At what aperture lens has the most resolving power?

[Europan]

According to the optical laws resolving power increases with the opening. https://www.asprs.org/wp-content/uploads/pers/1947journal/mar/1947_mar_64-85.pdf

 

[AgX]

There is no useful rule of thumb.

Yes.
But one may put up a rule of thumb per basic design of the lens:

So for a 2.8 Tessar-type the optimum would be 2-3 stops up.

For a 1.2 double-Gauss it would be 6 stops.

For a slow double-Gauss as in macro ones, it would be 1-2 stops


But this still is not the whole story as this only indicates the maximum effect of stopping down . But it depends also on how much the gain per stop is. At some lenses the the values per f-stop can be very close to each other at the higher stops.

In general: "the first cut is the deepest" (if someone remembers that song...)

 

[cliveh]

2 stops down usually hits the sweet spot.

 

[wiltw]

According to the optical laws resolving power increases with the opening. https://www.asprs.org/wp-content/uploads/pers/1947journal/mar/1947_mar_64-85.pdf

Max theoretical resolving is wide open because diffraction effect is at the very minimum for any lens when wide open. But practically speaking, other factors (not diffraction) limit a lens' performance when wide open, and these other issues do not diminish until about -2EV to -3EV aperture size.

 

[outwest]

On the other end of the spectrum, I seem to remember something called the "Rule of 4". Dividing the focal length of the lens by 4 would give you the point at which diffraction may become an overriding problem. Anyone else ever hear of that? I believe it is based on the fact that it is the point at which the effective aperture is 4mm and there was a thing about apertures smaller than 4mm.

 

[Pieter12]

https://www.dxomark.com/Lenses/

 

[RalphLambrecht]

That might be true and that's a lot to swallow right that article.


Thank you! Taken together with this - I figure that in reality I should focus on the inherent contrast in the scene and camera shake to saturate them "mere" 160l/mm films. Let alone 260+

Just did some measurements and my projection setup is:
> Meridian PC 45mm f/2.8 MC lens
> 42mm image diagonal of Gepe 7013 Slide mount with metal mask
> ~2,5m throw distance (measured from slide) that projects ~2,13m diagonally. This gives me what, ~59x magnification right?
> 2,2m viewing distance. Measurements are approximate.
It would also be nice to know exactly what this projection lens can do - is it a limiting factor in this quest for sharper 1,20x1,8m projection or not? Guess I can ignore these figures in the future for sure


Thanks!


That's what I got from links above too, thanks!

please note: resolution ismeasuredin 'lp'(line pairs)/mm not'l'(lines)/mm; just sayin...

 

[Dismayed]

It's not just the lens. Shoot a larger format of film and you won't have to enlarge as much.

 

[Ivo Stunga]

please note: resolution ismeasuredin 'lp'(line pairs)/mm not'l'(lines)/mm; just sayin...

Noted.

It's not just the lens. Shoot a larger format of film and you won't have to enlarge as much.

Agreed. But in practical terms - medium format projection is both - more dead and more expensive in every step of it, but the acquired results - not that impressive to justify all the expense and sourcing of pretty much alien projectors and glass slide mounts. Especially in a world where Gepe discontinued their more popular/demanded 7013 mounts I'm preaching. Inaccessible for me.

 

[Steve York]

Rule of thumb film days was best aperture was 2-3 stops down from wide open.

 

[Ivo Stunga]

Rule of thumb film days was best aperture was 2-3 stops down from wide open.

And that's quoted elsewhere too. Now I know to play around the range of 4-8 and see what I like best. Oh, and you might want to check your signature - flickr is missing its com Nice stream though!

 

[AgX]

Rule of thumb film days was best aperture was 2-3 stops down from wide open.

This has been repeatedly said now. And long ago it got stuck to my mind.

But read my post #28.

As said in another post"it all depends"... But stopping down 2-3 stops can't harm anyway.

 

[DREW WILEY]

Interesting thread. Nice to have a refresher course on mythology. Lots of generalizations and stereotypes. Very few specifics. Let's see, what did I do yesterday when I wanted some serious resolution yet a fast operating camera? Left the Nikon home and simply grabbed my Fuji 6X9 RF instead. No math needed.

 

[Ian Grant]

As Drew says thers some myths, the major issue is resoltion tests are done with test charts but that's a flat field. Outside of copying work what we are really looking for is getting the best sharpness in often quite different conditions, so that's a balance between achieving sifficient DOF and the so called sweet spot, as it's a 35mm thread that's typically f5.6/f8 and the stops either side.

Resolution and sharpness aren't always the same, a 50mm Summicron has far higher resolving power than a 50mm Nikkor but the inherrent micro contast of the Nikkor may well give a higher apparent sharpness and with 100 ISO films the resilts are subtle diffrences..

