The great 70-200mm f/2.8 shootout, Canon vs Nikon, Panasonic and Sony

Nov 2, 2016
849
648
That's the problem with much of the photo industry today; so much choice yet in reality so little choice.

Say I identify that I need a 35-150 f2.8. Where are they? Or perhaps a 50-135 f2. Nope.

But here are half a dozen 70-200 f2.8, now just buy one like a good little consumer.
You actually think that manufacturers are going to make every odd size a few people want? It’s very expensive to design and manufacture lenses. They have to know, or at least expect, that enough people will buy it. I’d you “find” that you need a lens that pretty much nobody else needs, it’s never going to be made. What if someone else “needs“ a 30-155 f 2.5, or a 40-170 f2? They should make those also?
 
  • Like
Reactions: 2 users
Upvote 0

justaCanonuser

Grab your camera, go out and shoot!
Feb 12, 2014
1,035
931
Frankfurt, Germany
DPR misses the most important things, forest for the trees, gear reviewers who aren't true photographers (this is why for example they hyper focus on dynamic range more than anything else in a camera body, and why their sample galleries look like a 5-year-old took them). In this case, they don't test what a 70-200 f/2.8 is primarily for: portraiture. Meaning, how well does it render people, how is the bokeh, how is the contrast? These are arguably the most important qualities for this lens (along with sharpness which they do test) which can't be measured on a test chart, it requires taking many shots and comparisons to get a feeling of overall rendering. For quality reviews from people who actually do photography, I find Bryan Carnathan at tdp and Dustin Abbott much more insightful.
I agree with others here that I wouldn't regard the EF 70-200 f/2.8 primary as a portrait lens, but one certainly can shoot quite nice portraits with it. The background shouldn't be too busy, then you get a not too bad bokeh either. So, it is a quite versatile lens, and what I in particular like about my Mk II version is its quite reliable AF performance with a 5D3/4 combo.

But I totally agree with your comment about DPR, Bryan Carnathan and Dustin Abbott, Woodman. Gear-heads and real photographers are not necessarily living in the same ecosystem - a skilled photographer could even get eye-catching images with a 100 years old Kodak Brownie ;)
 
  • Like
Reactions: 1 users
Upvote 0
I agree with others here that I wouldn't regard the EF 70-200 f/2.8 primary as a portrait lens, but one certainly can shoot quite nice portraits with it. The background shouldn't be too busy, then you get a not too bad bokeh either. So, it is a quite versatile lens, and what I in particular like about my Mk II version is its quite reliable AF performance with a 5D3/4 combo.

But I totally agree with your comment about DPR, Bryan Carnathan and Dustin Abbott, Woodman. Gear-heads and real photographers are not necessarily living in the same ecosystem - a skilled photographer could even get eye-catching images with a 100 years old Kodak Brownie ;)

My point about dpr is that they're just scratching the surface, and coming to conclusions based on these surface tests (eg. autofocus rated by racking speed... really? What about bokeh, iq/rendering, build quality, etc). Because of this, this Canon "victory" means nothing for me, as does most dpr ratings. If dpr is all we've got, fine, but thankfully there are other resources available that focus on the more important things of a lens or camera body.
 
  • Like
Reactions: 1 users
Upvote 0

Aussie shooter

https://brettguyphotography.picfair.com/
Dec 6, 2016
1,183
1,817
brettguyphotography.picfair.com
My point about dpr is that they're just scratching the surface, and coming to conclusions based on these surface tests (eg. autofocus rated by racking speed... really? What about bokeh, iq/rendering, build quality, etc). Because of this, this Canon "victory" means nothing for me, as does most dpr ratings. If dpr is all we've got, fine, but thankfully there are other resources available that focus on the more important things of a lens or camera body.
They did consider several factors though. Racking speed, bokeh, size/weight, video performance, sharpness etc. But the reality is that ANY professional 70-200 you buy is for all intents and ourposes going to perform equally well in all situations. No viewer on the planet is going to look at an image and deduce which lens it was taken on. TBH. I highly doubt 99.9% of viewers would care as long as the image is nice.
 
