Hass lens - contrast or dynamic range query

[Sim2]

Hi there,

Only recently got back into medium format and using a Hasselblad 150mm F lens, so may be a bit of a "muppet" question but....

Have got used to the replication of the dynamic range of a scene with my 35mm Canon lenses, using the same lighting set-up, film and dev on similar scenes with the Hass lens I am finding a greater dynamic range i.e. more detail in the shadows that previously would be blocked up black and tone in brightest areas that might have been "clipped" white on the 35mm.

Question: Does the Canon lenses have a smaller dynamic range to boost contrast and apparent sharpness, or do the Hasselblad lenses have an inherently larger dynamic range than smaller format lenses. Is this a function of a larger neg size?

Thoughts/explanations welcomed!

Sim2.
*Having a theoretical moment*

 

[ic-racer]

The MTF curves for the Zeiss lenses are available at the Carl Zeiss site for you to check. Not sure where the Canon data would be.

 

[wiltw]

Same scene, more grains or more color clouds which represent the same amount of subject area, due solely to the size of image captured.

 

[edtbjon]

MF lenses are less sharp and less contrasty than 35mm lenses. That is easy to check at the CZ site. (Digi FX size lenses are even sharper than 35mm lenses and LF lenses are less sharp than MF lenses. This leads to the interesting fact that those guys who makes converters to use 'blad lenses on 35mm cameras spends a lot of money and effort to gain nothing.)

//Björn

 

[Sim2]

Have seen the MTF charts for Zeiss and for Canon but they might not be the easiest of things to understand for the non-scientist! Suffice to say that the Hass lenses do not go as far up the graph as the Canon lenses - but that was to be expected. I have understood that the smaller the format/neg the better resolution a lens has to achieve to match the sharpness of a larger format lens. Not certain but think this might be a function of lens design/optimization.

In the pursuit of the required resolution for a small format is the reduced contrast/dynamic range a function of that pursuit?

 

[keithwms]

Sim2, even if the 35mm and the MF lens have the same contrast, the MF kit will win in terms of tonality.

When I say tonality, I mean transitions between tones, and that includes the transitions into deep shadows or into bright highlights. The more info you capture, the more smooth and subtle those transitions will be.

One way to think of it is this: suppose for simplicity that the film is recording only white and black dots, i.e. a bit-mapped image. The more dots you have, the greater number of greys you will be able to render. Regardless of the number of dots, the pure white and pure black tones will be rendered equally; but: the transitions depend a lot on how many dots you have. Too few and you get posterization.

So... what you are seeing may not have much to do with contrast- it probably has more to do with the advantage that you get from casting the same field of view (I assume) over a much bigger piece of film. Transitions will simply be rendered more smoothly when you do this, so whereas in 35mm a transition might look quite sharp in the image, in MF/LF it will tend to be smoother. (N.b. this can erroneously be interpreted as less sharpness.... rest assured the amount of info you get in MF LF is so vast that sharpness is very rarely an issue).

Remember also that if you want more or less contrast, you can always develop and print accordingly.

By the way, Björn, the mamiya 6/7/7ii lenses are as sharp and contrasty as the top of the line 35mm lenses. Really. 120 lp/mm almost all the way across the frame in many cases. But I agree that shooting MF lenses on 35mm gear is kind of silly unless you are using a shift adapter or such.

 

[Q.G.]

Have seen the MTF charts for Zeiss and for Canon but they might not be the easiest of things to understand for the non-scientist! Suffice to say that the Hass lenses do not go as far up the graph as the Canon lenses - but that was to be expected. I have understood that the smaller the format/neg the better resolution a lens has to achieve to match the sharpness of a larger format lens. Not certain but think this might be a function of lens design/optimization.

In the pursuit of the required resolution for a small format is the reduced contrast/dynamic range a function of that pursuit?

No.

It's not a function of lens design. Merely of format.
But it is indeed true that, to cram the same amount of detail in a smaller frame, that detail needs to be packed more densely. So 35 mm lenses need to be able to produce higher resolution images than lenses made for larger formats to match the results.

That (contrary to what Björn may seen to suggest) is not necessarily also the case (i can't see anything saying something to that effect on Zeiss' site either).
But sometimes, it indeed is.

