Update about the RF 85mm F1.4 VCM

Hi All

Around the time the 85mm F1.4 VCM was launched, there was some chatter about whether it was 'actually' an 80mm, as it appears to exhibit a wider field of view than the 85mm F1.2 L USM does. There was also some concern about whether this discrepancy was a result of digital correction.

I've now had a chance to test them alongside one another. From my testing, it looks to me like it's approximately an 86mm lens (and might be considered approximately an 85.5mm if you decided to exclude the corrections).

It's the 85mm F1.2 that's longer than billed. Or, at least, can often behave that way.

My testing was conducted with a chart 1m tall, giving a working distance of around 3.5m. I am not claiming these results perfectly represent infinity focus (which focal length is usually specified) and I'd rate my measurements as pretty good, rather than being conducted to clinical levels of precision.

At this working distance, the F1.4 VCM gives a 28.2 degree diagonal AoV (with corrections), which would be consistent with it being an 86mm lens. There is little focus breathing, so I'd expect it to give a similar measurement at infinity.

The 85mm F1.2 L USM gives a tighter field of view. It shows a roughly 27 degree diagonal AoV, equating to a roughly 90mm lens at the ~3.5m working distance. However, it exhibits A LOT of focus breathing, such that it's entirely likely that it would match the 87mm focal length given in the patent for its lens formula, if focused at infinity.

This rather dramatic breathing continues, such that it gives an AoV significantly tighter than a 90mm lens at head-and-shoulder focus distances.

So it's not that the new 85mm F1.4 VCM is wider than expected, it's that the 85mm F1.2 L USM is significantly narrower than you'd expect of an 85mm lens, and the difference between the two lenses gets greater, the closer you focus.

I hope this is useful info.

Richard - DPReview.com

Thanks Richard. How does the RF 85 F2 fare in this respect?

Richard,

Chris N. commented the new F1.4 had onion bokeh causing a bit of nervous backdrop, and that Canon could have done better

comments? see below

RF F1.4 VCM @ F2 appears a bit nervous bokeh in the backdrop

https://www.dpreview.com/sample-gal...-rf-85mm-f1-4-l-vcm-sample-gallery/4524108301

RF 85 F2 IS @ F2 appears smooth bokeh in the garden

https://www.dpreview.com/sample-gal...rf-85mm-f2-is-macro-sample-gallery/1847886099

Richard Butler said:

Hi All

Around the time the 85mm F1.4 VCM was launched, there was some chatter about whether it was 'actually' an 80mm, as it appears to exhibit a wider field of view than the 85mm F1.2 L USM does. There was also some concern about whether this discrepancy was a result of digital correction.

I've now had a chance to test them alongside one another. From my testing, it looks to me like it's approximately an 86mm lens (and might be considered approximately an 85.5mm if you decided to exclude the corrections).

Click to expand...

Interesting. Then I will assume that the 85mm f2 was also built with the same FL characteristics of the 1.2 because based on what some YouTubers have shown the f2 looks the same as the 1.2 but this new 1.4 VCM lens looks wider. You can clearly see it.

Richard Butler said:

It's the 85mm F1.2 that's longer than billed. Or, at least, can often behave that way.

My testing was conducted with a chart 1m tall, giving a working distance of around 3.5m. I am not claiming these results perfectly represent infinity focus (which focal length is usually specified) and I'd rate my measurements as pretty good, rather than being conducted to clinical levels of precision.

At this working distance, the F1.4 VCM gives a 28.2 degree diagonal AoV (with corrections), which would be consistent with it being an 86mm lens. There is little focus breathing, so I'd expect it to give a similar measurement at infinity.

The 85mm F1.2 L USM gives a tighter field of view. It shows a roughly 27 degree diagonal AoV, equating to a roughly 90mm lens at the ~3.5m working distance. However, it exhibits A LOT of focus breathing, such that it's entirely likely that it would match the 87mm focal length given in the patent for its lens formula, if focused at infinity.

This rather dramatic breathing continues, such that it gives an AoV significantly tighter than a 90mm lens at head-and-shoulder focus distances.

So it's not that the new 85mm F1.4 VCM is wider than expected, it's that the 85mm F1.2 L USM is significantly narrower than you'd expect of an 85mm lens, and the difference between the two lenses gets greater, the closer you focus.

I hope this is useful info.

Richard - DPReview.com

Click to expand...

Richard,

An easy way to determine accurately the focal length of a lens at infinity is to take an image of a star field. If using a fixed camera (e.g. on a tripod) It’s best to aim at or near a celestial pole (North or South) and limit the exposure to a few seconds so that the stars don’t trail.

The image can be uploaded to nova astrometry net, which will determine where the camera was pointing and the image scale (in arcseconds / pixel).

