Brian Caldwell

Sorry - I overlooked the EF camera mount restriction. I hope Canon, Blackmagic, etc. will wake up one day and do the straightforward engineering needed to fix this.

Just add a 0.71x SB to your 24-105 and you have a 17-75mm f/2.8

That's *really* big!

If you're doing a wide angle Iscorama-style attachment with a large diameter front, then my statement about the variable diopter being "easy" obviously needs amendment, since the mechanics does become much more difficult. Although I'm a big fan of big lenses, they tend to require exotic/expensive items like precision linear bearings to keep them moving straight and without play. Maybe not entirely relevant here, but its interesting the the new Cooke anamorphics also use a pretty basic variable diopter focusing scheme, although the net power is negative rather than afocal: http://pdfaiw.uspto.gov/.aiw?PageNum=0&docid=20140300973&IDKey=735679DAC5C4&HomeUrl=http%3A%2F%2Fappft.uspto.gov%2Fnetacgi%2Fnph-Parser%3FSect1%3DPTO2%2526Sect2%3DHITOFF%2526p%3D1%2526u%3D%25252Fnetahtml%25252FPTO%25252Fsearch-bool.html%2526r%3D5%2526f%3DG%2526l%3D50%2526co1%3DAND%2526d%3DPG01%2526s1%3Diain.IN.%2526s2%3Dneil.IN.%2526OS%3DIN%2Fiain%252BAND%252BIN%2Fneil%2526RS%3DIN%2Fiain%252BAND%252BIN%2Fneil

I think the death of anamorphic film projection and the resulting appearance of dirt cheap anamorphic projection lenses (e.g. Schneider Cinelux etc.) was also an important factor. After all, the variable diopter part is easy, its the cylindrical lenses that are hard to make. It will be interesting to see what happens once the supply of grossly underpriced projection lenses dries up.

Note, at the time the Isco patent was issued in 1970 the law was that the lifetime of a patent was 17 years from the date of issue. When a patent expires the time-limited right of the assignee to a monopoly also expires, and the technology becomes part of the public domain. So, its been completely legal to manufacture and sell Iscorama style systems since 1987, almost 30 years now!

As long as you use a fullframe (24x36mm format) master lens, then both the 0.71x ULTRA and 0.64x XL will work fine. The ULTRA is a little better corrected in the outer part of the image, but this is mainly important if you want critically sharp results at f/1.0 with a Sigma ART or Zeiss Otus, and would not be noticeable with a slow lens like the Canon 24-105. If you plan to use an APS-C lens such as the Canon 17-55/2.8, then you should be aware that you may encounter slight vignetting when shooting HD (not 4k) with the 0.64x XL. The reason for this is that even though the XL itself covers the entire m43 image just fine, the image circle of the lens is too small to cover after it is multiplied by 0.64.

Can a camera that only offers a EF or PL lens mount truly be considered mirrorless? True, there is no mirror, but if you throw away the possibility of short BFL optics what is the point?

Hi Dan: Thanks for the info! My only point of reference was the Panavision 25mm, and it definitely seemed like a lot of distortion to me. Very interesting to learn that Hawk and others are worse.

The Panavision G 25mm does have severe barrel distortion. I haven't seen the new Cooke 25mm yet - does it have the same weird mix of pincushion and barrel distortion that the other Cooke anamorphics have?

The linear portion non-parfocal behavior can be fixed with a back focus adjustment. This is easy to do with a Speed Booster. With a plain adapter you'll have to luck out and be able to do it by shimming the flange instead of having to remove metal. This will give the long and short focal length positions the same focal position. What happens in-between is the non-linear portion, and is controlled by the cam that moves the compensating lens group inside the lens. If you have a significant lack of parfocality in the in-between region the standard way of fixing it is to re-machine the compensator cam, but only super high-end cine lenses get this treatment.

Not sure if any of this will help in your situation, but try keeping the focal length as short as possible, using the smallest feasible aperture (note that diffraction may not be as objectionable compared to image degradation due to heat haze), shoot when the ground and air are close to the same temperature (e.g., at dawn, after a rainstorm, etc.), and maybe try some image stacking/averaging techniques used by amateur astronomers.

I certainly wouldn't argue that old lenses might not be preferable in certain circumstances. After all, there are a number of cine rental houses that make tons of money from large inventories of vintage glass. However, the author of the article in question should at least do careful comparisons to make his points, because it seems to me that lighting is the dominant effect that he's demonstrating, not lens differences. And his attempts to appear scientific by making those ternary charts is just a complete joke. I'm shocked that so many people on his blog and elsewhere just accept them as having some sort of meaning without questioning. Any 35/1.4 FF SLR lens plus a 0.71x Speed Booster will give a 25/1.0 for aps-c.

I appreciate your comments, and you certainly make some valid points. However, 4:3 will never actually become obsolete as it is always available from 16:9 via cropping. And a 4:3 crop of 16:9 will always have more pixels than a 6:5 crop. So, in my view the updated 6:5 ratio is simply a nod to an old (and soon-to-be obsolete?) film standard, and its unlikely to replace the ARRI Alexa 4:3 standard. After all, the recent explosion in anamorphics used for feature films is largely due to the native 4:3 mode in the Alexa. Regarding the GH4, you are probably correct regarding the market. Which is one reason I want to make a lower cost Iscorama-type attachment available. But the prime lenses I have in mind will be really special, and should appeal to people who view the camera as a small accessory that you attach to your lens rather than the other way around. Bear in mind that its pretty simple to switch from a PL mount design to m43. Another little thought: 1.79x on a 1:1 sensor (or a 1:1 crop of a wider sensor) gives you 1.79:1 = 16.11:9. In other words it gives almost a perfect rectangle for anamorphic HDTV.

