Snake Oil For your Lens

[Mainecoonmaniac]

Is this a bunch of BS?

http://www.metabones.com/

 

[E. von Hoegh]

Is this a bunch of BS?

http://www.metabones.com/

Of course. Penis enlargement for optics.

 

[Mainecoonmaniac]

Extenze for your lens?

Of course. Penis enlargement for optics.

For those inadequate photographers that need that little something?

 

[rawhead]

Short answer: No, it's not a bunch of BS. Theoretically very sound.

It's very different from "penis enlargement" because "penis enlargement" suggests that the penis is enlarged while the rest of the body stays the same size. This technology only works because (a) it takes a lens from a larger (135) format and uses it on a smaller (APS-C) format, and (b) it utilizes the big difference in the flange focal distance between SLR (EF mount) and a mirrorless (E mount).

Making the lens focal length 1.5x shorter and adding 1-stop in aperture while moving from FF to APS-C is, in fact, only letting that lens act on the APS-C sensor exactly as it would on FF. Really, it's not even magical.

 

[Dali]

Nothing new here, this is a focal length reducer. Clear aperture being the same as the f.l. is reduced yields to a lower f/d ratio. This is (was?) widely used with Schmidt-Cassegrain telecopes which have a rather high f/d ratio (usually 10 or 11) for photography purpose mainly. I doubt it increases the image quality though...

 

[rawhead]

I agree, the only part about their claims that's dubious is that this will somehow increase image quality. The best we could hope for is that image quality degradation is suppressed to a minimum.

 

[E. von Hoegh]

Short answer: No, it's not a bunch of BS. Theoretically very sound.

It's very different from "penis enlargement" because "penis enlargement" suggests that the penis is enlarged while the rest of the body stays the same size. This technology only works because (a) it takes a lens from a larger (135) format and uses it on a smaller (APS-C) format, and (b) it utilizes the big difference in the flange focal distance between SLR (EF mount) and a mirrorless (E mount).

Making the lens focal length 1.5x shorter and adding 1-stop in aperture while moving from FF to APS-C is, in fact, only letting that lens act on the APS-C sensor exactly as it would on FF. Really, it's not even magical.

You'd better recheck your theory. It "reduces" the focal length by adding an extra element to an existing optical system - the extra element has not been calculated to work with that system. It's the reverse of a telephoto exender, with all the attendant problems and even more aberrations. "Improves MTF"? Please.

 

[E. von Hoegh]

For those inadequate photographers that need that little something?

Well, I'd say it's for the current crop of less-than-knowledgeable equipment junkies who think gadgets equal skill and believe everything they read.
I'm surprised there was no claim for improved bokeh.

 

[Mainecoonmaniac]

Improve MTF? Is MTF "Male Transfer Function"?

 

[E. von Hoegh]

Improve MTF? Is MTF "Male Transfer Function"?

"Money Transfer Factor"; from the gadgetographer's account to metabone's.

 

[rawhead]

It's the reverse of a telephoto exender, with all the attendant problems and even more aberrations. "Improves MTF"? Please.

Nope, untrue. A telephoto extender goes between the rear element of the lens and the image sensor and tries to do its magic while remaining in the same system. That is where most of the stress comes from. To repeat, this technology is different because it utilizes the difference in the flange focal distances of an SLR system and a mirrorless.

If introducing a lens between (1) source of light, and (2) the image sensor necessarily introduces attendant problems and aberrations, then why even bother taking a photograph in the first place? I mean, all your Summicron is doing is introducing problems and aberrations, no?

Here's the deal.

You have light rays emitting from the rear of your lens. It's going to hit the sensor plane after ~45mm and create an image circle that will cover FF.

It is theoretically very sound and possible to introduce a set of optics within those ~45mm that will converge & concentrate those light rays so that after those ~45mm, the image circle will be smaller and just cover APS-C without significantly reducing the image quality.

Like it's been said, this concept has been used before most notably in astrophotography (of the best kind, I may add). If it's such a degrading "snake-oil" concept/technology, why would some of the most expensive astrophotographic imaging equipment use it?


Now, wether or not the actual execution of the concept here is entirely satisfactory remains to be seen. But seeing that one of the most talented lens designers living today is involved in the development of the product, my bet would be on him than people who apprently don't understand some basic principles of physics.

 

[rawhead]

With regards to improvement in IQ: this, while being (admittedly) a rather dubious claim, again is not necessarily impossible. Assume a set of optimally designed optics that will do what I described above without reducing IQ by simple introduction of more glass & air.

Then, what would happen is that when compared to using that lens with a glass-less adapter on the smaller (APS-C) format (this is the important point), all the aberrations etc. of that lens will ALSO be compressed, so that when the image is printed at the same paper size (IMPORTANT: NOT AT THE SAME MAGNIFICATION), then theoretically, you will see much less of those aberrations.

