New SLRmagic 25mm f/0.95 tested on the GH3 (along three other 0.95 lenses).

Share

We need speed we love speed! And 3D-Kraft satisfied our need by testing four f/0.95 lenses on the new Panasonic GH3 (Click here to read the test). These are in short the qualities and drawbacks of the lenses:

SLR Magic HyperPrime CINE 25mm T0.95
Vignetting and distortion are almost invisible which lets me speculate that the usable image circle might be significantly larger than what a FourThirds sized sensor requires. The lens will be available soon on eBay. Save this search on Slidoo to get notified when it’s available on eBay (you have to login).

Voigtlander Nokton 25mm f/0.95
A tiny bit less shapr than the SLRmagic and a tiny bit more vignetting. Price check at Amazon, Olympus US store, Adorama, B&H, eBay.

P.Angenieux Paris 25mm Type M1 f/0.95
The lens produces some geometric distortion and vignettes already quite strong on mFT sensors. These characteristics are probably the reason why some people tend to say this lens produces a very cinematic look. Click here to check out the price and auctions on eBay.

Schneider-Kreuznach Xenon 25mm f/0.95
Sharper than the Angenieux and with less Vignetting and distortion. Click here to check out the price and auctions on eBay.

 

Overall this SLRmagic sounds like an impressive piece of glass! And it’s almost half of the price of the Nokton ($799 but with an initial rebat $649 only). Chapeau!

Current SLRmagic lenses for Micro Four Thirds:
SLR Magic 12-36×50 ED Straight Spotting Scope lens (w/ hand grip) is now available on eBay (Click here).
SLR magic 11mm f/1.4 lens you can preorder on eBay (Click here).
Toy Lens 26mm f/1.4 lens on eBay (Click here).
SLR Magic 35mm f/1.7 MC lens on eBay (Click here).
Noktor SLRmagic 50mm on eBay (Click here)

Share
  • Twaddler Belafonte

    It’s amazing how the Schneider and Angenieux lenses produce literally nauseating, disorientating out of focus elements. Truly horrid!

    • Kristo

      Yes…but for 2 lenses back from 1920s the result is very nice…

    • Matt

      That’s because the Schneider is using a centre-filter. They’re using the CCTV version of the Schneider Xenon for their tests, which IS INDEED HORRID. You can just get the lens without the filter or remove the filter yourself and you get bokeh better then all the other lenses tested here. I can’t believe they’d do a serious test with that filter in place, are they crazy? That’s meant to help auto exposure CCTV cameras, not actual image quality.

      I have personally tested quite a few 25mm f0.95 lenses, and the Schneider is by far the best both in sharpness and bokeh.

      Of course with skewed and unprofessional tests like these no one will ever know. Why would you even test lens sharpness when you’re A) Not keeping ISO the same for every shot and B) shooting at ISO 1000!?

      This is a horrible test.

      • Bob

        I don’t believe you tried the Schneider before unless you ised the CCTV version. FYI The Schneider ND is not removable.

        • Matt

          It very much is. You simply remove the front or rear group and you can remove the ND center filter. Yes, it requires knowledge, no, it’s not impossible. I’ve done so on many Schneiders.

          • bananas

            hahaha @Bob got owned

            • Andrew

              bananas, Bob did not get owned. Matt here is talking about modifying a lens with his own knowledge. Would you take your lens front and rear group apart to take or put things in? If it is a filter in the front like most lenses or rear like some other lenses then it is part of the lens design. If you need to take a lens apart to take something out then Bob is not correct. Bob is just saying something like you cannot take the stereo out of his car. It is a situation where someone insists it can be removed with some knowledge but that does not necessary mean a normal person can easily take the stereo out of their own car.

              • Matt

                Bob is incorrect about a multitude of things, calling the filter non-removable and saying it’s the ‘cine’ version of this lens, which was not ever made.

                You keep talking about the Schneider lens as if it’s some sort of consumer lens for a Canon DSLR.

                It’s an industrial lens meant for CCTV usage. You can indeed, take it apart and remove the filter. Is it meant for someone clueless to do it? No, you’re supposed to be knowledgeable about the things you use in an industrial setting.

                You talk about my ‘hacking’ but completely ignore that your Schneider Xenon 25mm f0.95 was hacked even further by removing it from it’s motorized contraption as it was sold.

                I’ve said so OVER AND OVER AGAIN. But you keep ignoring this:

                Schneider sold these WITH filter ONLY in motorized form.

                Schneider sold these WITHOUT filter ONLY in manual focus form.

                If you have one with manual focus and aperture WITH a filter, you’re using a hacked up lens.

                This is not me making this up, this is in Schneider’s literature.

          • Andrew

            Matt, the thread is getting so long I can no longer see the reply button. We are also discussing details that is not well known by everyone. Please email support@slrmagic.com and write attn: to Andrew so we can continue the discussion.