  • Like
Reactions: 1 user
Upvote 0
DPR misses the most important things, forest for the trees, gear reviewers who aren't true photographers (this is why for example they hyper focus on dynamic range more than anything else in a camera body, and why their sample galleries look like a 5-year-old took them). In this case, they don't test what a 70-200 f/2.8 is primarily for: portraiture. Meaning, how well does it render people, how is the bokeh, how is the contrast? These are arguably the most important qualities for this lens (along with sharpness which they do test) which can't be measured on a test chart, it requires taking many shots and comparisons to get a feeling of overall rendering. For quality reviews from people who actually do photography, I find Bryan Carnathan at tdp and Dustin Abbott much more insightful.

In fact he did talk about bokeh, bokeh balls (onion rings and shape) and LoCA at 2:54 and included photos of his daughter:

Of course he also talked about sharpness, which does include subjective perceptions about contrast (the two are not mutually exclusive). I agree the sample photos are not particularly inspiring. I disagree that the 70-200 is primarily used for portraiture.

Personally speaking, so-called "rendering" is highly subjective and connoisseurs often describe it in non-technical terms (colourfully apt descriptions). Rendering, other than a faithful facsimile of the subject usually means the lens is under corrected in some manner (e.g., residual spherical aberration -- characteristic of fast aperture lenses). Anyone who is willing to spend over $2000 on a lens will undoubtedly demand a highly corrected, "sterile" lens. One can instantly "add" (i.e., subtract accutance) haze/veiling flare in post.
 
  • Like
Reactions: 1 user
Upvote 0
In fact he did talk about bokeh, bokeh balls (onion rings and shape) and LoCA at 2:54 and included photos of his daughter:

Of course he also talked about sharpness, which does include subjective perceptions about contrast (the two are not mutually exclusive). I agree the sample photos are not particularly inspiring. I disagree that the 70-200 is primarily used for portraiture.

Personally speaking, so-called "rendering" is highly subjective and connoisseurs often describe it in non-technical terms (colourfully apt descriptions). Rendering, other than a faithful facsimile of the subject usually means the lens is under corrected in some manner (e.g., residual spherical aberration -- characteristic of fast aperture lenses). Anyone who is willing to spend over $2000 on a lens will undoubtedly demand a highly corrected, "sterile" lens. One can instantly "add" (i.e., subtract accutance) haze/veiling flare in post.

Bokeh balls go only so far, just as racking speed goes only so far (there is much more to bokeh and autofocus than that). They are just scratching the surface. Rendering may be subjective, but the quality of rendering, and the differences in this quality, is real. It's why Zeiss Otus lenses and various L lenses have a special feel to them, regardless of what the test charts say.
 
Upvote 0
Bokeh balls go only so far, just as racking speed goes only so far (there is much more to bokeh and autofocus than that). They are just scratching the surface. Rendering may be subjective, but the quality of rendering, and the differences in this quality, is real. It's why Zeiss Otus lenses and various L lenses have a special feel to them, regardless of what the test charts say.

Real and yet unarticulated or unquantified by either Dpreview or you. Do tell us about those differences in palpable, if not technical, terms, particularly their impact on portrait photography, it's consumers and your wallet.
 
Upvote 0
Real and yet unarticulated or unquantified by either Dpreview or you. Do tell us about those differences in palpable, if not technical, terms, particularly their impact on portrait photography, it's consumers and your wallet.

Like I said, it is not measurable, to be clear, I mean not measurable on the consumer side (I'm sure lens engineers know exactly what gives rendering characteristics that make both a $200 50 prime versus a $2000 50 prime), but since we are not lens engineers, all we can do is to take many shots with one lens, and then another, and compare, ideally under similar circumstances. It is one of the first things I did when I got into photography, I asked what's the difference between a cheap pancake prime versus a massive expensive prime versus a zoom that covers the same range (besides the obvious differences of speed). I got all of them and took thousands of shots and compared. I got into 3rd party lenses and compared them to Canon and even got into a few Zeiss lenses. Yes there are differences in sharpness, but over time, you get to see many other characteristics of a lens. They are there, they are real, and they are not measured, at least by review sites and consumers.
 