Use the formula below to determine focal length:

Focal length = 206.265 * pixel size / Image scale

--
Andy

(apologies for the previous half-post)

Andy Lucy said:

An easy way to determine accurately the focal length of a lens at infinity is to take an image of a star field.

Click to expand...

Ha ha. Less easy in Seattle between the months of October and April.

Richard - DPReview.com

Alastair Norcross said:

Thanks Richard. How does the RF 85 F2 fare in this respect?

Click to expand...

The 85 f2 STM is a bit longer as well, as documented here:

Toggle between photo 2 & 3: This shows the STM to be narrower than the VCM

Toggle between photo 6 & 7: This shows the VCM to be very similar to the 70-200 at "85mm"

Toggle between photo 5 & 6: This shows the STM to be narrower than the USM zoom (as expected from the observation above)

Richard Butler said:

Around the time the 85mm F1.4 VCM was launched, there was some chatter about whether it was 'actually' an 80mm, as it appears to exhibit a wider field of view than the 85mm F1.2 L USM does. There was also some concern about whether this discrepancy was a result of digital correction.

I've now had a chance to test them alongside one another. From my testing, it looks to me like it's approximately an 86mm lens (and might be considered approximately an 85.5mm if you decided to exclude the corrections).

It's the 85mm F1.2 that's longer than billed. Or, at least, can often behave that way.

My testing was conducted with a chart 1m tall, giving a working distance of around 3.5m. I am not claiming these results perfectly represent infinity focus (which focal length is usually specified) and I'd rate my measurements as pretty good, rather than being conducted to clinical levels of precision.

Click to expand...

How do you measure the working distance? For angle-of-view computations you should measure the distance between the chart and the entrance pupil of the lens.

Richard Butler said:

At this working distance, the F1.4 VCM gives a 28.2 degree diagonal AoV (with corrections), which would be consistent with it being an 86mm lens. There is little focus breathing, so I'd expect it to give a similar measurement at infinity.

Click to expand...

See below.

Richard Butler said:

The 85mm F1.2 L USM gives a tighter field of view. It shows a roughly 27 degree diagonal AoV, equating to a roughly 90mm lens at the ~3.5m working distance. However, it exhibits A LOT of focus breathing, such that it's entirely likely that it would match the 87mm focal length given in the patent for its lens formula, if focused at infinity.

Click to expand...

The RF 85mm F1.2 L is present in the optical bench hub of photonstophotos.net. Its focal length is 86.3 mm at a working distance of 3.5 m, only marginally shorter than its infinity focus value of 86.5 mm.

Mind you, a lens with a constant focal length will give a narrower angle of view as you focus closer. That is simple photographic optics. Since the VCM lenses have little "focus breathing", their focal length has to vary with the working distance in order to keep a constant angle of view. The closer the focusing, the shorter the focal length.

Richard Butler said:

...My testing was conducted with a chart 1m tall, giving a working distance of around 3.5m. ...

The 85mm F1.2 L USM gives a tighter field of view. It shows a roughly 27 degree diagonal AoV, equating to a roughly 90mm lens at the ~3.5m working distance. However, it exhibits A LOT of focus breathing, such that it's entirely likely that it would match the 87mm focal length given in the patent for its lens formula, if focused at infinity.

...

Click to expand...

I'm not seeing any dramatic focus breathing.
Note that there is a Focus button and slider in the PhotonsToPhotos Optical Bench.
At ~3.5m it looks like this:


link

That 27.06 degree Angle of View is pretty close to the 27.94 degrees at infinity.

Regards

--
Bill ( Your trusted source for independent sensor data at PhotonsToPhotos )

bclaff said:

Richard Butler said:

...My testing was conducted with a chart 1m tall, giving a working distance of around 3.5m. ...

The 85mm F1.2 L USM gives a tighter field of view. It shows a roughly 27 degree diagonal AoV, equating to a roughly 90mm lens at the ~3.5m working distance. However, it exhibits A LOT of focus breathing, such that it's entirely likely that it would match the 87mm focal length given in the patent for its lens formula, if focused at infinity.

...

Click to expand...

I'm not seeing any dramatic focus breathing.
Note that there is a Focus button and slider in the PhotonsToPhotos Optical Bench.
At ~3.5m it looks like this:


link

That 27.06 degree Angle of View is pretty close to the 27.94 degrees at infinity.

Regards

--
Bill ( Your trusted source for independent sensor data at PhotonsToPhotos )

Click to expand...

There are not many focus breathing tests for that lens out there, but here:


It looks to have quite a lot of focus breathing.

bclaff said:

Richard Butler said:

...My testing was conducted with a chart 1m tall, giving a working distance of around 3.5m. ...

The 85mm F1.2 L USM gives a tighter field of view. It shows a roughly 27 degree diagonal AoV, equating to a roughly 90mm lens at the ~3.5m working distance. However, it exhibits A LOT of focus breathing, such that it's entirely likely that it would match the 87mm focal length given in the patent for its lens formula, if focused at infinity.