      • Bob
        • Matt

          That’s what the Schneider CCTV version looks /TODAY/. The looks of Schneider lenses changes over time, as with any brand that keeps the same optical schemes. The one they’re using is one not available anymore and older, when they had a different styling.

      • Matt: Regarding the ISO it was explicitely said in the test that the shooting took place right after sunset (with decreasing light conditions) in order to have that special light mood and that these conditions were not primarily intended to do a sharpness test. These fast lenses are typically used under even more difficult light conditions where ISO may be usually between 800 and 1600 if you want to keep exposure times short in order to avoid motion blur of people. And that may affect sharpness, dynamic range and colors but not bokeh as you complain about.

        Regarding the ND filter of the Schneider-Kreuznach: There are two versions of that lens, the CCTV version and a “high end” version. As I do not own that lens, I was affirmed by the owner that the lens used in test is the optically superior version which contains the filter built-in. If you remove it by taking the element apart, the lens would not work as expected anymore. If you insist in your position, please show us references showing pictures with and without that filter and of course show me examples of that lens used wide open with a non swirly bokeh!

        • Matt

          “were not primarily intended to do a sharpness test.”

          Then why do you compare sharpness and call one lens ‘better’ when you’re not intending to compare them? It just becomes a half-assed test. If you’re going to test lenses, you do it under the same circumstances.

          If you take two shots with different lenses, one at ISO 400, and one at ISO 1000, you cannot compare *anything*. You cannot compare sharpness, you cannot compare contrast. You cannot compare bokeh, as that will also change depending on the sunlight to light ratio, which makes bokeh more harsh or less harsh as fluorescent lights will stand out more. However, my complaint about bokeh comparisons comes from using the center filter on the Schneider.

          I have various 25mm Schneider Xenon’s with their filter removed, alas, I have no proper test results with and without. But let me write up a little about the filter:

          The filter exists for CCTV usage, it’s not a center filter as you’d expect to limit vignetting. It’s a small glass filter with a black dot in the middle. It is used to offer more aperture settings as a small adjustment in the aperture causes a more dramatic change in the light entering the lens. Obviously, CCTV usage also means that bokeh quality is not something to care about.
          Otherwise, the Schneider 25mm f0.95 CCTV is *identical* to any other version. If at the time you ordered a Schneider 25mm f0.95 from Schneider, you simply had the option of having a center filter installed or not. Nothing changes, otherwise. The center filter is not part of the optical formula, as it is pure optical glass with no curvature whatsoever.
          The second reason why using this lens with the center filter for tests is that with the filter installed, it’s simply not as fast. See it as a second aperture. Of course, it’s still an f0.95 lens, as the f-stop is a simple calculation, not one that actually says how much light reaches the sensor. With the center filter installed, you cannot get t0.95 amount of light, at all. It’s about a whole stop to two stops back from that, though you’d have to check official Schneider documentation to compare that. You’re simply using an f0.95 lens with a permanent f1.2 / f1.4 aperture installed. This might also be where the misconception comes from that this version would be ‘sharper’, as it’s simply stopped down somewhat with an oddly shaped aperture. There is no advantage gained from it, and it’s only limiting the lens in speed and quality of the bokeh.

          I strongly suggest forming an opinion based on a proper version of this lens, or at least removing the center filter from the lens before comparing it. It’s a great lens. Whether it’s better then the Magic SLR one, I do not know, as I have not used it (obviously), but I’d be curious to find out. This however, is not the way.

          • Matt, I strongly refuse your allegation, that this test is flawed. If the primary intention were to compare sharpness/contrast, I would shoot some of those boring charts and some other static objects at base ISO and perfect light – and the article explains and clarifies that! But what would you learn from such a chart test for this type of lens? This is not the real life situation where you use that kind of lenses at all.

            The test shows a typical available light situation where you would benefit from such a fast lens. The open aperture shots of the Voigtlander Nokton where taken with ISO 500 and those of the SLR Magic HyperPrime were shot at ISO 400 which does not make a significant difference regarding image quality on the current sensor of the OM-D and th GH-3. And of course that is about the advantage you have with the HyperPrime CINE due to its transmission of T0.95 over the Nokton that has a transmission of about T1.1 at open aperture.

            Of course it makes sense to compare also crops at these real life situations as well – and compare sharpness in that situation at least for the Nokton and the HyperPrime. You also can clearly see from the pictures that the Angenieux would never reach that amount of contrast and sharpness at open aperture even if the shots would have been taken with base ISO. The pictures also show that the Schneider-Kreuznach makes a very good job regarding sharpness even at ISO 800 and that is about the disadvantage in low light situations that you have to take into account due to the ND filter as you already explained. So it is not the 100% reference sharpness test with MTF charts etc. but it still tells a lot about what you can expect from these lenses in typical available light situations where other aspects like bokeh quality, vignetting etc. count for most users at least with the same weight as sharpness.