Last edited:
Upvote 0
Like I said, it is not measurable, to be clear, I mean not measurable on the consumer side (I'm sure lens engineers know exactly what gives rendering characteristics that make both a $200 50 prime versus a $2000 50 prime), but since we are not lens engineers, all we can do is to take many shots with one lens, and then another, and compare, ideally under similar circumstances. It is one of the first things I did when I got into photography, I asked what's the difference between a cheap pancake prime versus a massive expensive prime versus a zoom that covers the same range (besides the obvious differences of speed). I got all of them and took thousands of shots and compared. I got into 3rd party lenses and compared them to Canon and even got into a few Zeiss lenses. Yes there are differences in sharpness, but over time, you get to see many other characteristics of a lens. They are there, they are real, and they are not measured, at least by review sites and consumers.

What you're referring to is "mystique", which is no description at all. It is neither qualitative, semi-quantitative or quantitative. Such "reviews" are not based on any established evaluation criteria and therefore negate objectivity (although they can be excused if they provide decent SOOC comparison photos). How can anyone rationally part with $2000 or even $200 or be dissuaded from doing so on that basis?

Practically speaking, no one, including yourself or Dpreview, is going to buy all the aforementioned 70-200s, take thousands of photos with each and then describe their merits for one genre of photography (i.e., portraiture). In the absence of a universally accepted test methodology, brick wall/ test targets/ MTF curves and even pictures of a household cat may have to do.
 
Upvote 0

usern4cr

R5
CR Pro
Sep 2, 2018
1,376
2,308
Kentucky, USA
What you're referring to is "mystique", which is no description at all. It is neither qualitative, semi-quantitative or quantitative. Such "reviews" are not based on any established evaluation criteria and therefore negate objectivity (although they can be excused if they provide decent SOOC comparison photos). How can anyone rationally part with $2000 or even $200 or be dissuaded from doing so on that basis?

Practically speaking, no one, including yourself or Dpreview, is going to buy all the aforementioned 70-200s, take thousands of photos with each and then describe their merits for one genre of photography (i.e., portraiture). In the absence of a universally accepted test methodology, brick wall/ test targets/ MTF curves and even pictures of a household cat may have to do.
OK - "even pictures of a household cat may have to do" ?

I'll send you all some I took a few weeks ago with my R5, RF 70-200 f2.8L, with a B+W XS-Pro Clear Nano filter. Handheld, animal eye AF, electronic 1st shutter & mechanical 2nd shutter, raw on all. DXO Photolab4 (& deep prime) to my taste in post. Originals are slightly cropped for best look, then resized to 2.5k pixels longest side to download here, and are followed by a 1:1 crop around the face & eyes to show sharpness.

*EDIT* - I've been reading a lot more about the benefit/downsides of using any clear protective filter in general, including the highest quality ones. I'm now going to take off the filters for most usage, and only put them on when in dusty/sandy environments.

iso 400, f8, 1/750", 70mm
A01_1452_1.jpg

A01_1452_2_1:1.jpg

iso 400, f8, f/1000", 70mm
A01_1490_1.jpg

A01_1490_2_1:1.jpg

iso 400, f8, 1/500", 70mm
A01_1518_1.jpg

A01_1518_2_1:1.jpg

iso 100, f2.8, 1/1000", 135mm
A01_2067_1.jpg

A01_2067_2_1:1.jpg

iso 400, f8, 1/1000", 200mm
A01_1428_1.jpg

A01_1428_2_1:1.jpg

OK - I might not have photos from other 70-200 bodies & lenses to compare to, but I just thought I'd have some fun sending these off to all of you to enjoy. :)
 
Last edited:
  • Like
Reactions: 3 users
Upvote 0
What you're referring to is "mystique", which is no description at all. It is neither qualitative, semi-quantitative or quantitative. Such "reviews" are not based on any established evaluation criteria and therefore negate objectivity (although they can be excused if they provide decent SOOC comparison photos). How can anyone rationally part with $2000 or even $200 or be dissuaded from doing so on that basis?

Practically speaking, no one, including yourself or Dpreview, is going to buy all the aforementioned 70-200s, take thousands of photos with each and then describe their merits for one genre of photography (i.e., portraiture). In the absence of a universally accepted test methodology, brick wall/ test targets/ MTF curves and even pictures of a household cat may have to do.