Geoffrey Crawley in his abridge article on developer in 1961 stated:

" Sharpness "-the overall impression of a print or projected image, measured scientifically as
"acutance ", seen from normal viewing distance.

" Definition "-the extent to which fine detail is recognisably rendered in a print, etc. When
acutance of fine detail is good, then definition is good.

" Acutance "-the contrast at the edge of significant detail, a scientific measurement of the density
gradient at that point.

" Resolving Power "-the scientific measurement of the actual fineness of detail recordable by a
lens, film, or developer, or any combination of these three.

I find his thinking at cross purposes in some ways. Acutance is something added by developers and seperate to the sharpness and definition or resolving power of a lens. Very high acutance developers like Kodak HDD (only made and sold in Europe), Ilford Hyfin, and similar gave extreme acutance but poor grain and fonality and quite low definition/rsolution.

Crawley did his definition/resolving power lens tests with Pan F, I did my own in the eraly 1970's with EFKE Kb14 (later changed from the 14 DIN Tingsten light name ot Kb20 later Kb25 the ASA Tungsten speed). EFKE Kb14 waas alledged to give LF quality on 35mm film, well with careful exposure and development it came close to smewhere between 120 and 5x4 FP4 or Plus X but at the tme the emulsion had extremely poor hardening and needed delicate handling, I ended up using a hardening stop bath and later adding hardener to the developer.

As Drew saysit's easier to move up formats and that's what I did in 1976 going to MF and then 5x4 within a year, and hey EFKE R14 and Pl14 were available athough it was a decade before I used then in LF. I don't shoot a lot of EFKE film but still have maybe 150 sheet of 10x8 EFKE (Adox) 25, and a few boxes of 5x4 & Quarter plate.

When thinking of Reolving Power there's that mriad of other concerns, what film, what developer, and it really boild down to knowing your lenses and equipment and honing technique.

Ian

 

[gone]

I think for most photographers, having a lens w/ a lot of resolution runs a poor second to lens IQ, or image quality. There isn't any known correlation between the two of them either. I once owned some Zeiss lenses that were very sharp w/ lots of resolution, and didn't like them. The bokeh was not that great; pretty busy. Which is why I have a Leicaflex lens on my Nikon. It may not have the most resolution, it's nearly 50 years old, but I bought it because it takes marvelous photos wide open w/ beautiful bokeh, and stopped down it has a 3D look that few lenses can produce.

 

[AgX]

The same time the lens with the bokeh that you rejected might be be the better choice for a use where a gain of off-focus resolution is important.

It all depends...

 

[Down Under]

Photo books in the 1960s and 1970s advised to close down a lens about two stops from its largest aperture - so my f/3.5 lenses were at their best at f/5/6 and so on - but all my 'phot lit' from that era is long gone and I don't recall any of the technical explanations for this. Maybe someone of my same vintage could explain it better than I can.

I also recall reading a Leica guide book from about 1960 that advised to use a lenses at no smaller aperture than f/5.6. Something to do with diffraction or loss of sharpness or whatever it was. I think I may still have that book, but just now the effort of having to dig thru about 50 cartons in the garage is too much to think about. So it all stays somewhat vague in the back of my brain.

I still use all my DSLR Nikon D and 35mm Nikkor lenses mostly set at f/5.6 for my general photography and no smaller than f/8 for landscapes, which with my Nikon digitals auto-set at ISO 200 lets me shoot sans tripod without the danger of shake, which seems to be worse as I get older.

My MF lenses (Rollei TLRs) tend to produce sharp images at whatever setting I use. In the past when I was younger and could handhold my cameras without any shake, I kept my Planars and Tessars set at 1/60 at f/12.5 with 120 ISO 64 Ektachrome and it did the trick. For my architectural work it was f/16 or even f/22 and almost always a tripod and filters, also a polariser, but with tripod work sharpness was not a problem.

In my general photography I've found not all the 'planes' (depth of field) were in sharp focus. Two of the three worked best for me. Usually I focussed to keep the foregrounds slightly out of focus, which produced the best images for what I did.

Again, I can't provide a technical explanation for all this, but it serves me well, which is what 90% of my photography is all about - to get the best image. the other 10% is book learnin' and I've already had a lifetime of that, so it's good enough for me.

 

[Down Under]

Interesting thread. Nice to have a refresher course on mythology. Lots of generalizations and stereotypes. Very few specifics. Let's see, what did I do yesterday when I wanted some serious resolution yet a fast operating camera? Left the Nikon home and simply grabbed my Fuji 6X9 RF instead. No math needed.

Ah, yes. Like the brigade of "there is but one God of photography and thy name is Hasselblad." But then what else could we expect??