            • Matt

              “but it still tells a lot about what you can expect from these lenses in typical available light situations where other aspects like bokeh quality, vignetting etc. count for most users at least with the same weight as sharpness.”

              You cannot. Vignetting becomes more apparent when the photograph is lighter or the ISO is higher, the bokeh is very dependent on focus distance on the model, the outside light to artificial light ratio. Bokeh will seems harsher if you’re shooting artificial light against a dark background vs a light background, as there’s nothing to blend the light with. The less light, the slower the shutter speeds, the more movement in your model, the less sharpness, and so forth.

              I’m all for testing lenses in real life situations, but you *have* to keep major variables identical otherwise there is simply nothing to compare.

              If you don’t want to compare sharpness, I understand. But if you don’t plan on doing that, then don’t write about it. Despite saying you don’t want to compare them or it’s not the point of the test, you continuously write about it. Even now you’re talking about the sharpness of the Angenieux lens, despite the horrible testing conditions.

              You even have an entire part of your comparison to compare sharpness.

              Yet the test isn’t about sharpness.

              I’m sorry, do these tests for yourself all you want and I don’t care about it, but don’t publish them with such horrid testing conditions and implications. Be aware of how your articles will influence people’s opinions. If you don’t want to compare sharpness, don’t write about comparing sharpness. Look at the comparison shots shot at f1.4: You lose more sharpness by the change in ISO then the change of lens.

              This is all of course loose from all the other points and using a CCTV lens with an internal center filter. There’s just no point in comparing lenses like this. I don’t even like LPM charts or comparing lenses, I prefer to be out and shoot with whatever works, but I can’t help but feel duped by such an article as yours which just ends up promoting your friends’ lens.

              • RFC1925

                Got to agree with Matt here, there are too many uncontrolled variables in this test for a meaningful comparison.

          • Matt, you say you have several Schneider-Kreuznach with an without ND (and with removed ND as well when I understand your writing correctly). So why do you say you are’nt you able to proof with test results? Just writing and confusing readers with unproofed personal points of view does not help.

            I have the information, that the ND is made directly on one of the lens elements so it can NOT be removed.

            • Matt

              I got plenty with center filters at the time, but I removed each and every one. There is simply no use for a center filter for any creative imaging use, it’s purely made for CCTV usage.

              However, since you insist so much, I’ve gone ahead and re-installed one of the center filters in one of the lenses: Here you go: http://i.imgur.com/W1G4gnk.jpg

              Nothing too scientific, but clear enough to make you understand that the center filter is NOT an optical element (It’s called a filter for a reason, and is installed at the aperture’s location inside the lens) and that you’re limiting the lens’ light income and harming it’s bokeh. The difference would be more clear at night where bokeh becomes distinct from single light points, but this will have to do. The design of the center filter has changed over time with retooling of the line, but all that has changed is the filter’s fading on the edge of the dot and the size of it.

              Secondly, on the lens you were using, the center filter is completely uncoated. You’re adding two more air to glass surfaces to a lens, uncoated, which will greatly harm contrast and sharpness.

              If you order one from Schneider, you can choose as a last minute option whether to get one with or without. They install the filter to your desire. The optical formula is still identical.

              You talk about confusing people with statements without proof, when my whole point is that your statements are without proof. The irony! Get back to that Schneider lens, open it up, and you’ll find a completely flat center filter inside of it.

              • Thanks for the helpful image.

                But I still would not call the bokeh without filter “smooth” and I am sure, the newer lens designs will look better in that aspect.

                • Matt

                  That is beside the point. I’m quite certain both the Voigt and the SLRMagic will be better in this regard.

                  The point is that your test is flawed in many regards and many ignorant statements have been made regarding the lenses used.

                  Even if a scientific test and a flawed test have the same end result, does not mean you are correct. You simply cannot compare these things if you don’t test them correctly, or you’ll always end up confirming your own opinions that were present before the shoot.

                  Next time do some research instead of assuming what you hear is correct, or you’ll just end up being a marketing puppet for someone else, and I doubt that’s the goal of your articles (I hope it’s not, at least).

                  • Matt, if I had used a modified Schneider-Kreuznach, people would call the test flawed as well. The version of that lens was tested as you get it from stock.

                    I can not proof, if you can order versions without ND filter and if the rest of the lens would be identical and if it would behave like your modified version.

                    For real life (low light) situations, it can also be meaningful to show that the camera is forced to double ISO in order to compensate the decrease of light caused by the filter and the vignetting.

