Veydra Cine Mini Prime MTF Optical Bench Tests

I have my preconceived notions, just like anyone else. A long while back the video techs told me we were stocking Veydra Primes in multiple focal lengths for m4/3 mounts. I just rolled my eyes and passed on by. Another boutique lens that would have poor resolution, ridiculous copy-to-copy variation, and a shelf-life-until-broken measured in weeks. Not interested.

But I noticed we were stocking more and more of them because they rented well; and added them in E-mount, too. I also saw they rarely came to repair. Then I did a little checking and found that our techs, who can check out any gear they want for their weekend shoots (it is an excellent perk, isn’t it?) were taking Veydras home pretty often. So I figured it was time to test them.

For those of you who don’t follow this blog, this is not a lens review. There are plenty of places to find that. This is the results of optically testing multiple copies. It won’t tell you a thing about how it focus breathes, how it handles, or whether it has that ‘film’ look. I will simply tell you how well it resolves (because you can roll your artsy eyes all you want; sometimes you have to make things look sharp), and about sample variation.

If you haven’t seen Veydra lenses, they are very small. They’re the Corgi’s in the Kennel of Cinema Lenses; they have all the features of the big dogs, but you think ‘really cute’ instead of ‘scare away burglars.’ Actually, maybe that’s a bad analogy, because Corgi’s cost as much as the big dogs, but Veydra’s are way cheaper than most Cine Primes. You can buy a set of these for what one nice Zeiss CP.3 lens would cost. On the other hand, like a Corgi’s short legs, the Veydras won’t cover a full-frame sensor; they’re either m4/3 or Super 35 image size and come in m4/3 and Sony E mounts.

In this post I’ll look at the three standard-to-telephoto- range lenses; Veydra 35mm T2.2 Mini Prime; Veydra 50mm T2.2 Mini Prime; and Veydra 85mm T2.2 Mini Prime.

MTF Testing

For those of you who don’t speak MTF, just remember higher is better and solid and dotted lines (sagittal and tangential) of the same color close together is better. The lower frequencies (red and orange) are more about contrast, while the higher frequencies (blue and purple) are more about fine detail resolution. For those of you who do speak MTF, I realize you’re going to have a field day criticizing that last sentence, but it is a decent generalization.

All of these lenses can cover up to a Super 35 format, so we tested them to +/- 14mm from the center. However, many people are using them on an m4/3 format. So I’ve left them in color in MTF graphs out to 10mm away from the center (m4/3 format), then black and white after that (APS-C format). It makes a difference, particularly on the wider lenses.

Veydra Mini Prime 85mm T2.2

This is an outstanding resolution, both in the center and off axis. Even at the full 14mm from the center distance, these are really high MTFs.

Olaf Optical Testing, 2018

Veydra Mini Prime 50mm T2.2

The 50mm lens has an amazingly high resolution in the center and maintains it very well through the micro 4/3 frame. At the edges of the Super 35 (or APS-C) sensor, there is some fall off, though.


Olaf Optical Testing, 2018

Veydra Mini Prime 35mm T2.2

Again, this is really good performance in the center half of the frame. It fades a little in the outer part of an m4/3 frame and a lot in the outer 1/3 of a Super 35/APS-C frame.

Olaf Optical Testing, 2018

Sensor Glass Effects

I am a geek, which means whenever I think of putting the same lens on different sensors, I worry about the effects of different sensor glass thickness. In this case, the lenses can be used on an m4/3 camera (4mm to 5mm glass thickness) or a Sony camera (2mm glass thickness). In theory, this might, or might not, cause the lenses to have very different MTF away from the center. So for each focal length, we compared m4/3 and E mount lenses, using 4mm glass for the m4/3 and 2mm glass for the E-mount.

Veydra Mini Prime 85mm T2.2

Olaf Optical Testing, 2018

Veydra Mini Prime 50mm T2.2

Olaf Optical Testing, 2018

Veydra Mini Prime 35mm T2.2

Olaf Optical Testing, 2018

There are some slight differences, but they are not significant for the 50mm and 35mm lenses. There may be a difference in the 85mm results, with increased sagittal-tangential separation off-axis, but it’s very small and could merely be the smaller sample sizes (a few copies each) used in this comparison.

How About Some Comparisons to Other Lenses?

Well, of course! I can’t use my handy computerized comparison tool here because most of the Cinema lenses we’ve tested are done for full-frame lenses. So I’ve done a little cutting and pasting from stock data. It should be adequate to make some comparisons, but remember, the graphs are of different size sensors. If you mounted a FF lens on a micro 4/3 or Super 35 camera, you’d only be using the central portion of the image, so that’s what you should compare. To make that a little easier I’ve gone with the ‘color the graph out to 10mm away from center’ technique I pioneered above.

Compared to Zeiss CP.2 Lenses

If you’re a CP.2 fan, you’re going to have to talk about the look of the lenses, or how they handle, or something similar to explain why you like them. Because, to be quite blunt, the Veydras just kick some serious resolution butt in this comparison. To be fair, though, Zeiss CP.2s have never been about absolute resolution. On the other hand, when you move up to 4k and eventually higher footage, the resolution is going to be more important than it was at 1080p, especially on smaller sensors.