And this is where experience counts, where photographers are separated from gear reviewers, who already have a feel for various lenses because they already have taken thousands of pictures, they already have a feel of where a lens fits in the "pecking order". Sounds like you're hell bent on sticking with gear reviewers, your choice of course. I'll go with real photographers and their insights (Bryan Carnathan, Dustin Abbott, Alik Griffin, and a host of Canon ambassadors like Peter Mckinnon, Vanessa Joy, Irene Rudnyk, etc. But they're biased right?) Go with what works for you.
 
Last edited:
Upvote 0
And this is where experience counts, where photographers are separated from gear reviewers, who already have a feel for various lenses because they already have taken thousands of pictures, they already have a feel of where a lens fits in the "pecking order". Sounds like you're hell bent on sticking with gear reviewers, your choice of course. I'll go with real photographers and their insights (Bryan Carnathan, Dustin Abbott, Alik Griffin, and a host of Canon ambassadors like Peter Mckinnon, Vanessa Joy, Irene Rudnyk, etc. But they're biased right?) Go with what works for you.

Funny, a couple of those "photographers" I would be more inclined to call "gear reviewers". Either way, they only provide one data point for any lens that I might be interested in.

Vanessa Joy? Never heard of her until now. Interesting that she compares an EF 50/1.2 L and a 50/1.8 (I've owned both).


What point do you think she's making?

In any case, you seem "hell bent" on avoiding addressing my points. If it's all about the experience, as you say, then you will know that most lenses will get the job done, aesthetically, artistically, professionally or recreationally -- of course in the hands of a reasonably competent photographer.
 
  • Like
Reactions: 1 user
Upvote 0
Jan 29, 2011
10,675
6,121
Funny, a couple of those "photographers" I would be more inclined to call "gear reviewers". Either way, they only provide one data point for any lens that I might be interested in.

Vanessa Joy? Never heard of her until now. Interesting that she compares an EF 50/1.2 L and a 50/1.8 (I've owned both).


What point do you think she's making?

In any case, you seem "hell bent" on avoiding addressing my points. If it's all about the experience, as you say, then you will know that most lenses will get the job done, aesthetically, artistically, professionally or recreationally -- of course in the hands of a reasonably competent photographer.
She is a photographer, an actual working pro who shot weddings with 1DX II‘s long before she got on Canon’s radar. She has an impressive website of current wedding images, she is working the social media angle but then that is how many people supplement their income nowadays, but first and foremost she is a professional photographer who knows her business and her gear. Her husband is a pretty decent videographer too.
 
Last edited:
Upvote 0

jd7

CR Pro
Feb 3, 2013
1,064
418
What you're referring to is "mystique", which is no description at all. It is neither qualitative, semi-quantitative or quantitative. Such "reviews" are not based on any established evaluation criteria and therefore negate objectivity (although they can be excused if they provide decent SOOC comparison photos). How can anyone rationally part with $2000 or even $200 or be dissuaded from doing so on that basis?

Practically speaking, no one, including yourself or Dpreview, is going to buy all the aforementioned 70-200s, take thousands of photos with each and then describe their merits for one genre of photography (i.e., portraiture). In the absence of a universally accepted test methodology, brick wall/ test targets/ MTF curves and even pictures of a household cat may have to do.

I would agree that it is worth paying attention to measurable characterists of lenses. Quantifying differences can certainly give you useful information, and for example can help avoid wasting money on gear which really isn't better than cheaper gear, or avoid overlooking gear which would actually be good. However, at least if the purpose of photography is to make art (perhaps may be different if, for example, you do forensic photography and your aim is to record a scene as faithfully as possible?), surely what ultimately matters is how viewers react to an image? And if in the end you believe your intended audience will prefer images from one lens than another, why ignore that just because you cannot point to some quantiifable difference in a particular lens characteristc? Obviously, getting carried away with particular gear giving some particular "look" risks buying into hype and wasting money on gear which isn't making any real difference, etc, but in my view that is not a reason to simply say that unquantifiable (or at least unquantified) differences can never be real.
 
