[falconeye]

look at that person first - OP has long road ahead of him to get there

Thanks for the link.
However, my sample #3 with 1.8 micron detail does already outperform the results I have seen by fotoopa. The point isn't the numeric macro ration (27:1 in his case) but the resolution actually achieved.

If you turn the macro ratio up too much (as fotoopa did by using afocal photography), you end up with blurred images (at 100%) and you detoriate the MTF figures of your lens used as it may be designed for infinite focus.

Currently, I am able to do 3:1 macros with sharp 100% crops on a K20D (i.e., 6:1 with sharp 50% crops). The physical limit is 10:1 anyway. At least as long as you use visible light

[deudeu]

Are you trying to make some fake bills?
Go for the 10 Euro ones, it is not as profitable, but they are less likely to look for them!

Just joking of course, This is a very interesting experiment you are doing here, but depending on what you are trying to take pictures of, wouldn't be easier to hook your camera up to a microscope?

[thePiRaTE!!]

@ deejjjaaa - thanks for the fotoopa links - fascinating stuff.

@ falconeye, best of luck in your endeavor- should be interesting!

[falconeye]

wouldn't be easier to hook your camera up to a microscope?

That's a good question!

I actually don't like microscope images, they look too artificial. In the end, I don't want to do scientific research, I want to take nice photographs at a scale not normally accessed. I.e., I want to preserve the 3D look of our world.

A 100x microscope lens has a focal length of 2mm. And I would have to mount it directly to a lens on my camera as I want to take images directly. I think, the small focal length and small image circle make it impossible to take photographs outside of laboratory conditions. But this very idea to use microscope lenses is under evaluation as well

So, may I repeat my question?

Any photo lens with >500 lp/mm?

[philmorley]

what about the canon lens that does 5x / 5:1 magnification ???

Canon | Macro Photo MP-E 65mm f/2.8 1-5x Manual Focus | 2540A002

[roentarre]

The sharpest tool in photography is edge sharpening in photoshop, not a lens.

All lenses are soft among the consumer groups these days.

[falconeye]

what about the canon lens that does 5x / 5:1 magnification ???

Phil, thanks for the interesting pointer. I studied this option a bit now and this is what I am thinking about:

The MP-E 65mm f/2.8 1-5x is, almost exactly, a DA 35mm Ltd. internally reverse-mounted to the EOS mount. So, it is reverse-mounted to the mount but you won't see it because it is done within the lens objective.

In consequence, the macro magnification doesn't go from 1:0 to 1:1 and focus from infinity to 70mm. But from 1:1 to 5:1 and focus from 70mm to ~35mm.

The problem is that this is an innovative product idea but technically well known and won't give you new options -- except if the lens also has stellar optical performance. Here I read:

"It produces images that look great overall. There is some noticeable sharpness loss at the extreme end on this range (5:1), but it isn't anything to write home about IMO."

I think I have already surpassed the options this lens could offer me (I got 6:1 which looks rather sharp than soft). Thanks again for bringing this option to our attention. I wasn't aware of it.

[falconeye]

The sharpest tool in photography is edge sharpening in photoshop, not a lens.

I cannot agree. The Zeiss I used obviously delivers on the promise of 300 lp/mm where even a K20D sensor would be happy with 97 lp/mm.

And with respect to Photoshop: By sharp, I mean good resolution, not good contrast. Otherwise, just convert to binary black and white and you have a razor "sharp" image still lacking any detail.

[ll_coffee_lP]

Very interesting work...and love the shots of the currency.

Can I be so bold as to ask what you plan to shoot at this resolution? I'm very curious in what you are hoping to achieve.

c[_]

[troyz]

Falconeye:

Have you tried using a lens with a shorter focal length (e.g. reverse a really good 20mm-ish lens onto your 300) to get a greater magnification (instead of optimizing the resolution at the lower magnification)?

If you go for the short focal length, you might get better bang-for-the buck with a non-retrofocus rangefinder lens (e.g. Cosina/Voigtlander 21) instead of a retrofocus SLR lens.

[Aristophanes]

I am on the research of the sharpest lens. ………

According to photozone.de, the three top performers for the K mount are:

From my understanding, Photozone.de tests with single samples of loaner lenses. Not new lenses either, so this is not factory sampling.

It's a terrific site and a wonderful objective look at lenses, but for your purposes, the sample variation cannot pin down what you want. He's not testing enough lenses across the product run to nail it down. Furthermore, the variations under your criteria fall within a % or less in some cases, so individual lens variations between manufacturers could actually be attributed to wear of a particular donated lens, etc. There's a lot of uncontrolled factors in the tests.

Hinted at in the preamble to many of the tests is the idea that the results mimic the wealth of anecdotal information provided by thousands of shots from thousands of photographers. He's confirming what is already known.

[falconeye]

Can I be so bold as to ask what you plan to shoot at this resolution? I'm very curious in what you are hoping to achieve.

Well. I am a curious person. I think one has to keep an open eye at any scale to get some good shots. The level of magnification I am talking about means subjects which pass unnoticed to the unaided eye. Unlike 1:1 macro subjects which you would identify as good subjects.

So, I do not yet know what I will discover.

What I am hoping to achieve is photography which is novel (to a certain extent at least). Besides technique, this means searching for new subjects as well.

Have you tried using a lens with a shorter focal length (e.g. reverse a really good 20mm-ish lens onto your 300) to get a greater magnification (instead of optimizing the resolution at the lower magnification)? [...]non-retrofocus rangefinder lens

No, but maybe it is a good idea. I didn't limit my quest to SLR lenses. Physically, it's resolution which matters -- only. The focal length is irrelevant. Of course, a larger magnification means an even smaller field of view not making things easier Which is why shorter focal length should be accompanied by better resolution as otherwise, you only get more blur.

E.g., the 24mm Canon lens was remarkable. But the 35mm Zeiss even more and at equal resolution, the Zeiss would enable better photography.

Following your suggestion, I have looked at some of those, e.g., the Leica Elmarit-M 21mm f/2.8. (f/4 lenses cannot be as sharp as the 35mm Zeiss which is sharper at f/2.8 than at f/4!). Leica publishes MTF figures and to my disappointment, the 40 lp/mm MTF isn't better at f/2.8 than at f/5.6. Still no competitor to the 35mm Zeiss.

Zeiss itself claims that their SLR lenses resolve up to 300 lp/mm and their rangefinder lenses (ZM series) up to 400 lp/mm. So, they are somewhat better indeed. But which of the ZM lenses is the top performer? Maybe, the Biogon T* 2,8/21 ZM?


Any concrete lens suggestion?

[falconeye]

for your purposes, the sample variation cannot pin down what you want.

Yes, but this was the only start I had. And the fact that only one lens seems to resolve better at f/2.8 than at f/4 (and another one close at least) seems to be at least one reliable result. A lens with a significant factory variation cannot be top notch anyway.


Anybody knows about good removable P&S lenses?

[troyz]

Any concrete lens suggestion?

Sorry, no specific suggestions. I've never used the reversed-lens-coupling-ring method and the short-focal-length lenses I've used (some of which have darn good reported MTF measurements) aren't necessarily well-corrected in the system you are building.

[falconeye]

(some of which have darn good reported MTF measurements) aren't necessarily well-corrected in the system you are building.

This is the nice thing: A reverse-coupled lens is used exactly as it was designed for: at infinite focus. So please, what are the darn good reported MTF measurements (I need > 500 lp/mm)?