Indeed, all those ancient old myths. Even in the good ol' days when Leicas and Rolleis were the go and films were as slow as crawling caterpillars, Nikkors as they were then lenses could be used at f/5.6 and handheld. That Texas Leica has to be set at least at f/16 to get any decent sharpness for landscapes with everything sharp from your thumb in the lens to the distant horizon. I've used one of the beasts and I know. It's why I'm staying with my Rolleiflexes. Well, one of many reasons, anyway.

Better films from the 1980s on and nowadays digital flexibility has changed the game a lot, but let's not pooh-pooh the old learnin' too much. It was based on a lot of common sense and also much experimenting which meant shooting and loading a lot of film and then much time in the darkroom. And good photographers were producing superb images even when film speeds were stuck around the ISO 12-25 mark.

On the other hand, in the 1960s I shot a lot of (then ASA 200) Kodak Super XX with my uncle's Contax I, but most of the landscapes and family photos I did back then are not easy to print now, too harshly contrasty and lacking good sharpness. By the 1980s we had far better films and more scope for flexibility but even then the old tried and true, like Tri-X, Plus-X and Panatomc-X, still produced the goods.

In 2021 I shoot mostly at ISO 200 with my Nikon D800 and D700s and my results are sharp, show superb colors and pleasing tones, scan beautifully, and are easily printed with pleasing results. Now and then I put a roll of Kodak Panatomic-X thru a Nikkormat or a Contax G1 with good results. I don't enjoy the darkroom drudge as much as I did even ten years ago, but age has wearied me, as it inevitably does to all of us.

So yes, some of our mythology has changed, but surprisingly a lot of it is the same as it was. Long may it endure.

 

[DREW WILEY]

Well, I get very high sharpness from both of my Texas Lecias all the way from f/4 to f/16, with just a tad of loss at the extremes of either wide open or f/22. I've got plenty of inherited Rollei SL66 negs on hand, and they aren't close to the Fuji; but I don't happen to know what apertures my brother shot them at, or even the developer he used. The last decade of his life he borrowed my Pentax 6x7 because he got much better results with that, possibly an ergonics issue; but the Pentax lenses are really quite good, despite the all the Hassle-bland mythology. But with respect to my Fuji "Texas Leicas" (GW690ii and GW690iii, both with 90mm lens), it not a fair fight anyway, since 6x9 has twice the surface area as 6X6 cropped to a 645 rectangle.

But it's all relative. 4X5 blows away medium format, and 8x10 stomps 4X5. Godzilla always wins over Bambi. But there are various reasons I am multi-format. Just yesterday the Texas Leica was the perfect choice because I needed something which would work very fast due to people intermittently intruding into the scene. So a view camera would have been hopeless, while 35 mm would be clear out of its league given the size and quality of prints I'm anticipating. Just developed that roll of film (TMX 100) an hour ago, and so far, so good.

But in fairness to this thread, I should remark that when I need optimal sharpness and very high optical quality from my Nikon, I select the classic 55 micro. About all I use that for anymore is the copy stand. Out shooting instead, I mainly use 35mm format like an alter-ego to my larger format work, and seek small poetic, even grainy, prints instead of the big immaculate ones. Mostly use a classic 85/1.4, but sometimes an old single-coated A1'd 55 f/2 with just the right amount of softness to it. It's fun to switch things up from time to time.

Ian - I once used Efke 25 in 6X9 roll film backs for my 4x5 camera in order to get reasonably good 16X20-ish prints which wouldn't be embarrassed by being in the same portfolio box as prints made from 4x5 or 8x10 film. That's a hard task for any medium format film. No way on earth any kind of far smaller 35mm film is going to fill that bill. In terms of quality, large format provides all kinds of opportunities over smaller formats because one does not need to be as worried about grain per se. Ekfe 25 wasn't exactly versatile, and was slow, but while it was still around did provide the unique combination of a very long scale with high acutance. My lenses were fully equal to the task. But toward the end of Efke, serious quality control issues began showing up. Yes, you could get that kind of detail with Pan F, but nowhere near the contrast range.

For example, I remember when Kodak was initially advertising Tech Pan as "4x5 quality from 35mm film". Well, I guess if you went out and found the grainiest old sheet film on the market, had a very sloppy camera and filmholder, along with a marginal quality lens, and didn't know how to use any of that well .... maybe. But I happen to have 8X10 sheets of Tech Pan itself still on hand. Does that mean I can get "32X40 inch film quality" from a mere 8x10 camera? Ludicrous thinking. The tonality of Tech Pan was miserable anyway. It was a forensic and copy film to begin with. Quality has more than one connotation. One has to set priorities of what that means to them personally.

 

[AgX]

Photo books in the 1960s and 1970s advised to close down a lens about two stops from its largest aperture - so my f/3.5 lenses were at their best at f/5/6 and so on - but all my 'phot lit' from that era is long gone and I don't recall any of the technical explanations for this. Maybe someone of my same vintage could explain it better than I can.