...

Click to expand...

I'm not seeing any dramatic focus breathing.
Note that there is a Focus button and slider in the PhotonsToPhotos Optical Bench.
At ~3.5m it looks like this:


link

That 27.06 degree Angle of View is pretty close to the 27.94 degrees at infinity.

Regards

Click to expand...

The focus breathing might not be dramatic, but, when combined with the F1.2 lens being slightly longer than 85mm at infinity, it's easily enough to account for the differences various reviewers have noticed between the two lenses. It would produce approximately a 3mm difference in perceived focal length between focusing at infinity and focusing at 3.5m. If the 85 F1.2 is 87mm at infinity, it would look like 90mm at 3.5m. If the 85 F1.4 is, as Richard says, around 85 at infinity, and at 3.5m, you get a perceived difference of around 5mm at 3.5m, which is in line with what a lot of reviewers are claiming to see. The main point that Richard is making is that the VCM lens is, in fact, pretty much what it claims to be-- an 85mm lens. It's interesting that so many people, on initially seeing the reviews, immediately jumped to the conclusion that Canon was somehow cheating by putting out an 80mm lens as an 85mm lens. Some even declared on this forum that they were no longer interested in the lens, because of this terrible discrepancy. Whereas the truth is that the VCM lens is both more accurately labelled than the F1.2 lens, and has the advantage of much less focus breathing. No good deed, eh?

--
“When I die, I want to go peacefully in my sleep like my grandfather. Not screaming in terror, like the passengers in his car.” Jack Handey
Alastair
http://anorcross.smugmug.com
Equipment in profile

Alastair Norcross said:

bclaff said:

Richard Butler said:

...My testing was conducted with a chart 1m tall, giving a working distance of around 3.5m. ...

The 85mm F1.2 L USM gives a tighter field of view. It shows a roughly 27 degree diagonal AoV, equating to a roughly 90mm lens at the ~3.5m working distance. However, it exhibits A LOT of focus breathing, such that it's entirely likely that it would match the 87mm focal length given in the patent for its lens formula, if focused at infinity.

...

Click to expand...

I'm not seeing any dramatic focus breathing.
Note that there is a Focus button and slider in the PhotonsToPhotos Optical Bench.
At ~3.5m it looks like this:

That 27.06 degree Angle of View is pretty close to the 27.94 degrees at infinity.

Regards

Click to expand...

The focus breathing might not be dramatic, but, when combined with the F1.2 lens being slightly longer than 85mm at infinity, it's easily enough to account for the differences various reviewers have noticed between the two lenses. It would produce approximately a 3mm difference in perceived focal length between focusing at infinity and focusing at 3.5m. If the 85 F1.2 is 87mm at infinity, it would look like 90mm at 3.5m. If the 85 F1.4 is, as Richard says, around 85 at infinity, and at 3.5m, you get a perceived difference of around 5mm at 3.5m, which is in line with what a lot of reviewers are claiming to see. The main point that Richard is making is that the VCM lens is, in fact, pretty much what it claims to be-- an 85mm lens. It's interesting that so many people, on initially seeing the reviews, immediately jumped to the conclusion that Canon was somehow cheating by putting out an 80mm lens as an 85mm lens. Some even declared on this forum that they were no longer interested in the lens, because of this terrible discrepancy. Whereas the truth is that the VCM lens is both more accurately labelled than the F1.2 lens, and has the advantage of much less focus breathing. No good deed, eh?

Click to expand...

A lot of this argument is due to badly defined terms. If you expect a lens to follow the simple mathematical thin lens equation 1/dₒ + 1/di = 1/f then, as the object distance dₒ decreases, the image distance di has to increase, as does the magnification m = dₒ/di. It's di that governs the field of view on the sensor, not f , though at dₒ = infinity di = f. Also, if you think in terms of di being f plus a focussing rack or a virtual (imaginary) extension tube, di = f(1 + m) which is a quick way to estimate the amount of focus breathing a simple lens will give.

However, if you expect no focus breathing as a lens focusses, di has to remain constant, so f has to decrease with dₒ for the thin lens equation to work. In my view focal length is only a very approximate indicator of field of view once you start compensating for distortion and focussed distance.

You're using think lens equations of a real world lens. Thick equations might be more appropriate.

Alastair Norcross said:

Alastair Norcross said:

...

Click to expand...

The focus breathing might not be dramatic, but, when combined with the F1.2 lens being slightly longer than 85mm at infinity, it's easily enough to account for the differences various reviewers have noticed between the two lenses. It would produce approximately a 3mm difference in perceived focal length between focusing at infinity and focusing at 3.5m....

Click to expand...

