I just don't get the 35mm vs bigger format thing.

[markbarendt]

yes, it would be dead obvious using tri x and hp5+ seeing they are the grainiest films available.
what i have suggested about the differences between formats, for me, was true using low / no grained films
tabular grained films, like tmy+tmx, pan-x and tech pan. but then again at a good "viewing distance"
35mm pinhole images shot on 3200 film can be enlarged to 32x40 ... and look perfect.
its all about what you want, what you like and what you do ...

Yep, under the right conditions, with the right materials in play, shot, processed, and printed well, not printed too big, not viewed too close, ..., it is hard to see a difference between formats.

Reaching that standard is doable.

What it is not is a direct comparison of format to format in a scientific sense; it's a comparison of system to system in a practical sense, and that's ok.

And yep, it boils down to what we can live with vs the hoops we are willing to jump through and time we are willing to spend and what we like.

 

[removed account4]

So I have real problems with an assertion that states enlargements from 8mm motion picture and 16x20 inch (or any other size) large format negatives are indistinguishable. They could only be indistinguishable if the negatives that produced them contained the exact same information. And all other things being equal, they simply cannot.
Ken

what i have a real problem with is people purposefully misquoting others to mislead ...

maybe you should read what i said ?

he enlarged 8mm film to 16x20 and had it on display at some sort of party people from the index were having
or whatever ... ansel adams was there looking at the print and thought it was made with a large format camera....
jerry laughed from what i remember showed him the film ... it was smaller than 110 film ....

i ( and he ) never said indistinguishable. and it wouldn't be hard at all to hang a 16x20 from 8mm film
and view it from a certain distance and think it was made by a large format camera.
since you seem to be hellbent on discrediting the now deceased chemist ( and me ),
maybe you should contact morgan morgan publishing and speak with mr morgan and ask if he remembers the party
and you can ask for the chemist's credentials.


i said enlargements from 35mm-4x5-120 film can produce similar looking images ..
i have a portfolio of portraits made with the 3 formats, all printed a bit bigger than 8x10 on 11x14 paper.
the difference in format is not noticeable.
but then again, i was told i was doing it all wrong, and i am pretty much a liar, both in the same thread.
so maybe not ...

 

[flavio81]

Optically speaking, aberrations scale directly with focal length and/or aperture (depending on how you look at it). For example, scaling up the design for a 50mm f/2 focal length lens to 250mm f/2 focal length lens will increase the aberrations, blur size, etc. all by exactly 5

This is misleading.

A 50mm f2 lens, in 35mm format, has an angle of view and depth of field comparable to a 100mm f4 or even 100mm f5.6 lens in 6x7 format. A 100mm f4 lens of excellent correction for the 6x7 format is no big deal.

Moreover probably that 100/4 lens, wide open, will give an end image far sharper and better resolved than the 50/2 wide open; using their respective formats.

Furthermore, there are some tests on the net that dispel the myth that larger format lenses are inferior performing. The best MF lenses are as good as the corresponding 35mm lenses.

 

[frank]

what i have a real problem with is people purposefully misquoting others to mislead ...

maybe you should read what i said ?




i ( and he ) never said indistinguishable. and it wouldn't be hard at all to hang a 16x20 from 8mm film
and view it from a certain distance and think it was made by a large format camera.
since you seem to be hellbent on discrediting the now deceased chemist ( and me ),
maybe you should contact morgan morgan publishing and speak with mr morgan and ask if he remembers the party
and you can ask for the chemist's credentials.


i said enlargements from 35mm-4x5-120 film can produce similar looking images ..
i have a portfolio of portraits made with the 3 formats, all printed a bit bigger than 8x10 on 11x14 paper.
the difference in format is not noticeable.
but then again, i was told i was doing it all wrong, and i am pretty much a liar, both in the same thread.
so maybe not ...

John, in this post you are escalating the language unnecessarily to a personal level.

We're trying to find out how, in your photos from different formats (are you using the same film for each?) you are not seeing a technical advantage in the larger formats, because I and others do see a very noticeable difference between 135 and larger formats.

You're being a little vague/contradictory when you say both:
1) I never said indistinguishable
2) the difference is not noticeable

And you introduce viewing distance as another variable when viewing 16x20 print from 8mm film.

 

[Old-N-Feeble]

I like to pixel-peep/grain-peep very large prints so minimal grain and maximum detail are very important to me. When I move in close for an intimate look at detail I want to feel like I'm 'there'.

 

[Nodda Duma]

This is misleading.

A 50mm f2 lens, in 35mm format, has an angle of view and depth of field comparable to a 100mm f4 or even 100mm f5.6 lens in 6x7 format. A 100mm f4 lens of excellent correction for the 6x7 format is no big deal.

Moreover probably that 100/4 lens, wide open, will give an end image far sharper and better resolved than the 50/2 wide open; using their respective formats.