Olaf Optical Testing, 2018


Olaf Optical Testing, 2018


Olaf Optical Testing, 2018


The graphs speak for themselves. If you are shooting on a crop sensor body and want the best resolution, the Veydra’s are simply better. You may well prefer the ‘Zeiss look’ or any of a dozen other factors. But if you want to claim the Zeiss out-resolves the Veydra, you are as incorrect in your preconceived notions as I was before I tested these. Of course, being incorrect never stopped anybody from saying stuff on the internet.

Field Curvature

While most people debate the MTF graphs, the field curvature (MTF vs. Field vs. Focus) graphs give far more practical information. We limited these to the m4/3 range because, well, that’s the range we were testing when we did the field curvature tests. These are single-copies, not the average of all tested copies.

Veydra Mini Prime 85mm T2.2

The overall field is very flat, but if you look closely, there’s a very slight sagittal curve in one direction, with a tangential curve in the other. It’s very mild, though. This particular copy has a slightly tilted field also, which some do, because real lenses, not manufacturer’s claims. This is not something you’d notice in actual use, however.

Olaf Optical Testing, 2018

Veydra Mini Prime 50mm T2.2

Because you’ll wonder about it, the little blob at the bottom of this graph is what happens if you get a tiny piece of reflective dust in the optical path of the MTF bench; it’s an artifact of testing a not a real thing. Otherwise, we have a very nice flat field here with just the slightest bit of curve in both fields.

Olaf Optical Testing, 2018

Veydra Mini Prime 35mm T2.2

Field curvature tends to be more apparent in wider lenses, and while there is more curvature in the 35mm Mini Prime, it’s still just moderate for a 35mm lens.

Olaf Optical Testing, 2018


The catch with the vast majority of less-expensive lenses is usually copy-to-copy variance. I will point out that it’s also an area of personal frustration for me, so I tend to put a lot more emphasis on it than many of you do. People look for that ‘great copy’ of the cheapest lenses with all the effectiveness of Monty Python’s Quest for the Holy Grail.

So, I fully expected to see a lot of variation in the Veydra Mini Primes because that’s what we always see in inexpensive lenses. If you’re sensing a pattern of “Roger’s preconceived impressions were entirely wrong” here, you are correct. The Veydras controlled copy-to-copy variation exceptionally well; far better than that other discount cine lens brand and better than a lot of quite expensive prime lenses. Using our internal variance number the Veydras all rated in the ‘excellent’ range.

Veydra Mini Prime 85mm T2.2

Olaf Optical Testing, 2018

Veydra Mini Prime 50mm T2.2

Olaf Optical Testing, 2018

Veydra Mini Prime 35mm T2.2

Olaf Optical Testing, 2018

So What Did We Learn Today?

We learned for the umpteenth time that doing scientific testing is a great way to shoot down Roger’s assumptions. We knew the Veydras were conveniently small at a very reasonable price. I was really surprised at how optically excellent they were and even more surprised at how consistent their performance is from on copy to another. These are currently sort of cult lenses that most people don’t consider very often. Most people, as is so often the case, are wrong for that. These are really very good lenses. Emphasis on “really” and “very.”


Roger Cicala and Aaron Closz

March, 2018

Author: Roger Cicala

I’m Roger and I am the founder of Hailed as one of the optic nerds here, I enjoy shooting collimated light through 30X microscope objectives in my spare time. When I do take real pictures I like using something different: a Medium format, or Pentax K1, or a Sony RX1R.

Posted in Equipment
  • NDA stuff I was referring to earlier. Not these, although these were a nice surprise.

  • Brandon Dube

    Nope, something else entirely. Testing at 900cy/mm. You can see some of the results in, ehh, three weeks or so.

  • Jiri Vrozina

    Why would i invest in Veydras?
    No superwides in a set nor Macro.
    For single operator that long focus throw is a nightmare. Give me Milvus set with 120’ degree focus throw anytime.
    But if one wants to enjoy good FAST optics (t2.2 is a slooow in 2018 for MFT),
    Then new fast SLR Magic cine hyperprimes or new fast Voigtlanders are the way to go.
    Lastly, last 12 months Veydras promissed 25mm Anamorphic and 8.5mm Wide Prime.
    Neither happened. I am sure they can at least bring out some Macro prime.

  • l_d_allan

    Just curious … are these Veydra cine lenses what you were referring to in an article previous to this one (Sony 24-105 Feb 14, 2018) …

    RC >> So, yes I am bitter and feeling a bit sorry for myself. I mean, two weeks ago I’m doing some double-naught-secret-agent testing on a spectacular wide aperture prime lens with a resolution that made my jaw drop at frequencies I wasn’t sure was possible to test.

    Or something else? NDA issues involved?