 

[Sim2]

Yes, the relationship between tonality or gradations of tones and format size is understood, my current observation is that something that is pure white (burnt out) on 35mm has tone in MF and something that is pure black in 35mm has tone in MF.
I do not think this is the tonal changes, but the range of tones (dynamic range) that are delivered by the lens to the film.
Perhaps best I stop now, as it appears I am unable to phrase the question adequately for my query to be understood.

 

[ic-racer]

I'm still not sure where the Canon data is but to do a comparison you need to define what you are comparing, such as:

Same same focal length comparison vs same angle of view comparison (So, 50mm vs 50mm or 50mm vs 80mm?)
Same Aperture comparison vs same DOF (pupil diameter) comparison. (So, F8 vs F8 or F8 vs F11?)
Same field of view vs disparate field of view (that is are you only going to compare up to the 42mm circle? Or go all the way out to 80mm with the MF lenses and look at distance from the center as a percent?)

The 'winner' will be different depending on how you compare them.

 

[Sparky]

Question: Does the Canon lenses have a smaller dynamic range to boost contrast and apparent sharpness, or do the Hasselblad lenses have an inherently larger dynamic range than smaller format lenses. Is this a function of a larger neg size?

It's a strangely worded question - but, well- lenses - all lenses - have INFINITE dynamic range. Exposure determines that. It's the FILM that you have to worry about. As much of a critic I am of it, it would be a good thing for you to study the basic principles of the zone system - and you can understand how to select your wanted dynamic range and expose and process your film to capture it.

 

[Sparky]

Yes, the relationship between tonality or gradations of tones and format size is understood, my current observation is that something that is pure white (burnt out) on 35mm has tone in MF and something that is pure black in 35mm has tone in MF.
I do not think this is the tonal changes, but the range of tones (dynamic range) that are delivered by the lens to the film.
Perhaps best I stop now, as it appears I am unable to phrase the question adequately for my query to be understood.

I think you're thinking of 'digital' phraseology here... pure white tonality is determined by your printing technique/exposure/development etc. You can print any 'tone' on film to any other tone in a print... it's totally up to you.

 

[keithwms]

Lenses - all lenses - have INFINITE dynamic range.

That statement confuses me.

I imagine a subject brightness range, and that range gets passed through a lens, and the resulting range (before any film or recording medium is involved) need not necessarily have the same brightness range.

Let us take the case of a non-neutral lens. How about an extreme case, i.e. an old lens with some yellowing. That lens will not transmit the subject brightness range as completely and as such will have lower contrast, simply because there is no pure white. What is white in reality will not be white on the film side of the lens.

A less extreme case would be any lens that doesn't have correction; it will not weight the colours equally and hence will not depict white (=an equal combination all wavelengths) as it should.

A perfect lens would, I guess, do what you say: it'd have perfect contrast and perfectly neutral transmission. I am thinking that only an apo-corrected lens comes close to this ideal.

Another way to put it: no lens is perfect... all lenses have MTF that is imperfect and only approach perfection under certain conditions (e.g. frame center, f/8). Given the real-world MTFs, I don't see how to say that they 100% faithfully transmit the subject brightness range.

...or am I missing something?

 

[Sparky]

Well- I'd say you're looking at the picture from a completely different perspective than, let's say, traditional film photographers. As you may be aware closing down an aperture lessens the transmission of light through a lens - but does it limit the highlight we can achieve on film? NO! We just give it a longer exposure. Likewise - ANY lens can capture light from, say EV -5 or so up to an infinite band of brightnesses, given that it's shutter can control the overall exposure to compensate.

One may get film to capture (reliably) an extra 5-10 stops (log exposure) of brightness in the subject the lens is pointed at, simply by reducing development time and increasing exposure to compensate (google H+D log curves to see the mechanism of what I'm talking about). This is something called 'pulling' film. One can actually go a fair bit more extreme than this - but the results tend to look rather 'mushy' since we're really not used to seeing reality portrayed over such an extreme of brightnesses compressed into a fairly small scale in terms of print reflectance values (hope i'm not talking too 'nerd' here). Anyway - the point I'm trying to make here is that lenses (any lens), through time of exposure, is able to capture any level of brightness, and concentration of photons (if you like) and therefore is capable of capturing any RANGE of those brightnesses. FILM is the flexible factor here.