I probably should have stayed away from this topic, but I found the article so utterly revolting that I just couldn't help myself. Its difficult to even begin a thorough critique, and I'm not going to do it here. Suffice to say that he relies almost entirely on non-standard, non-optical terminology (e.g., 3D, tonality, micro contrast, flat), and he never bothers to define these terms in a precise and unambiguous way. Lenses can be fully characterized by things that are well-defined and can actually be measured, such as MTF, veiling glare, distortion, transmission vs. wavelength, etc.. Why rely on pseudo-scientific nonsense when proper characterization methods are well established? He then goes on to show images that supposedly illustrate his points, but none of the images are actually carefully done comparisons under fixed lighting conditions, post processing, etc., so they are absolutely worthless as far as I'm concerned. He also makes a bunch of ternary diagrams that make absolutely no sense at all - "line of realism"??? WTF??? And his obsession with the number of elements in a design is absolutely bizzare.

Sorry, but somebody's got to say it. This has to be the worst piece of trash writing ever done about lenses. The author has absolutely no clue what he is talking about, and should be ignored.

The main technical hurdle is that focal reducers reduce the physical length of a lens in addition to reducing the focal length. So if you simply build a focal reducer from cylinders you would wind up with many millimeters of astigmatism throughout the image. An even bigger concern, as valid points out, is that you don't get any of the desirable anamorphic artifacts with a rear anamorphic attachment.

Thanks, Andrew. The Canon 24-70/2.8II is actually the one that I own, and I like it a lot. However, I picked the Tamron as an example because it has optical stabilization, which would make up for the shortcoming in the A6300.

I certainly agree with both your main points: 1) there is no magical aesthetic to MF, and 2) MF glass is not generally higher performing than FF glass. The newer Zeiss lenses are just flat-out amazing in terms of IQ, but are expensive (similar to many MF lenses). The Sigma ART lenses are an amazing blend of high performance at a modest price. And it does appear that Canon, Nikon, Tamron et al. are starting to catch up. I think it really is a golden era in photographic optics, but most of the gold is being bestowed on FF rather than MF.

Both. The problem with medium format lenses is that they are by and large very slow. As in f/2.8 on the fast side, and often way down around f/3.5-f/4.5, even for prime lenses. After you add a focal reducer the speed is improved, but still just isn't very exciting. I know there are a handful of f/1.9 - f/2.0 medium format lenses out there, but performance is not good enough to warrant a special Speed Booster IMO. By contrast, the latest FF optics by Zeiss, Sigma et al are breathtakingly good at very large apertures. The older medium format lenses just aren't in the same league, and I doubt there is enough of a market for them to ever catch up. At one point I thought about doing a Speed Booster for the Leica-S lenses since they're actually pretty good. However, they're also very expensive and you'd have to reverse-engineer the electronics to control the aperture.

I just wish there were at least one example of a speedboosted medium format lens that can't be equaled or bettered by a FF lens.

So, does this mean we need to do yet another custom Speed Booster?

I'm not entirely sure about the IQ - its something I would want to test. Its worth noting that inexpensive stills glass is often better than very expensive cine glass. For example, I once did a projection bench comparison of an older generation Nikon 70-200mm f/2.8 zoom against a $20k Zeiss 80-200 T2.9 CZ.2, and the Nikon was clearly superior, especially in the corners. I suspect there is a good reason why you *never* see MTF charts of cine glass! I haven't found a way to do it in a compact form factor while maintaining even remotely decent image quality. There is an infamous Kodak patent that addresses this problem, but the example given in the patent is so incredibly poor that its just not enabling at all. You could do it with a relay type system, which I would call a Keplerian style, or real image focal reducer. However, you then wind up with a huge ungainly and under-performing setup just like the old Nikon E3. Its incredibly nice to learn about this (!!), and I'd like to know as much as possible about your work. BTW, a Speed Booster XL will take an f/2.8 zoom down to f/1.8, which is probably around T/2.0 for most zooms.

What really matters is the BFL, not the flange distance. In this case, as in most cine zooms, I think the Fuji 14.5-45/2.0 will fit onto a PL-to-EF adapter, so you can think of it as an EF lens. So, in this sense its equivalent to 24-70mm lenses from Canon, Tamron, or Sigma with an EF mount.

The Fuji Premier 14.5-45mm T2.0 ( http://www.bhphotovideo.com/c/product/840519-REG/Fujinon_HK3_1X14_5_F_HK3_1X14_5F_14_5_45mm_T2_0_ZOOM.html ) is pretty well known in the professional cine world. But, if you've ever hefted one in person you know its not only priced like a small house (~$100k), its nearly the same size and weight as well! So, naturally, I'm curious to know how it might perform on the upcoming A6300 in comparison to something much cheaper. Like the Tamron 24-70mm/2.8 with a Speed Booster, which gives you a 17-50mm/2.0. Obviously, the latter is practically $free compared to the former, and you do give up the 14.5mm - 17mm range along with perhaps 1/4-1/3 stop difference in aperture due to f/# vs T/#. But, you gain autofocus, image stabilization, a little extra reach at the long end, and an unbelievably huge reduction in size/weight/cost. I would not be surprised in image quality is nearly equal.