Again, this isn't anything magical. Think of the purple fringing you see around, say, tree branches against a cloudy sky background. The purple fringing is going to be less significant when you compare an image taken with that lens on FF (it's native magnification) than an image taken with that lens on APS-C (with 1.5x apparent focal length, which magnifies everything including purple fringing).

If, on the other hand, you bring the two images to the same magnification, then, you should have the same amount of aberrations, etc. You would expect exactly this, because we assumed in the beginning that the optics introduces no IQ degradation. In reality, it probably will, so when the images are shown at the same magnification, then there's probably going to be more aberrations with this product.

So you see, to repeat, there's really nothing magical or unbelievable or snake-oil-like thing being proposed here at all. Indeed the concept is simple enough that even I came up with it independently a few years ago while chatting with my brother.

 

[ntenny]

Just going from first principles, it seems like a focal reducer could improve the MTF precisely because it reduces the size of the image---the same reason small enlargements look sharper than big enlargements. The extra element necessarily introduces some aberrations, but if the improvement due to the smaller image scale exceeds those aberrations, it could be a net win.

But I don't know how the performance of these things shakes out in the real world. Should we be fitting focal reducers to large-format lenses and mounting them on our 35mm cameras?

-NT

 

[rawhead]

Here's an early review and some sample images & video. Too early to make a final judgement, sure, but there's nothing here that supports the "snake-oil" hypothesis, either:

http://philipbloom.net/2013/01/13/speedbooster/

 

[Steve Smith]

Am I correct in assuming it takes the 'normal' view a lens would have on 35mm format and focuses the whole of that image onto a smaller sensor?

If so, it seems logical. It's something I thought about a few years ago when I was temporarily a digital photographer - not as a business idea, just one of the idle thoughts I have occasionally!

Some people might be querying the faster claim. It's because the lens now actually has a shorter focal length but the aperture diameter range stays the same. Simply, aperture (f No.) = focal length / aperture diameter).

Should we be fitting focal reducers to large-format lenses and mounting them on our 35mm cameras?

That's a thought I had a few minutes ago. It's easy to test the theory. If you have a large format camera and you can find a positive Dioptre (close up lens) which will fit either on the front or back cell, you will find that the bellows extension is shorter for infinity focus than without the positive lens. If you had a big enough ground glass, you could see if the image area increased too but this is not relevant to the system being discussed here as it is for smaller than normal sensors.


Steve.

 

[E. von Hoegh]

Just going from first principles, it seems like a focal reducer could improve the MTF precisely because it reduces the size of the image---the same reason small enlargements look sharper than big enlargements. The extra element necessarily introduces some aberrations, but if the improvement due to the smaller image scale exceeds those aberrations, it could be a net win.

But I don't know how the performance of these things shakes out in the real world. Should we be fitting focal reducers to large-format lenses and mounting them on our 35mm cameras?
-NT

For the price, $600, I'd say not.

 

[E. von Hoegh]

Am I correct in assuming it takes the 'normal' view a lens would have on 35mm format and focuses the whole of that image onto a smaller sensor?

If so, it seems logical. It's something I thought about a few years ago when I was temporarily a digital photographer - not as a business idea, just one of the idle thoughts I have occasionally!

Some people might be querying the faster claim. It's because the lens now actually has a shorter focal length but the aperture diameter range stays the same. Simply, aperture (f No.) = focal length / aperture diameter).


Steve.

That's the only claim which holds water.

 

[rawhead]

Am I correct in assuming it takes the 'normal' view a lens would have on 35mm format and focuses the whole of that image onto a smaller sensor?

Yes you are correct. The only thing that this product purports to achieve is to take the "normal" view a lens would have on a 135 format and give you the exact same view (FOV, DOF, etc) on the smaller APS-C sensor.

If so, it seems logical. It's something I thought about a few years ago when I was temporarily a digital photographer - not as a business idea, just one of the idle thoughts I have occasionally!

It is very logical, and simple, theoretically. Simple enough that you and I could come up with it completely independently.

Some people might be querying the faster claim. It's because the lens now actually has a shorter focal length but the aperture diameter range stays the same. Simply, aperture (f No.) = focal length / aperture diameter).

Exactly. And, it's really NOT going to make the lens "faster" per se. Because we're dealing with a smaller sensor which, all else being equal, gives you more noise per square inch of the same size print, having a faster F value doesn't really give you an advantage over shooting with that lens on FF with 1-stop extra ISO.

That is, with everything else in the sensor technology (and processing technology) being equal, taking a 50mm F1.4 lens (on FF) and:'

(a) shooting it on an APS-C camera with this adapater as a 33mm F1.0 lens at ISO 100, 1/125s, and
(b) shooting it on an FF camera straight-up as a 50mm F1.4 lens at ISO 200, 1/125

should optimally give you identical results. Identical in terms of FOV, DOF, AND Noise levels.

 

[Steve Smith]

And, it's really NOT going to make the lens "faster" per se. Because we're dealing with a smaller sensor.

