DXOMark Reviews Zeiss Otus 85mm f/1.4

[Loswr]

Effectively, the lens goes from a score of 38 to 49 solely based on the increased resolving power of the Nikon. See attached.

You fell victim to one of the classic blunders. The most famous is: Never get involved in a land war in Asia. But only slightly less well known is this: Believing that DxO's optical metrics are the primary determinants of the Lens Score.

I know, as you show in the screenshot DxO sticks that Score right on top of the list of lens metrics, suggesting the Score is a synthesis of those measurements. It's not. The Lens Score is primarily based on 'performance in 150 lux illumination' (the light level found in a dim warehouse). What that means in practice is the the Lens Score is heavily influenced by the Sensor Overall Score of the body on which it is tested. So, while some of the Zeiss 85/1.4 Otus' higher lens Score on the D800E comes from the higher P-MPix value, the relative difference in the sensor Score also accounts for much of that difference. For example, comparing the two current 600mm lenses, we see that the Canon lens is sharper, has higher transmission, less distortion, equal vignetting, and less CA...the Canon lens is optically superior, but the two lenses get the same score because the D800 sensor scores higher (and I should point out that DxO has explained that their Sensor Score is independent of resolution, so it's not the 22 vs. 36 MP accounting for that difference). Also worth noting that their chosen conditions for the lens score – dim light – would generally require a high ISO setting, yet their Sensor Score is biased toward low ISO (2 of the three subscores are considered only at base ISO). Bias on top of bias. That's DxO's biased scores – aka BS – for you...

Nice work, DXO, you have again failed to earn any respect as a neutral review source.

That part you got perfectly correct.

 

[RGomezPhotos]

I saw a great video on the practical differences between the Zeiss Otus 85mm and the Nikkor 85mm. This Nikkor is one of the best lenses and I was impressed by how much better the Zeiss performed over it. Coupled with the D810, the detail, contrast and CA performance was very impressive.

I'd only get one if I were a full-time professional and using a Nikon D810.

 

[TWI by Dustin Abbott]

I'm actually surprised by the light transmission issue here. The front element on the Otus is HUGE (86mm, if I recall). I would expect light transmission to be better.

Still, watching this video makes it clear how HUGE of a difference there is in microcontrast with this lens: https://www.youtube.com/watch?v=S_-vUXkOaOY&list=UUL5Hf6_JIzb3HpiJQGqs8cQ

Where I reviewed the Zeiss Sonnar T 135/2, it literally destroyed my beloved 135L in this regard.

 

[jdramirez]

I'm actually surprised by the light transmission issue here. The front element on the Otus is HUGE (86mm, if I recall). I would expect light transmission to be better.

Still, watching this video makes it clear how HUGE of a difference there is in microcontrast with this lens: https://www.youtube.com/watch?v=S_-vUXkOaOY&list=UUL5Hf6_JIzb3HpiJQGqs8cQ

Where I reviewed the Zeiss Sonnar T 135/2, it literally destroyed my beloved 135L in this regard.

It's probably just me, but I wasn't blown away by the images in the video. Maybe I am now just jaded.

 

[Sporgon]

Has anyone noticed that the 'true' max apertures as indicate by the Tstop is nowhere near the manufacturers claim? The Zeis is closer to a f1.8 lens than f1.4 and the rest fare no better....

T value is not aperture value, an f1.4 lens is a "true" f1.4 if the apparent aperture diameter is focal length divided by 1.4. The T value relates to actual light transmission and is pretty much irrelevant with TTL metering stills cameras.

Aperture value is always lower than T (transmission) value because however good the glass is you always lose some.

Quite a few of the latest EF lenses do seem to have a T value that is the same as the aperture: the 24-70 IS, 40 pancake, 24/28/35 IS primes. These are all slower lenses but it does look like Canon are achieving a very high light transmission efficiency - you know - to make up for the sensor.........

 

[TWI by Dustin Abbott]

Has anyone noticed that the 'true' max apertures as indicate by the Tstop is nowhere near the manufacturers claim? The Zeis is closer to a f1.8 lens than f1.4 and the rest fare no better....

T value is not aperture value, an f1.4 lens is a "true" f1.4 if the apparent aperture diameter is focal length divided by 1.4. The T value relates to actual light transmission and is pretty much irrelevant with TTL metering stills cameras.

Aperture value is always lower than T (transmission) value because however good the glass is you always lose some.

