work around its weaknesses. Regardless of the price tag its up to each of us to choose which imaging properties we want (or want to avoid) and decide if a lens is 'worth it'.



Oct 14, 2017 at 11:52 AM
RCicala
Offline
• • • •
Upload & Sell: On
p.3 #12 · p.3 #12 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


Let's pretend that Sony made a lens that was as about good as the Canon, but with perhaps more sample variation. And that manufacturers hand-pick copies that are sent for review. So we would expect the two tested lenses (Canon and Sony) to be very similar.

A system MTF test (camera plus lens, which would be DxO or Imatest) is not the same as a lens test (because MTF system = MTF lens X MTF camera).

That would mean that if a good copy of the Sony lens was tested on a better camera than the similar Canon lens, that we would expect DxO to rate the Sony higher than the Canon because they are testing the system, not the lens. The camera MTF is WAY more complicated than just pixel count. And if .jpgs are tested rather than raw, or the raw is sharpened, it's so different as to be far more important than the pixel count.

So DxO says a good copy of the Sony 16-35 on a Sony camera is slightly better than a good copy of the Canon 16-35 on a Canon camera. I'm surprised that here on a Sony board, people are surprised. . . .

Now I'll turn on cynical snarky mode and mention that I believe 9 of the last 11 products reviewed by DxO have been declared "BEST ONE EVER!!!!" This includes two consecutive cell phone cameras (the iPhone 8 was best one ever for like 4 days) and the Nikon D850 (which required removing the Pentax 645D from their database before reviewing the Nikon because the Pentax rated a point higher). And DxO's software isn't selling as well as it used to . . . . . . .



Oct 14, 2017 at 11:56 AM
GMPhotography
Offline
• • • • • •
Upload & Sell: On
p.3 #13 · p.3 #13 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


Honestly I think DXO scores do more harm than good, so I never ever paid any attention to them. As someone that tests lenses for end use. I only pay attention to Rogers work and testing from other members that I trust. Obviously I trust myself and Iíve been around lenses for years but I go by final output images so thatís one equation but I do trust Rogers tests on his very expenses toys that he uses besides that he is also not bias to brand or even money to be gained from testing. I have yet to try this lens but like any other zoom on the planet Iím sure there are some weak spots, copy variation and things of this nature. One other big problem I see are comparisons to other brands. Now maybe Iím crazy but I shoot Sony and really Nikon/Canon mean absolutely nothing to me so in my head at least there is zero reason to even compare. I shoot to make money all I care about is having a brand that I trust with a system that goes with it. Sure I love third party lenses but that will never include a Nikon/ Canon in the mix. I donít have this lens or tested it so Iíll just stay out of it with a warning be careful what you read.


Oct 14, 2017 at 01:04 PM
chiron
Offline
• • •
Upload & Sell: On
p.3 #14 · p.3 #14 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


In addition to the other sources of testing & reporting variations as described by Roger, I also wonder to what extent lens tests are shooting at a moving target.

To what extent do Sony and other manufacturers tweak their manufacturing processes in an on-going way to improve problems with copy variation that derive from the difficulties of high-precision manufacturing of very complex but still mass-produced objects?

I remember that not too long ago Roger reported that Canon lenses had been showing some excellent copy variation outcomes with a series of lenses. Did this reflect a change in how Canon was manufacturing their lenses? Presumably so. And famously (for those of us who try to follow Roger), the Sony 70-200 2.8 GM was designed to perform at a level of critical tolerances and assembly that manufacturing processes were having difficulty meeting. Will Sony be able to improve those manufacturing processes over time?

Each design must present unique as well as familiar problems for the actual manufacturing process. I wonder if Sony and others might improve these processes in an on-going way that gradually improves copy variation and lens performance because it gets each copy closer to the design spec.

Not that lens hoods are the kind of high-tech manufacturing I have in mind here, but the initial imperfections in some of the lens hoods for the 16-35 2.8 GM might be a very minor example of what I am talking about. Some of these hoods were off just a bit, presumably from manufacturing variances, and that seems to have been fixed.



Oct 14, 2017 at 01:19 PM
Schlotkins
Offline
• • • •
Upload & Sell: On
p.3 #15 · p.3 #15 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


I don't know there is a large conspiracy theory here. The way I see it:

- Roger tests at infinity. Lensrentals does not. Lenses can behave much differently at different focus distances

- Roger tests 10. Dxo does 1. So there's about a 20% chance that Dxo will either have much better or much worse results than Roger.

- The "aggregate" score stuff from Dxo isn't very helpful when thinking of a wide open MTF graph for zoom lenses.

If you look at the correlation between lenses Roger thinks are good and Dxo, I think the correlation is pretty darn high. It's when you get to the extreme ends and exact ordering that things change.

Chris



Oct 14, 2017 at 01:58 PM
mttran
Offline
• • • • •
Upload & Sell: Off
p.3 #16 · p.3 #16 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


Holger wrote:
There are a good number of reasons. Sure. But that is not what we talk about. It is your exclamation that it is a literally "lottery" to get an in-focus shot. For people knowing what they are doing it certainly is not.


Holger, here is the sequence:

1) Fred opened the thread with (canon & sony) MTF comparison
2) Molson said "buy five, get one sharp" from sony
3) I submitted....same scenario on canon since real time MTF is not that best from canon when compared to sony since dslr has to deal with its unstable mirror(s).
4) You came in and distorted my comments with "weirder" remark
5) I patiently explained dslr is an unstable system with oof facts that's well shared in dslr forums for years.
6) You came back with more distortion on my comments....please, read again and tell me where I said "in-focus is a lottery shot"

If I remembered correctly, you did the same thing last week to someone that caused the relate thread got locked....you need to understand the subject before distorted the whole system MTF thread. Please, stick with science stuffs and ghost out your fanboyslism comments.