In theory it should because it is gathering the light from a wider area than the sensor would normally see and putting it onto the same sensor area so the light density (if such a term is correct) will increase. This doesn't take into account any light loss through the additional lens though.

I remember when my father owned an Olympus E20. It came with wide angle and tele convertors which fitted to the front of the lens. I remember him telling me that with a front mounted wide convertor there was no increase in aperture like you get with a tele convertor. There was however, no claim that it made the lens faster though!


Steve.

 

[rawhead]

But I don't know how the performance of these things shakes out in the real world. Should we be fitting focal reducers to large-format lenses and mounting them on our 35mm cameras?

-NT

This is an interesting question. Let's assume that we can utilize this concept & technology without limits, for the time being (e.g., with the product in question, it's limited to F1.2 lenses that will act on APS-C as an F0.9 lens).

There are certain lenses in the large format world for which we have no equivalent in smaller formats, and so people go out of the way to shoot them on LF & film in this day & age. The Kodak Aero-Ektar on 4x5 (or 5x7) comes to mind (my fave, by the way: http://www.flickr.com/search/?q=aero+ektar&w=88929764@N00&s=int).

178mm F2.5 lens on 4x5 translates to something like a 53mm f0.8 lens on full frame (ref: http://www.pointsinfocus.com/tools/depth-of-field-and-equivalent-lens-calculator/). We do not have such a lens at our disposal on FF.

IF there can be a reducer like this that can attach to this large format lens and concentrate the light so that it will act like a 53mm f0.8 lens on my Canon 5Dmk2, then EFF YES, I would purchase such an adapter for $600, in a friggin' nanosecond.

 

[rawhead]

Steve: yes, I understand that it does make it "faster" in that light density is going to increase. My point is that the lens becoming 1 stop faster on an APS-C camera is not going to give you any real-world advantage over shooting with the same lens on FF 1-stop slower F value, BUT 1 stop extra ISO.

 

[Steve Smith]

You're probably right about real world advantages. At the moment I'm just seeing it as an interesting item for discussion.

I don't know if anyone has posted this yet but here is a detailed explanation: http://www.metabones.com/images/metabones/Speed Booster White Paper.pdf


Steve.

 

[polyglot]

Well, I'd say it's for the current crop of less-than-knowledgeable equipment junkies

You need a mirror, E.

These devices are merely reverse teleconverters and can produce as much resolution as a modern small digital sensor (about 4um), with which they are intended to be used. There's nothing magic or bullshit, they don't need to be designed in conjunction with any particular lens: is your film designed with your lens in mind? It just takes the focused image and shrinks it by 1.5x linearly, which gives exactly the same effect as a one-stop shorter+faster lens with less coverage. The recorded image on APS will be identical to the recorded image on 135.

Just because you don't understand it doesn't make it wrong.

 

[rawhead]

Here's one area that I would LOVE to see this technology utilized for. And I understand I'm standing on dangerous grounds here, since, while not limited to digital, the usefulness of what I propose really only applies to the digital world.

So, you know how many Hasselblad (and other 6x6 MF) shooters are really hoping for (but may never get) a full-size 56mm x 56mm sensor medium format digital camera with which to use their legacy Hasselblad/Zeiss lenses? Again, they'll probably never get one since a square digital sensor would be a very niche market in an already niche digital MF world.

What that means is that the legacy Zeiss glass can only be used cropped on a 645 and smaller sensors.

However, let's say that in the future, we will have (1) mirrorless digital MF cameras. You can substitute this relatively easily today because we already have digital MF backs that can be attached to any box. Hartblei's H-cam comes to mind. Let's say that then there is (2) an adapter exactly like this for MF that will concentrate the image circle by 1.5x.

Then, you could have a digital MF sensor like one used in CFV-16 (37mm x 37mm) and use a legacy Hassie lens on it as it was originally intended without any crop factor, and with the same DoF. I.e., e.g., Zeiss Planar 110mm F2.0 will act like a 73mm F1.4 lens on a CFV-16, effectively giving you the same FOV/DOF as the lens would give you if you shot it with a 56mm x 56mm sensor. Of course, this could be applied to any of the larger digital MF sensors (full size 645) too, and a legacy square lover would just crop to square.

More intriguingly, if you consider that many of the Zeiss lenses have an image circle that is comfortably larger than that required to cover 56mm x 56mm (that's why you can use them with a Flexbody), it is possible that such an adapter used with legacy lens will cover a fullsize 645 sensor and give you, in this case, an even wider FOV and even shallower DoF than they were originally designed to on traditional 6x6 film.

I bet legacy Hassie users (and many others) would be very, very, very happy to shell out $600 or even $1600 for such an adapter (again, provided there's a mirrorless MF digital body with which to use it).

I'm of the camp that thinks ALL digital cameras of the future will/should become mirror less, so I'm very intrigued by this idea.

 

[Sirius Glass]

Of course. Penis enlargement for optics.