  • Unless you want slow as molasses, or have at least $25,000 to shell out, the Fuji MK zooms in 18-55mm and 50-135mm, both at T2.9, have you covered. (Unless your video camera uses the long flange EF mount.)

  • Well, you *can* have you cake and eat it too— in the form of Sigma’s cinema primes, to name one example. If you’re a single operator then the Zeiss Milvus line would be a big step up from the performance shown here by the CP.2’s, and at half the cost or less.

  • Ashamed all zee vay to zee bank…

  • I can’t say first-hand, but one of the guys who works here uses them for photo work on m4/3 a lot and loves them.

  • chrisgull

    […rolling artsy eyes…] ;p
    Great article as always, thanks for posting. I notice that Veydra specializes on cine lenses. Would these lens designs translate well to still-image lenses?

  • Munchma Quchi

    Zeiss should be ashamed. VERY VERY ASHAMED.

  • DrJon

    Excellent point, thanks, I shouldn’t write quite so OTOH. I’d still like them to test beyond the sensor limits for systems supporting IBIS, as I’ve certainly had one edge or the other get blurry come times with IBIS on with my GH5, but never both edges and I have plenty of images where the edges are fine (of the same subject from the same spot).

  • Stefanie Daniella

    smaller optics need not worry about difficulties of dealing with glass beyond a much cropped image circle

    the optical physics of glass means a cropping out excess outer perimeter glass, and carrying over remaining “core” optical geometries (for downsizing diameters) and tweaking it for thinner elements (downscaling length is more complicated than for diameters)

    as for other aspects beyond sharpness
    the wider-angle the lense, the more difficult it gets optically

    all the lenses looked at, fared superbly, given they were all medium to long FL (none wide-angles)

    plus, give these are aimed at cine, this means for m43, many 4k uhd are further cropped if camera uses only a straight 1:1 sensor read, without oversampling

    lenses need only to perform excellent on MTF to just under 10mm radius

    no point for folks to compare how “bad” primes are on larger sensor systems, when the optics must rate high all the way out to 20mm+ radius on MTF measures (far more difficult to design)

  • Janus Ng

    10mm is not necessary for IBIS only. The diagonal length of m43 is 21.60mm. Half of it is 10.80mm.

  • Marcel van Leeuwen

    Well i heated that the CP’s are in essence just rehousings of the original still lenses and not designed for cine work, just as so many other “cheap” cine lenses like the Samyang/Xeen lenses. And the Veldra’s are designed from the ground up as cine lenses, so there is that to keep in mind.

  • Devil’s Advocate

    No worries, it was just a ‘thought I’d let you know’ post 🙂

  • Athanasius Kirchner

    This is extremely helpful, honestly. Cheap, high-quality cinema primes have become a thing in the last few years, and it’s great to see Veydra helping to broaden the category.
    Now where are the cinema zooms? @_@

  • Sorry about that – because we couldn’t use our automated tool I did cut and paste for those and slapped “M” on there late at night without thinking. I’m swamped doing two different projects now and just don’t have time to redo the grafts from scratch right now. My apologies.

  • trackofalljades

    YES! I absolutely loved this article, but please do that. I was totally confused for a moment there. 😉

  • Devil’s Advocate

    You might want to change the ‘M mount’ caption in some of the graphs to read ‘M43’ to prevent confusion in the future. Aside from that, great work as ususal 🙂

  • It’s going to take a while; we don’t have as many copies of the wide ones so I have to wait on them rotating in and out of stock to get enough copies tested. Although the 25s are done and are very similar to the 35s.

  • Les

    Now I really want to see a test of the wide Veydras. 35mm is a long focal length for video on M4/3.

  • DrJon

    The 10mm for m43 is interesting, as I assume that leaves a margin for IBIS movement (half a m43 sensor being 8.65mm). Is there any aspect of measuring IBIS movement here or just a guess (sorry, scientific approximation)?
    Oh and very interesting, thanks.

  • A very good point Søren. But also part of the reason I was so very surprised the copy-to-copy variation was so low on these.

  • Well, full-frame coverage with some of them. And, of course, I can’t comment on the ‘look and feel’ part. But from the lab perspective, honestly, they blew me away, far better than I expected.

  • Carleton Foxx

    I’m sorry, my brain locked up trying to process this information. There must be a catch, right?
    So besides snob appeal, what are we giving up by renting one of these lenses instead of a Zeiss, a Super Baltar, or that gorgeous-looking Angenieux zoom?

  • Samuel H

    Also at T/2.2, which probably means f/2.0, they are not as fast as the lenses in the previous article. But it’s still a very strong showing by these little monsters.

  • Søren Stærke

    Do not underestimate optics designed for small sensors. If you design a lens for a 35 mm full-frame sensor, and then shrink it to half size (o it only covers 21 mm in diameter) then remember you also shrink the spot size by the same amount! Shrinking the spot size to half the size makes the MTF go up quite a bit. Unfortunately the tolerance requirements go up just as much as they impact the smaller optics way more.

  • Samuel H

    Wow, those are some unexpectedly impressive results. Thanks for testing!

Follow on Feedly