Quite a few of the latest EF lenses do seem to have a T value that is the same as the aperture: the 24-70 IS, 40 pancake, 24/28/35 IS primes. These are all slower lenses but it does look like Canon are achieving a very high light transmission efficiency - you know - to make up for the sensor.........

LOL - but a solid point, and part of why I was surprised that the T-stop lagged so much with the Otus 85. I raised the point in my 35IS review that the light transmission between the Sigma 1.4Art and the 35IS wasn't all that big. The 35IS feels like it has a wider aperture than f/2 (although it actually just REALLY has an f/2 aperture, instead of a f/2.3 or such.)

It's probably just me, but I wasn't blown away by the images in the video. Maybe I am now just jaded.

I know what you mean. Several of them seemed really underexposed, but the crops certainly showed the great detail rendering.

 

[infared]

Interesting.

I take everything from DxO with a grain of salt, but I include their info in my mix and assess accordingly.

I will not be buying an Otus 85mm because of manual focus and price.

I own a Canon 85 f/1.2L II and absolutely love it.

Perhaps a new Sigma Art could challenge all of this. (along with some focusing issues!!! LOL!)

Side note: DxO rates the existing Sigma 85mm at higher sharpness than the Canon 85L II.
(that is there overall rating, not shown here..they only have the sharpness numbers)

...but this reviewer (below) put 4 or 5 copies of the old Sigma against the Canon L II and totally
disagreed with the DxO findings. Just food for thought.

http://cannonfastreviews.com/canon-85mm-f1-2-l-ii-vs-sigma-85mm-f1-4-ex

 

[drjlo]

I have to but ask for whom is 85mm Otus meant? I mean, where is the significant difference (when there is significant PRICE difference) in optical perfomance to its peers?

Well, according to Mr. Thein at the link review above:

"I’m sure many keystrokes will be wasted to explaining why alternative X at $1000 is better. If you have to even ask why, this is not the lens for you."

Nice, eh?

 

[BozillaNZ]

I know, DXO will always rate Canon lower than others so what? the more they try to push me the more I will stick. What can they do about it?

Yes! Nikon is the best but no, I won't buy one thabkyouverymuch!

 

[jdramirez]

Has anyone noticed that the 'true' max apertures as indicate by the Tstop is nowhere near the manufacturers claim? The Zeis is closer to a f1.8 lens than f1.4 and the rest fare no better....

T value is not aperture value, an f1.4 lens is a "true" f1.4 if the apparent aperture diameter is focal length divided by 1.4. The T value relates to actual light transmission and is pretty much irrelevant with TTL metering stills cameras.

Aperture value is always lower than T (transmission) value because however good the glass is you always lose some.

Quite a few of the latest EF lenses do seem to have a T value that is the same as the aperture: the 24-70 IS, 40 pancake, 24/28/35 IS primes. These are all slower lenses but it does look like Canon are achieving a very high light transmission efficiency - you know - to make up for the sensor.........

LOL - but a solid point, and part of why I was surprised that the T-stop lagged so much with the Otus 85. I raised the point in my 35IS review that the light transmission between the Sigma 1.4Art and the 35IS wasn't all that big. The 35IS feels like it has a wider aperture than f/2 (although it actually just REALLY has an f/2 aperture, instead of a f/2.3 or such.)

It's probably just me, but I wasn't blown away by the images in the video. Maybe I am now just jaded.

I know what you mean. Several of them seemed really underexposed, but the crops certainly showed the great detail rendering.

Yeah, it reminded me of my 100L macro.

 

[eml58]

When I'm shooting this Lens on my 1Dx or 5DMK III or a7r (Delivery first week November because it's Australia), I'm pretty sure what DXO think, or don't think of the Lens will not be high on my thought process, I'm reasonably sure it'll be Subject, Focus, f/stop, exposure, satisfaction and a Glass of good Red.

 

[deleteme]

I'd only get one if I were a full-time professional and using a Nikon D810.

While I admire the quality of the lens I can scarcely imagine a full time pro wanting to buy one. Manual focus for what is a portrait lens seems to be a recipe for frustration.
The other reason a pro would not buy it is because $4500 is a lot of working capital tied up in a tool that no client will appreciate.

 

[Mt Spokane Photography]

Title of Nikon + Otus review = "Outstanding performance"

Title of Canon + Otus review = "Sharpness limited by sensor"

Though the second statement is potentially fairly made point -- that the lens can outresolve the Canon sensor -- the tone of the words is damning. Nice work, DXO, you have again failed to earn any respect as a neutral review source.