Furthermore, there are some tests on the net that dispel the myth that larger format lenses are inferior performing. The best MF lenses are as good as the corresponding 35mm lenses.

I'm sorry but it's not misleading. Maybe there's a misunderstanding of what it means to scale a lens.

A scaled 50mm f/2 would be a 100mm f/2 in your example. Or conversely, compare the 100mm f/4 to 50mm f/4. Not only does the focal length double but the aperture doubles, maintains the same f/#.

In fact, in lens design references, the prescriptions are typically all scaled to 100mm focal length for more direct comparison of aberrations and performance.

There's nothing I posted that should be construed to say anything about large format lenses being inferior performers. The optics will do exactly what they were designed to do. Also, be careful of what you think those MTF plots are telling you. They are not apples-to-apples comparisons if the f/#'s are different.

 

[Ken Nadvornick]

Yep, under the right conditions, with the right materials in play, shot, processed, and printed well, not printed too big, not viewed too close, ..., it is hard to see a difference between formats.

Reaching that standard is doable.

What it is not is a direct comparison of format to format in a scientific sense; it's a comparison of system to system in a practical sense, and that's ok.

And yep, it boils down to what we can live with vs the hoops we are willing to jump through and time we are willing to spend and what we like.

Absolutely, yes. No issues with this summary at all. As the man said, one is entitled to one's own opinions, but one is not entitled to one's own facts.

When one alleges fact, credibility is won or lost by comparison to other established fact. When one alleges opinion, credibility is established by comparison to the reputation of the individual alleging that opinion.

So simply go look at the pictures in question. It will become immediately apparent if the alleged opinion works for you, and thus the opinion has merit for you, or it does not.

If many are in agreement with the alleged opinion, the individual offering it may in time accrue sufficient credibility to acquire the reputation of being an expert. If few are in agreement, well, we don't need to explicitly go there...

But in both cases the accrual or denial of credibility rests beyond the control of the individual doing the alleging. It can come only externally from those doing the judging.

This, of course, is the process that happens every time a creative work is published for public consideration.

Ken

 

[Ken Nadvornick]

"All other things being equal" can't be applied to this situation. About all that can be said with certainty for a given print size and film type is that graininess is reduced as film size increases.

Yes, I do realize this. And in cases where it does prove impractical or impossible to factor out all of the other variables, then meaningful comparisons simply cannot be made. If in such cases one persists in alleging meaningful comparisons, then credibility suffers even further.



Ken

 

[tomfrh]

he enlarged 8mm film to 16x20 and had it on display at some sort of party people from the index were having
or whatever ... ansel adams was there looking at the print and thought it was made with a large format camera

Damn! Gotta get me some of that 8mm stuff. Can't believe I've been lugging 35mm around all my life for no reason...

 

[Nodda Duma]

I have some free time now that I'm on lunch break, so I decided to set up a model in ZEMAX to maybe better show how aberrations scale.

One comment before I begin: ZEMAX is sort of the de-facto industry standard optical design program, used by the great majority of optical designers (and non-optical designers) in the world today.. and has been that way for the past 10 years at least. So it's a fair bet that any modern camera lens designed over the past decade was designed using ZEMAX (and the experience of the designer as well). One well-known example that I'm aware of is that the '03 Mars Rover optics were designed in ZEMAX by Greg H. Smith. He wrote a white paper available in the public domain: http://lkellogg.vttoth.com/LarryRussellKellogg/SPIE_paper.pdf Just in case you are wondering how accurate a design program can be... it is good enough to send optics to another planet!



I use a basic 50mm f/1.8 Double Gauss design to demonstrate the principle, but I'll stop it down to f/5.6 since practical large format lenses start at that f/#. I think this design is actually the Olympus OM-series 50mm f/1.8 patent prescription, if you're curious where it came from, but it is one of the designs in Zemax's (and Code V too I think) optical database.

The first output shows the prescription, layout, geometric spot diagram, and the MTF up to 33 lp/mm. Hopefully it's high enough resolution to read when I upload it, but if not PM me and I can email you the original. Note the effective focal length (EFFL), f/# (ISFN), and the entrance pupil diameter (ENPD) shown at bottom. 51.21 mm / 5.6 = 9.144 mm.

The spot diagram indicates the blur size you'd get if you imaged a point source such as a star. The colored lines on the MTF plot correspond to the field angles shown in the layout. Both are used as metrics for optical design performance, along with ray fans and OPD plots. Blue = center, green 14 mm off-center (on the film), and red = 21.6 mm off center (about the corner of the film).

Now I scale up the design to "fit" an 8x10 format. For fair comparison, I want to keep the same aspect ratio, so I scale for a 6.85" x 10" film plane. So I have to scale the lens to a focal length of 51.21 * (254/35) = 371.6 mm. This gives the same FOV for both lenses. The scaling factor is about 7.25.