As I tried to say so far: lenses react differently on stopping down.

There are 2 fenomenon involved:
-) growing diffraction
-) reduced aberrations

The maximum effect of stopping down is reached when both are at equilibrium.

The effect of growing diffraction is to foresee straight forward.
Aberrations are faults at the lens and their reduction by stopping down though are inherent to the individual lens design.

Above I pointed out that one may look at groups of lenses of similare designs.
I based this on published test results, But still within some groups the effect of stopping down varies.

And to repeat myself, what is the use of stopping down to that maximum effect, if the gain is so minime that you will not see it in practise? The effect by dividing exposure time in two, by stopping down one stop less, may have more visible effect.

We tend to ask simply for the best, the maximum, but things are more complex.

 

[reddesert]

On the other end of the spectrum, I seem to remember something called the "Rule of 4". Dividing the focal length of the lens by 4 would give you the point at which diffraction may become an overriding problem. Anyone else ever hear of that? I believe it is based on the fact that it is the point at which the effective aperture is 4mm and there was a thing about apertures smaller than 4mm.

There's nothing special about the 4mm physical aperture. The blur size on the negative caused by diffraction is proportional to f-number, not to physical aperture. (Because: diffraction spot size ~= focal length * wavelength / aperture diameter.) So the aperture where diffraction becomes an issue, depends on the f-number and the desired amount of enlargement to make a print. It would be possible to contrive situations where the rule-of-4 worked, for example 'normal lens for the format and roughly 8x10" to 11x14" print', but then you'd have to remember not to use it for other lenses, etc. It would be better to just associate an f-number to the format, like diffraction starts to take over at about f/16 for 35mm, f/22 for medium format, etc.

 

[grain elevator]

How often to you photograph flat subjects or with super wide angle lenses so that desired dof and perhaps hand-holdable shutter speed isn't the primary determinant of the aperture you use?

 

[Ian Grant]

Ian - I once used Efke 25 in 6X9 roll film backs for my 4x5 camera in order to get reasonably good 16X20-ish prints which wouldn't be embarrassed by being in the same portfolio box as prints made from 4x5 or 8x10 film. That's a hard task for any medium format film. No way on earth any kind of far smaller 35mm film is going to fill that bill. In terms of quality, large format provides all kinds of opportunities over smaller formats because one does not need to be as worried about grain per se. Ekfe 25 wasn't exactly versatile, and was slow, but while it was still around did provide the unique combination of a very long scale with high acutance. My lenses were fully equal to the task. But toward the end of Efke, serious quality control issues began showing up. Yes, you could get that kind of detail with Pan F, but nowhere near the contrast range.

For example, I remember when Kodak was initially advertising Tech Pan as "4x5 quality from 35mm film". Well, I guess if you went out and found the grainiest old sheet film on the market, had a very sloppy camera and filmholder, along with a marginal quality lens, and didn't know how to use any of that well .... maybe. But I happen to have 8X10 sheets of Tech Pan itself still on hand. Does that mean I can get "32X40 inch film quality" from a mere 8x10 camera? Ludicrous thinking. The tonality of Tech Pan was miserable anyway. It was a forensic and copy film to begin with. Quality has more than one connotation. One has to set priorities of what that means to them personally.

Efke 25 actualy had a daylight speed of 40-50 ISO, over exposure killed its sharpness, I used both it and Tmax 100 at 50 ISO with the same development time in replenished Xtol.

We agree on the claims of LF quality from 35mm Efke 25 a and also Technidol, after all Iused some 5x4 Technical Pan for work. The down side of the older Efke Kb14 was extremely poor hardening, that had improved by the time it became Efke Kb25 hardening had been imroved but it was still not as good as Ilford films, Techical Pan suffered from poor tonality, and while it had extremely high resolution it also had zero acutance

In the late 1980's I needed a sharp high resolution 120 film for my5x4's 6x9 roll film back and Agfa AP25 later APX25 was that versatile film, very fine garin, high resolution, great tonality, and prints from the 6x9 negatives stand up well alongside those from 5x4 AP/APX100 or Tmax 100. APX25 was probably the best slow film available.

Ian

 

[Nodda Duma]

There is no useful rule of thumb.

yep. Full stop.

learn your specific lens through use.

 

[Jim Jones]

Old rules of thumb have often outlived any practical application in an age that should include digital imaging. I often use the zoom kit lenses on my 3/4 frame digital cameras wide open because diffraction begins to limit resolution at smaller apertures, and these lenses perform surprisingly well wide open. So does an old Vivitar 400mm 1:5.6 Auto Telephoto. Back when Tri-X was the 35mm speed champion in photojournalism, stopping Leica lenses to smaller than f/8 introduced diffraction. If optimum resolution is important, we should test the lenses we use instead of relying on unreliable rules of thumb.