- A

That's why I'd like to see the lens rentals data from their machine that tests the lens on a stand-alone basis. No camera body involved that influences the results. Zeiss also generates their MTF curves from actual testing, so they are a good source of actual lens performance.

 

[Zv]

I'd only get one if I were a full-time professional and using a Nikon D810.

While I admire the quality of the lens I can scarcely imagine a full time pro wanting to buy one. Manual focus for what is a portrait lens seems to be a recipe for frustration.
The other reason a pro would not buy it is because $4500 is a lot of working capital tied up in a tool that no client will appreciate.

Yeah it does seem like a poor business decision. I guess you could make that money back if you're one of the top pros earning thousands per job. For the average working person it's like 2 months salary for almost no gain in extra clients or work. Don't forget you have to insure that bad boy too!

I have a feeling Sigma are going to bring out an updated Art version of theirs soon at a quarter of the price and 99% of the IQ. The current one gets the job done as it is for most folk looking for an affordable 85.

 

[Aglet]

I'd like to see Rodger Cicala run it thru his new test machine. While I'm sure that the lens is absolutely a gem, there are always weak spots in a product due to tradeoffs involved in design and manufacturing. Knowing what they are would be mostly out of my curiosity, since I won't buy one in any event.

Yes, altho this thing is PD sharp, I don't care for the swirly-circular bokeh in the background of some of those examples when it's wide open. It does improve as it's stopped down. This would not always be an issue but I don't like everything about how it renders OOF areas.
Give me the Samyang for $300 !

 

[talicoa]

DXO can score this however they want, but when they say things like this: "The new Otus 85mm is without question the most desirable and best performing 85mm portrait lens available" I have no desire to continue reading what they are selling.

What are they smoking? I still haven't figured out what use case there is for this lens. You can't hand hold this lens for closer portraits wide open. You can't. Not while focusing manually.

I will guarantee that I could take sharper photos with my Canon 85mm 1.8 when doing portraits within 10 ft of the subject. I have a pretty steady hand, but there is always enough movement with a standing subject to sway forward and back a few inches. With the Otus, you would have some very sharp ears and the tip of people's noses. That isn't really what I am interested in.

maybe some still life shots would work well on a tripod?

What am I missing? Why is this the Most Desirable and Best performing 85mm?

 

[Hjalmarg1]

I'd only get one if I were a full-time professional and using a Nikon D810.

While I admire the quality of the lens I can scarcely imagine a full time pro wanting to buy one. Manual focus for what is a portrait lens seems to be a recipe for frustration.
The other reason a pro would not buy it is because $4500 is a lot of working capital tied up in a tool that no client will appreciate.

Yeah it does seem like a poor business decision. I guess you could make that money back if you're one of the top pros earning thousands per job. For the average working person it's like 2 months salary for almost no gain in extra clients or work. Don't forget you have to insure that bad boy too!

I have a feeling Sigma are going to bring out an updated Art version of theirs soon at a quarter of the price and 99% of the IQ. The current one gets the job done as it is for most folk looking for an affordable 85.

Looking forward to see the Sigma 85mm f1.4 Art. The current version outresolve the mighty Canon 85mm f1.2L in some aspects and come close in others. The new 'Art' should be a winner.

 

[Eldar]

DXO can score this however they want, but when they say things like this: "The new Otus 85mm is without question the most desirable and best performing 85mm portrait lens available" I have no desire to continue reading what they are selling.

What are they smoking? I still haven't figured out what use case there is for this lens. You can't hand hold this lens for closer portraits wide open. You can't. Not while focusing manually.

What am I missing? Why is this the Most Desirable and Best performing 85mm?

We did shoot f1.4 portraits back in the old days, where no AF was available. I had the first 50mm f1.2L lens and have lots of good shots with it, wide open.

A major part of being a photographer then was to practice your manual focusing skills. People were also shooting all kinds of sports and wildlife at the time. Very few (if any) shot f1.4 though, but you still had to handle the movement issues. So adding a precision focusing screen to your camera (I use Ec-S on my 1DX), you can clearly shoot portraits handheld with this lens. But you have to practice quite a bit to master it. If you pick up these manual lenses only once every now and then, I agree, you will not make it. I have lots of close portraits taken with the Otus 55mm at f1.4 (and the Zeiss 135mm at f2.0), where focus is exactly where I want it, so I imagine I will be able to do it with the 85mm also.