Here's the scaled design. Note the f/# remains the same (the aperture scaled: f/# = focal length / aperture diameter).

Pay attention to the values circled... particularly the GEO radius (geometric radius) in the spot diagram. The spot radius values scaled by the same 7.25 factor. Also look at the MTF.

Now if you enlarge and make a print from the 35mm film by 7.25x, then compare to a contact print from the larger format, the blurs will all look the same... the only difference is from the effect of the film grain itself (assuming you did all the development the same of course).

Aberrations scale with focal length (for same f/#).

 

[removed account4]

John, in this post you are escalating the language unnecessarily to a personal level.

We're trying to find out how, in your photos from different formats (are you using the same film for each?) you are not seeing a technical advantage in the larger formats, because I and others do see a very noticeable difference between 135 and larger formats.

You're being a little vague/contradictory when you say both:
1) I never said indistinguishable
2) the difference is not noticeable

And you introduce viewing distance as another variable when viewing 16x20 print from 8mm film.

indisnguishable to me means exactly the same as no difference
not noticable to me means there may be a difference but it isn't large enough to make any difference.
with regards to the portraits - location work, some with available light some with lumedyne flash all processed
conventionally either by shuffling or in hand tanks with 1 min / 10sec/min afterwards agitation.
the if i handed you a print you would not be able to tell me what format made it, even if you stuck your nose to the print/s/

with regards to viewing distance, search nearly ever post in have made here in 12 years, i have never championed prints to
be looked at nose to image ( nearly because i like jewel prints since you need to really look at them up close )
i was not savvy enough to ask the chemist how far away mr adams was when he looked at his print
so who knows if he was up close or at a distance. i am not about to make up some BS just for the sake of backing HIM up.
be my guest, dont' believe his story. and also be my guest and buy big cameras, they are cheap, except for the film + lens part.

nothing i have said contradicts what i have said before ...

good luck with whatever format you like to use

 

[frank]

To nodda:
So, are you saying that larger formats are not capable of rendering finer detail than smaller formats because the blur sizes are the same?

 

[Nodda Duma]

To nodda:
So, are you saying that larger formats are not capable of rendering finer detail than smaller formats because the blur sizes are the same?

I'm saying scale up a lens design and the aberrations and blur sizes will scale with it. Nothing more, nothing less. Another way I'd be ok with saying it is that the same information is available at the image plane for both optics I describe above. What the *film* does with that information is something else entirely.

 

[frank]

I'm saying scale up a lens design and the aberrations and blur sizes will scale with it. Nothing more, nothing less. Another way I'd be ok with saying it is that the same information is available at the image plane for both optics I describe above. What the *film* does with that information is something else entirely.

But how does this apply to the thread topic? Connect the dots for me because my mind is swimming a bit.

 

[removed account4]

I'm saying scale up a lens design and the aberrations and blur sizes will scale with it. Nothing more, nothing less. Another way I'd be ok with saying it is that the same information is available at the image plane for both optics I describe above. What the *film* does with that information is something else entirely.

this makes perfect sense but then again i am bordering incredulous ... :munch:

 

[Nodda Duma]

But how does this apply to the thread topic? Connect the dots for me because my mind is swimming a bit.

I've found that a brace of beers really clarifies the mind when waxing philosophical

Several posts indicated thinking that better or worse detail in prints made from different formats is partly due to the optics. My post models what happens when a fair comparison is made between a large format lens and its small format equivalent, and shows that's not the case. The model shows that the same information is available to both formats, and it's up to the film to accurately record that information or lose some of the information to the grain.

In the real world, it's hard to compare because optics are rarely scaled directly. So a real world test would still involve variables that muddy the waters. However, if you have the right lenses I think you could get close to demonstrating the principle that the modeling shows.

 

[RobC]

Just in case you are wondering how accurate a design program can be... it is good enough to send optics to another planet!

Did it design the hubble mirror?

 

[Nodda Duma]

Did it design the hubble mirror?

The software doesn't design anything. It is simply a tool used by the person who does the design work so that a prescription can be developed in weeks instead of years.

The cause for the Hubble issue had nothing to do with the prescription. The cause was a flawed test setup used to control polishing coupled with ignoring independent test results and poor program management.

Here is a good read:

http://www.ssl.berkeley.edu/~mlampton/AllenReportHST.pdf

Hubble was designed long before the modern optical software packages were introduced. However, the Hubble prescription complete with corrective optics is provided with the software (along with several other basic designs). The corrective optics - COSTAR - was designed using Code V (the other commonly used optical design tool).

 

[RobC]

The software doesn't design anything. It is simply a tool used by the person who does the design work so that a prescription can be developed in weeks instead of years.