Upvote 0

I would agree that it is worth paying attention to measurable characterists of lenses. Quantifying differences can certainly give you useful information, and for example can help avoid wasting money on gear which really isn't better than cheaper gear, or avoid overlooking gear which would actually be good. However, at least if the purpose of photography is to make art (perhaps may be different if, for example, you do forensic photography and your aim is to record a scene as faithfully as possible?), surely what ultimately matters is how viewers react to an image? And if in the end you believe your intended audience will prefer images from one lens than another, why ignore that just because you cannot point to some quantiifable difference in a particular lens characteristc? Obviously, getting carried away with particular gear giving some particular "look" risks buying into hype and wasting money on gear which isn't making any real difference, etc, but in my view that is not a reason to simply say that unquantifiable (or at least unquantified) differences can never be real.

You're confounding the optical performance of a lens, which is largely, if not wholly measurable, with the subjective perceptions of its user and his audience. What do the latter really have to do with a "review", a systematic and objective evaluation against defined criteria? If what really matters in a review is how to impress an audience, then we should focus more on the brand mystique, price and size of our equipment. These qualities never fail to impress, even if one takes crap photos. Yes, let's make reviews about how to "manage" optics rather than to scientifically evaluate them.
 
Upvote 0

jd7

CR Pro
Feb 3, 2013
1,064
418
You're confounding the optical performance of a lens, which is largely, if not wholly measurable, with the subjective perceptions of its user and his audience. What do the latter really have to do with a "review", a systematic and objective evaluation against defined criteria? If what really matters in a review is how to impress an audience, then we should focus more on the brand mystique, price and size of our equipment. These qualities never fail to impress, even if one takes crap photos. Yes, let's make reviews about how to "manage" optics rather than to scientifically evaluate them.
No, I am not confounding optical performance and viewer perception at all.

In any event, if you are a professional and the impression your gear makes on your clients - due to its size, brand, price, etc - makes a difference, I think you'd be mad to simply dismiss that as entirely irrelevant, so yes, I think you can make an argument that those things are worth mentioning in a review.

However, I was thinking more in terms of the impression a photo creates on an audience when the audience doesn't know what gear was used to take it. And yes, how the image will be subjectively perceived is, in my view, ultimately more important than any objectively measureable characteristic, and therefore it has a a lot to do with a review of the gear. As I said, there is good reason to pay attention to objectively measurable characteristics. But in my view there is also good reason not to be a slave to them such that you ignore the possibily that non-measurable (or at least non-measures) characteristics can be important too. The difficulty, of course, is deciding how much weight to put on impressions which are hard, if not impossible, to support with objective data. That is not the same, though, as saying the lack of objective data to support something necessarily means it is irrelevant or wrong.

If you want to choose your gear based solely on measurable differences, and your only interest is in reviews which only consider such differences, you are welcome to take that approach, of course. Not everyone is going to agree that that is the only, or even a, sensible approach to take though.
 
Last edited:
Upvote 0

justaCanonuser

Grab your camera, go out and shoot!
Feb 12, 2014
1,035
931
Frankfurt, Germany
OK - "even pictures of a household cat may have to do" ?

I'll send you all some I took a few weeks ago with my R5, RF 70-200 f2.8L, with a B+W 77mm XS-Pro Clear MRC-Nano 007 Filter. Handheld, animal eye AF on all. DXO Photolab4 to my taste in post. Originals are slightly cropped to taste, then resized to 2.5k pixels longest side. 1:1 crops after each original are 2.5k pixels wide.

OK - I might not have photos from other 70-200 bodies & lenses to compare to, but I just thought I'd have some fun sending these off to all of you to enjoy. :)

Your images show all hairy details of life that photography can cat(ch) ;)
 
  • Haha
Reactions: 1 user
Upvote 0
how the image will be subjectively perceived is, in my view, ultimately more important than any objectively measureable characteristic, and therefore it has a a lot to do with a review of the gear.

No. Reviews are not focus testing, or at least rarely so. Again confounding subjective perceptions with the objectively measurable. A review is for the latter.

Even if focus testing was done for images produced by a given lens, would it be informative? For example, earlier I linked Vanessa Joy's comparison of the 50/1.8 vs 50/1.2. What do you think the average non-photographer tell you? What would seasoned photographers tell you about image quality of either lens in a "blind" test? How would that information weigh into your purchasing decision?

See?
 
  • Like
Reactions: 1 user
Upvote 0