Remember, focus breathing is a change in the Angle of View, not a change in the focal length. In fact, you can't eliminate focus breathing without changing focal length.

I don't have a clue about physical backgrounds , but if focus breathing is defined as change of field of view and you measured only very little change in your test bench, how does that fit to the video I linked earlier where the viewing angle seems to change a few degree?

Is it due to the range being observed?

Sittatunga said:

Alastair Norcross said:

bclaff said:

Richard Butler said:

...My testing was conducted with a chart 1m tall, giving a working distance of around 3.5m. ...

The 85mm F1.2 L USM gives a tighter field of view. It shows a roughly 27 degree diagonal AoV, equating to a roughly 90mm lens at the ~3.5m working distance. However, it exhibits A LOT of focus breathing, such that it's entirely likely that it would match the 87mm focal length given in the patent for its lens formula, if focused at infinity.

...

Click to expand...

I'm not seeing any dramatic focus breathing.
Note that there is a Focus button and slider in the PhotonsToPhotos Optical Bench.
At ~3.5m it looks like this:

That 27.06 degree Angle of View is pretty close to the 27.94 degrees at infinity.

Regards

Click to expand...

The focus breathing might not be dramatic, but, when combined with the F1.2 lens being slightly longer than 85mm at infinity, it's easily enough to account for the differences various reviewers have noticed between the two lenses. It would produce approximately a 3mm difference in perceived focal length between focusing at infinity and focusing at 3.5m. If the 85 F1.2 is 87mm at infinity, it would look like 90mm at 3.5m. If the 85 F1.4 is, as Richard says, around 85 at infinity, and at 3.5m, you get a perceived difference of around 5mm at 3.5m, which is in line with what a lot of reviewers are claiming to see. The main point that Richard is making is that the VCM lens is, in fact, pretty much what it claims to be-- an 85mm lens. It's interesting that so many people, on initially seeing the reviews, immediately jumped to the conclusion that Canon was somehow cheating by putting out an 80mm lens as an 85mm lens. Some even declared on this forum that they were no longer interested in the lens, because of this terrible discrepancy. Whereas the truth is that the VCM lens is both more accurately labelled than the F1.2 lens, and has the advantage of much less focus breathing. No good deed, eh?

Click to expand...

A lot of this argument is due to badly defined terms. If you expect a lens to follow the simple mathematical thin lens equation 1/dₒ + 1/di = 1/f then, as the object distance dₒ decreases, the image distance di has to increase, as does the magnification m = dₒ/di. It's di that governs the field of view on the sensor, not f , though at dₒ = infinity di = f. Also, if you think in terms of di being f plus a focussing rack or a virtual (imaginary) extension tube, di = f(1 + m) which is a quick way to estimate the amount of focus breathing a simple lens will give.

However, if you expect no focus breathing as a lens focusses, di has to remain constant, so f has to decrease with dₒ for the thin lens equation to work. In my view focal length is only a very approximate indicator of field of view once you start compensating for distortion and focussed distance.

Click to expand...

All of this is why I stressed 'perceived' focal length. We all know what people are trying to say, and there's no need to get overly technical.

bclaff said:

Alastair Norcross said:

bclaff said:

...

Click to expand...

The focus breathing might not be dramatic, but, when combined with the F1.2 lens being slightly longer than 85mm at infinity, it's easily enough to account for the differences various reviewers have noticed between the two lenses. It would produce approximately a 3mm difference in perceived focal length between focusing at infinity and focusing at 3.5m....

Click to expand...

Remember, focus breathing is a change in the Angle of View, not a change in the focal length. In fact, you can't eliminate focus breathing without changing focal length.

Click to expand...

Of course I know this. That is why I stressed 'perceived' focal length. We all know what people are talking about. There's really no need to go into lots of technical details. The bottom line is that the people who jumped to the conclusion that the VCM lens is actually a lot shorter than other 85mm (claimed) lenses, and a lot shorter than 85mm, were simply mistaken. They were misled by looking at angles of view at fairly short distances. It was obvious that at least some people were just looking for something, anything, to criticize Canon for. And it turns out that the thing they were jumping on as a criticism is actually a fine achievement of lens design (minimal focus breathing).

bclaff said:

You're using think lens equations of a real world lens. Thick equations might be more appropriate.

Click to expand...

I know. It would have been quicker to say that, once you get away from infinity focus, statements relating focal length to angle of view (particularly for internally focussing lenses) are wide open to interpretation. And, as you replied to Alastair, you can't eliminate focus breathing without changing focal length.

Laqup said:

I don't have a clue about physical backgrounds , but if focus breathing is defined as change of field of view and you measured only very little change in your test bench, how does that fit to the video I linked earlier where the viewing angle seems to change a few degree?

Is it due to the range being observed?

Click to expand...

Haven't had time to view the video but if it's a well constructed test then it doesn't matter what the Optical Bench says