Now - this is a very different model from the equivalent in the digital world, where the CCD (or whatever sensor) can only capture a FIXED range of brightnesses falling upon it. The lens is used as the sole modifier of brightness RANGE. In THIS case, your dynamic range or SBR as interpreted BY the lens, is only controlled BY that lens. And yes, well, due to various design factors, some lenses create more or less flare (which affects shadow depth, etc) and you are much more dependent on lens design for this purpose. Additionally - different lenses also have different transmissivity as it's called - in the film industry, it's a common practice to calibrate a lens taking this into account (resulting in T-stops in place of F-stops). This has nothing to do with contrast but purely how much light is blocked out passing through the glasses in the lens assembly itself.

There is SOME truth in where your coming from. I'm not negating that. All other things being equal (always developing for a given 'normal' target) some lenses will create more negative contrast than others... but controls you can get through development are radically more powerful than having to rely on a given lens for this purpose. This is why you'll find little or no conversation on these boards about using different lenses for the purpose of contrast control. And tons of threads on development techniques for the same purpose.

 

[Q.G.]

...or am I missing something?

Of course not, no.

Given a scene with a given brightness range, different lenses behave differently.
Most will have no problem with the absolute range, given that the bright and dark patches are large enough, i.e. the changes from dark to light bits has a low spatial frequency.

But the smaller the detail, the greater the chance that it will be lost.
Tiny specular highlights against a dark background will have a greater chance of not being 'flattened' then little dark spots in a bright background.

Which takes it back to being a format thing. Larger formats use longer lenses, because they have to project images with a greater magnification to fill the larger format with the same angle of view.
Tiny detail will therefore be not so tiny as in the image a lens on a smaller format has to project, and thus have a greater chance to survive going through the lens.

So even assuming a lens on a smaller format with better MTF behaviour at high spatial frequencies, the larger format and an assumed lens with less good MTF behaviour will still capture the same detail.
It's what has been said before: 35 mm format lenses have to be better than MF lenses just to keep up. (And they by no means always are.)

 

[Sparky]

Here is a typical model for understanding film exposure. You'll see that the ideal that you're referring to is thrown out the window as soon as the light hits the film. The film and paper interpret light in their own very special (and very strange) way. The slope of the middle of the curve determines your effective contrast (speaking relatively) and you'll see that more stops of light can be captured at reduced contrast. Likewise - less are captured with a steeper slope, resulting from greater development times. Most photographic lenses are designed to produce the best result within the purview of these characteristics. It is a strange irony (but makes some sense technologically) that digital sensors are designed to replicate the response of film as closely as possible. This was probably due to the preponderance of 'traditional' lenses already on the market. I strongly suspect that as digital 'evolves' - we will be less and less able to use traditional lenses with that medium - as the quest for 'higher performance' sensors places more unique demands on the glass sitting in front of it... much as the design of a digital body and sensor aspect ratio is based on the traditions established by the 35mm shooter.... when in fact it is not terribly appropriate.

 

[Q.G.]

You can only have one film response per scene (per film even) though.

So image, if you will, one scene, one film, one 'processing', but (by some magic) two frames on it, one exposed through an MF lens on MF format, the other through an 35 mm format lens on 35 mm format.

What, would you say, will they both show? How, if at all, will they differ? And why?

 

[Sparky]

I think that the differences in manufacturing with given equipment would preclude you from drawing any rational conclusions. I was only trying to explain what I meant by lenses being able to give an infinite dynamic range. But I have found in my experience that I need to scale back my dev times with 35mm MORE than with MF to give an equivalent contrast result. I'd suspect that most such results are due to inadequate lens shading with medium format. I also suspect that the retrofocus design of most 35mm lenses results in more 'innate' shading against flare with wide angle designs...

 

[keithwms]

Well- I'd say you're looking at the picture from a completely different perspective than, let's say, traditional film photographers.

No, that is precisely the perspective from which I am approaching this. I mention pure white and pure black tones just to simplify the argument.

Here is a typical model for understanding film exposure.