Oct 14, 2017 at 03:21 PM
RCicala
Offline
• • • •
Upload & Sell: On
p.3 #17 · p.3 #17 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


Schlotkins wrote:
- Roger tests at infinity. Lensrentals does not. Lenses can behave much differently at different focus distances

Chris


One thing I just learned last week, which I was a bit surprised about. A very reputable manufacturer's lens designers and I were talking about testing and I brought up this point and asked if they saw much difference, they test mostly at shorter distances but do have optical benches and test at infinity sometimes.

They said with a well-designed modern lens they did not see much. None if the lens had a floating element. Older designs (decades) did show that difference.

I was surprised myself, and know of at least a couple of exceptions to the rule, but that was reassuring to hear.



Oct 14, 2017 at 03:38 PM
Gary Clennan
Offline
• • • • •
Upload & Sell: On
p.3 #18 · p.3 #18 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


I have learned so much from threads such as this! Thanks to everyone for your contributions.


Oct 14, 2017 at 03:52 PM
Fred Miranda
Offline
Admin
Upload & Sell: On
p.3 #19 · p.3 #19 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


Schlotkins wrote:
I don't know there is a large conspiracy theory here. The way I see it:

- Roger tests at infinity. Lensrentals does not. Lenses can behave much differently at different focus distances

- Roger tests 10. Dxo does 1. So there's about a 20% chance that Dxo will either have much better or much worse results than Roger.

- The "aggregate" score stuff from Dxo isn't very helpful when thinking of a wide open MTF graph for zoom lenses.

If you look at the correlation between lenses Roger thinks are good and Dxo, I think the correlation is pretty darn high. It's when you
...Show more

I agree Chris, but as Roger wrote, the Sony "lens" MTF never matched the Canon "lens" MTF. There is no way in my mind that a lower megapixel Sony "system" MTF would beat the Canon "system" by that much...I would love to see the real 1:1 crops evaluated by DXO.

Differences in distance performance should not be that dramatic especially with modern lenses. Unless DXO tests at MFD or very close to it we should not see such discrepancy. I think that infinity testing is the most reliable since the results represent most shooting distances.



Oct 14, 2017 at 03:53 PM
Holger
Offline
• • •
Upload & Sell: Off
p.3 #20 · p.3 #20 · DxOmark: FE 16-35mm f/2.8 GM, Highest-rated wide-angle zoom


mttran wrote:
Holger, here is the sequence:

1) Fred opened the thread with (canon & sony) MTF comparison
2) Molson said "buy five, get one sharp" from sony
3) I submitted....same scenario on canon since real time MTF is not that best from canon when compared to sony since dslr has to deal with its unstable mirror(s).
4) You came in and distorted my comments with "weirder" remark
5) I patiently explained dslr is an unstable system with oof facts that's well shared in dslr forums for years.
6) You came back with more distortion on my comments....please, read again and tell me where I said "in-focus is
...Show more




Here your comments:
"It is a easy to tell a bad/good lens from milc platform but not always from a dslr. This is why we see many oof threads in canikon forums "
For me a generalisation. What is the purpose of raising up the milc superiority and why do you need to discredit dslrs in a Sony thread on a Sony lens without need? Why do you think the mirror to be unstable? In most scenarios I shoot at 1/250s and faster, there is hardly any influence of mirror slab at all. The measurements are usually done in LV in MUP mode or with EFCS, so no mirror involved.

"^^^ but how we fix the unpredictable flapping mirror(s) that causes most oof in dslr regardless good or bad lenses."
Again a generalisation in my opinion. You mention "unpredictable". What do you mean here. We talk about a mechanical system tuned for a more than a decade. There can be influences due to mirror shock. They cause elastic "shock" waves of a certain measurable wavelength, possibly interfering at slower shutter speeds. Whether there is an influence is mirror mechanism dependent, Camera body dependent, focal length etc. But for each camera the frequency is very well predictable. That has been reported and measured over the years several times. On the D810, which I used for quite some time I saw it to be at 1/40s with a 50mm lens, for example. Always. So doing your science helps you avoid that SS. One can cope with it. As one can do with shutter shock, which affects mirrorless, too (A7r, Olympus, etc.) is something you heard, too?

"I have an impression that you have never had any oof from your dslr. I have a fair number of sharp one and lot of oof one from my 1/5 dseries (and all calibrated lenses) all the times, not weirds but scientific facts. I did not say that dslr can't deliver but the chances to have a best one is like picking a lottery. This is why dslr has MAF build in (but good only for fixed dof) and still required multiple shots to compensate the mirror(s) error(s). Don't believe me: try 35L,50L,85L..."
Again generalisation in my book. I even told you that I have more OOF shots with DSLRs. You talk about a "mirror error". What is that? You probably mean alignment? And here is your lottery comment which is an exaggeration, too. I don't have the feeling that to get great shot with a dslr is a lottery.

I wasn't personal in the last thread which was locked. You probably try to use that to discredit my opinion. Your quote "you need to understand the subject" is a funny one.
As I am using DSLRs and mirrorless (Sony A9) I am not a dslr fanboy, try not pushing me in that direction. I just don't like exaggerations. Regarding science: you should be careful in trying to play that card. I am scientist in my main job (full engineering science university professor, photographing weddings for several years now at weekends, so knowing that photographic side very well).



Oct 14, 2017 at 04:23 PM
1       2      
3
       4       5       end






FM Forums | Sony Forum | Join Upload & Sell

1       2      
3
       4       5       end
    
 

You are not logged in. Login or Register

Username     Reset password