 

[RGomezPhotos]

I'd only get one if I were a full-time professional and using a Nikon D810.

While I admire the quality of the lens I can scarcely imagine a full time pro wanting to buy one. Manual focus for what is a portrait lens seems to be a recipe for frustration.
The other reason a pro would not buy it is because $4500 is a lot of working capital tied up in a tool that no client will appreciate.

Yeah it does seem like a poor business decision. I guess you could make that money back if you're one of the top pros earning thousands per job. For the average working person it's like 2 months salary for almost no gain in extra clients or work. Don't forget you have to insure that bad boy too!

I have a feeling Sigma are going to bring out an updated Art version of theirs soon at a quarter of the price and 99% of the IQ. The current one gets the job done as it is for most folk looking for an affordable 85.

Yes, you have to GRADUATE to this level of lens. If two month's salary is $4500, then you haven't reached that level. I don't even know serious, semi-professional photographers who earn that little. That's where I'm coming from.

In that context, I'll say again that this lens is for the full-time professional, wealthy amateur or those maniacal enthusiast types who rather take the bus than buy a car to afford this lens. Especially love that last category of folks. I know full-time pros that need to budget for that super-lens. Because they are reaching that level in their business when they need the performance, functionality or durability that lens will give them.

And at that level, $4500 is simply another business expense. Laptops, computers, advertising, marketing, health insurance and other expenses are all going to cost as much if not more than that lens. Hell, most professionals I know who use a cell phone for business have a monthly bill at least $100/month. If you stop your service after 4 years, you got a worthless piece of plastic. But you still have a lens. And since I insure my gear, I can estimate that insuring this lens will cost about $5 - $7/month. That's a cheap lunch.

And for portraits, AF is a convenience. Even if you're shooting at f1.4, AF isn't going to nail focus much if any faster than manually. And at that aperture, your AF is going to hunt for focus every time you press the shutter. MF isn't. I'm cool if someone likes AF. No problem. But I've been MF for my fashion and portrait work, forever. I don't miss it. "Recipe for Frustration"? If you have really bad eye sight and don't have glasses. Yes, I guess it is.

Lastly, even if I were at that level to afford this lens, even if I shot with a Nikon D810, I don't know if I would get this lens. Detail is AMAZING... And if you do lots of close-ups, it might be worthwhile. But as I mentioned numerous times on canonrumors.com, my Canon setup is sufficient until I can graduate to Medium Format. Because 'the look' of MF is more important than pure sharpness. Unless Canon comes out with a sensor that can give me that same look, possibly with that 5-layer sensor that they patented recently, that's my plan.

 

[talicoa]

DXO can score this however they want, but when they say things like this: "The new Otus 85mm is without question the most desirable and best performing 85mm portrait lens available" I have no desire to continue reading what they are selling.

What are they smoking? I still haven't figured out what use case there is for this lens. You can't hand hold this lens for closer portraits wide open. You can't. Not while focusing manually.

What am I missing? Why is this the Most Desirable and Best performing 85mm?

We did shoot f1.4 portraits back in the old days, where no AF was available. I had the first 50mm f1.2L lens and have lots of good shots with it, wide open.

A major part of being a photographer then was to practice your manual focusing skills. People were also shooting all kinds of sports and wildlife at the time. Very few (if any) shot f1.4 though, but you still had to handle the movement issues. So adding a precision focusing screen to your camera (I use Ec-S on my 1DX), you can clearly shoot portraits handheld with this lens. But you have to practice quite a bit to master it. If you pick up these manual lenses only once every now and then, I agree, you will not make it. I have lots of close portraits taken with the Otus 55mm at f1.4 (and the Zeiss 135mm at f2.0), where focus is exactly where I want it, so I imagine I will be able to do it with the 85mm also.

So are you saying that a well practiced manual focusing photographer could use this lens as an effective portrait lens in close quarters wide open? What do you think the keeper rate would be hand held? What percentage of those keepers would use to the fullest extent this very expensive glass? I think it would be disappointingly low.
One of the first posts here had a link to some real world examples, I even went to the flickr site referenced. Guess what? No portraits. Some beautiful photos, but no portraits.

In the past photographers practiced their manual focusing. Nowdays it is the videographers out trying to perfect this skill.

I agree that this lens could have a place, but my real gripe was with DXO, who doesn't seem to understand the lens other than by its stats.