The curve is moot. We were talking about what lenses do, not what film or paper does. Anyway... it's a technical point, but, I still do not think any real lens records all of the contrast in the scene.

~~~

The real question here was about transitions, and for those, the larger format will win. Smoother tonal transitions.

 

[Sparky]

no no... sorry - you misunderstand me. Lenses record nothing. Film records. I'm just trying to say (with the curve and all) that film has it's own unique interpretation of the light that hits it. Lenses only focus. Film expresses that. if you had film that could record 200 EVs of information - ANY lens could convey that. I'm just trying to say the lens has no role in determining the range of stops that get recorded onto film.

 

[wclark5179]

This is an interesting discussion.

I've got a question, perhaps this isn't the spot to ask, but it's on my mind. Does the distance from the back of the lens to the film/sensor have a bearing on contrast/sharpness? I've heard it said that Rangefinder cameras have an advantage because of the distance between the back of the lens to the film/sensor plane is less than SLR cameras. Is this so?

Thanks! Have a great week.

 

[keithwms]

no no... sorry - you misunderstand me. Lenses record nothing. Film records.

Ok. We are making progress understanding each other

I'm just trying to say the lens has no role in determining the range of stops that get recorded onto film.

I'm still not agreeing with that. But the key to resolving our little dispute is that it may make little sense to think of the lens and the film independently. Maybe we should speak only of system resolution, to keep things simple.

Along the lines of what Q.G. was saying (and I was trying to say previously but apparently not with enough clarity), consider the following thought experiment:

Suppose that you photograph the lace of a dress from some distance. Suppose that the lace is bright white and the underlying cloth is black.

The camera with less system resolution will tend to average the two regions. The fine highlights and the neighboring blacks will tend to be averaged together to produce greys. Hence less brightness range is recorded. A TTL metering system will see grey and make an exposure recommendation based on that.

The camera with more system resolution will do a better job distinguishing the two regions, hence more brightness range is recorded.

I can't think of a way out of this effect, so again I assert, format size certainly does matter when it comes to tonal transitions.

 

[keithwms]

Does the distance from the back of the lens to the film/sensor have a bearing on contrast/sharpness?

Short answer: it used to. Designing a lens to provide extra space for a mirror used to be a real issue.

Longer answer....

The first thing to understand is that no lens is perfect. There are quite a few "corrections" that need to be made to a basic lens design in order to get optimal sharpness and neutral tone.

Google "retrofocus lens" and you will find an intro to the optics involved in designing a lens formula. "Formula" just means design. Behold the complexity of the modern lens formulas. There are many kinds of lens formulas and some are easier to correct for things like astigmatism, chromatic aberration, and other issues.

Prior to the use of computerized formulas and modern glass to produce superduper hyper ultra low dispersion, nanocoated, 10 group / 30 element, apochromatic aspherical, oh my God lenses, RFs did have a substantial advantage: the lens formula was simpler and hence it was easier to correct the optical flaws. There are less variables to juggle in the simpler design.

That advantage is mostly gone now... the lens formulas are computed and lens designers have access to a number of fancy, low dispersion glasses. With these tools they can accurately model the performance of a lens in computero across the full wavelength range... before it is ever built.

 

[ruilourosa]

well well


this discussion is really a bummer!

talking about non-retrofocus or retrofocus,35 or MF, seems resolution isn´t all in life, orelse everybody would pursuit it avidly and meaninglessly, tonality, distorsion, flare resistance, bokeh, weight, practicality, and personal taste.

Biogon and super angulon´s and grandagons ARE great designs and a 50 years old lens is way good! they are non retrofocus

Distagons with 50 years are OK...

Holgas have a non retrofocus desing and they are really bad in resolution!


shoting with 35mm is nice, but you have to live with some things
the same with MF
the same with LF
the same with DIGITAL
the same with Alternative Negative Processes
the same with art in general



everybody knows that if you want more resolution and better tonality and less grain you should move up in format


but what about doing what suits you, i like shooting in MF, also in 35 and also in LF

Digital is a problem, i have no patience for another black box


just photograph...

 

[Q.G.]

... and stay ignorant?

What's wrong in trying to understand something?
Especially when the question arose when doing the practical bit you advocate, i.e. it's a practical question?


And one more, perhaps most important question now: why join a forum when all you want to do is photograph, and not talk about photography?

 

[Q.G.]

I've got a question, perhaps this isn't the spot to ask, but it's on my mind. Does the distance from the back of the lens to the film/sensor have a bearing on contrast/sharpness? I've heard it said that Rangefinder cameras have an advantage because of the distance between the back of the lens to the film/sensor plane is less than SLR cameras. Is this so?

Apart from what Keith already said about this, there's one more thing.
Light fall-off.

Illumination levels, and with it contrast, drops off more rapidly (going from the center to the edges and corners of the frame) when the lens is closer to the film.
In this respect, rangefinder cameras and their lenses have it more difficult, compared to SLRs and their retrofocus lenses.