The cause for the Hubble issue had nothing to do with the prescription. The cause was a flawed test setup used to control polishing coupled with ignoring independent test results and poor program management.

Here is a good read:

http://www.ssl.berkeley.edu/~mlampton/AllenReportHST.pdf

Hubble was designed long before the modern optical software packages were introduced. However, the Hubble prescription complete with corrective optics is provided with the software (along with several other basic designs).

I was joking. Maybe I should have used a smiley.

But your answer indicates that the fact you have some software is irrelavant to the final outcome so I don't know why you mentioned it in the first place and used it as a yardstick for good lens design.

 

[Ken Nadvornick]

Several posts indicated thinking that better or worse detail in prints made from different formats is partly due to the optics. My post models what happens when a fair comparison is made between a large format lens and its small format equivalent, and shows that's not the case. The model shows that the same information is available to both formats, and it's up to the film to accurately record that information or lose some of the information to the grain.

So to clarify, if the same information is available at the film plane of two different formats (via your example of scaling), then a larger capacity container to hold that available information will, in general, preserve a greater amount of it than will a smaller capacity container, where more of it will be lost to the grain, all other things being equal.

Is that a fair reading?

Ken

 

[Nodda Duma]

I was joking. Maybe I should have used a smiley.

But your answer indicates that the fact you have some software is irrelavant to the final outcome so I don't know why you mentioned it in the first place and used it as a yardstick for good lens design.

I figured you were joking but it did deserve a good answer. And I must admit I tend to geek out on lens design..which is good because that's my chosen profession. I mentioned the yardstick because as a designer who learned my trade in "the real world", I'm always slightly skeptical of computer models until they've been validated by real-world performance. I was convinced long ago of this for Zemax. The Mars Rover design white paper I mentioned above is a good "proof is in the pudding" for skeptics... Images taken with those optics are in the public domain.

 

[rbultman]

Another way I'd be ok with saying it is that the same information is available at the image plane for both optics I describe above. What the *film* does with that information is something else entirely.

But the image plane is larger for larger formats, right? This means that for the same grain density on the film, the larger format has more grains on which to capture the information available on the film plane due to the projected image being larger.

To put it another way, when you enlarge, you are effectively enlarging the grain. If you have to enlarge less for a given print size, you are enlarging the grain less, i.e. the enlarged grains are smaller for a larger format than a smaller format.

If you had a film format that was so tiny that the film plane only contained a single grain, would the resulting print be grainy? Or, would there not be enough information in that single grain to even form a recognizable image on paper? Probably not.

At some point, the grain density is high enough that it does not matter anymore. Either your eye cannot discern the difference or the enlarging paper grain density is less than the projected grain density from the negative. As others have said, viewing distance matters too. I gotta believe that, at least theoretically, print quality should improve (less grain, smoother tones) as negative size increases.

Would the following help put this to rest? Perhaps someone here has an EOS film camera, Mamiya 645, an EOS body to Mamiya lens adapter, and suitable Mamiya lens. The rolls could be developed at the same time to remove processing as a variable. I assume the aperture settings would be the same but the lens-to-subject distance would need to be different in order to achieve similar images on the film plane. Make 8x10, 11x14, and 16x20 prints of each.
Regards,
Rob

 

[Nodda Duma]

So to clarify, if the same information is available at the film plane of two different formats (via your example of scaling), then a larger capacity container to hold that available information will, in general, preserve a greater amount of it than will a smaller capacity container, where more of it will be lost to the grain, all other things being equal.

Is that a fair reading?

Ken

Yep, absolutely.

I think what ends up happening is that. the grain size limits the resolution of smaller formats, and the MTF of the optics limits the resolution for larger formats. Stop down the lenses and I think it's still the same: the MTF of both lenses keeps getting better and better until you hit diffraction limit--but it is still the limiting factor for the larger format and not the grain. For the smaller format, the grain size still limits the resolution.

The result (and I think this is the truth we're looking for): For typical film the larger format print will blow away the smaller format because of the limits of the grain. Use a film where the grain doesn't limit the performance of the 35mm lens, and the 35mm optic will give you prints as good as anything.

Ok back to work for me.

Edit: ah yes rbultman you said it much better than I could.

 

[Thomas Bertilsson]

I guess I'm really lucky. The type of photography I do doesn't really require a lot of detail, so I'm in the camp of - who gives a shit?

None of my photographs become better because of more detail and resolution.

 

[Sirius Glass]

I guess I'm really lucky. The type of photography I do doesn't really require a lot of detail, so I'm in the camp of - who gives a shit?

None of my photographs become better because of more detail and resolution.

As I posted earlier, the detail becomes important IF a portion of the negative is enlarged enough. For those of us who always perfectly compose the photograph before exposure, we do not need to be concerned with such matters unless we are making a really large print.