                    The test did never claim to fullfill scientific laboratory conditions. The impression you can get out of this comparison is at least valid in terms of vignetting, character of bokeh (for the ND version of the Schneider-Kreuznach) and still gives you at least an idea regarding contrast and sharpness. A more scientific sharpness comparison may follow later, when the production version of the HyperPrime CINE is available.

                    • Matt

                      “Matt, if I had used a modified Schneider-Kreuznach, people would call the test flawed as well. The version of that lens was tested as you get it from stock.”

                      Very true, because the Schneider lens isn’t the reason the test is flawed as I have written before. The difference in ISO, shutterspeed, focus distance and what not are the reasons the test is flawed. I’m merely commenting on the Schneider due to personal experience, as both you and Andrew seem to be clueless about the center filter.
                      It did NOT come stock. It was an optional add on for industrial CCTV usage, as I have written over and over, but you ignore it once more.

                      “I can not proof, if you can order versions without ND filter and if the rest of the lens would be identical and if it would behave like your modified version.”

                      There is nothing ‘modified’ about my lens. It’s simply a removal of the center filter, as you would install a center filter. Proof of this is really simple, download the datasheet from the official Schneider Kreuznach website:
                      https://www.schneideroptics.com/Ecommerce/CatalogItemDetail.aspx?CID=1336&IID=5959
                      “Option: Optical filter”.
                      If that doesn’t prove enough for you, send Schneider an e-mail and you’ll quickly find out.

                      “For real life (low light) situations, it can also be meaningful to show that the camera is forced to double ISO in order to compensate the decrease of light caused by the filter and the vignetting.”

                      Certainly not, as no one in their right mind would use a center filter in such situations. A center filter is designed to increase difference in aperture against change in the aperture ring. By creating a dot in the middle (o) you can change the light income on your sensor change drastically, as the difference between f0.95 and f5.6 will cause the lens to be closed completely. This should give another hint this isn’t meant for imaging quality, as you can never reach a high aperture with a center filter installed, as it will close off the lens.

                      It’s also not ‘meaningful’ if the difference is because of a change in the light due to the sunset, and not the change in lenses.

                      “The test did never claim to fullfill scientific laboratory conditions. ”

                      And I never fault you for not doing it in a lab. I’m blaming you for comparing things you cannot compare when you change ISO, shutter speed and focus distance. I understand you feel the need to say you can, but you simply cannot. So much changes between these things. Bokeh is probably the thing you could compare the most, and I would probably not have replied all that much if it was just that.

                      But despite this being the thing you can compare, you write an entire paragraph about sharpness and image quality, despite these being completely incomparable. You cannot compare sharpness from an ISO 400 and ISO 1000 shot.

                    • Matt, as there is no reply option clickble for your last post, I will reply this way:

                      You may criticize that HyperPrime image was taken with ISO 400, Nokton with ISO 500 and Schneider with 800 although this is about the advantage that you have in real life at available light due to different transmission of those particular versions of the lenses. The impact of these differerent settings on bokeh is neglectible. There is an impact on sharpness but most of that, what my paragraph about sharpness says is, that it can not derived as reference from my test. You continously seem to ignore that.

                      You may not criticize or call it a flaw, that in that test a lens with center filter was used. It is the way it is and that version of the lens was not used on purpose in order to give the other lenses an advantage. You can not ignore that an unknown percentage of used lenses out there in the market have that filter and most of the users will not be able to remove it – or do not even know that there are different versions. You have to accept that.

                      The good thing from this discussion and the article is, that it can illuminate this aspect to people looking for that lens in the future. It is mentioned in my article (and of course it was already mentioned right from the beginning).

                    • Matt

                      “that it can not derived as reference from my test. You continously seem to ignore that.”

                      And you continuously say this, that you cannot derive it from your test. And you are right. Yet you still talk about which lens is sharper, even if it cannot be derived from your test. You have an entire paragraph talking about the sharpness of the lenses with 100% crops and talking about how sharp each lens is.

                      Yet it cannot be derived from your test.

                      If you cannot derive something from your test, you do not analyze it. It is pointless.

                      “due to different transmission of those particular versions of the lenses. ”

                      You shot during a sunset. Light changes rapidly, and difference from 5 minutes can cost you even half a stop, as I can assure you from my own shoots (As you will more then likely know yourself). You make it extra clear that light was changing rapidly, so there is no way whatsoever you can compare the real life transmission of these lenses when you cannot tell whether it’s the lens or the light that has changed. The only way you can compare lens transmission is if the light you’re shooting stays identical. Obviously, during a sunset, it does not.

                      “You may not criticize or call it a flaw, that in that test a lens with center filter was used.”

                      I call it a flaw that you use a CCTV lens with a center filter installed and call it a ‘optically superior version’. I see you’ve now gone ahead and changed that in your text, to where it says “in order to compensate light fall-off to the edges”.
                      This is incorrect. A center filter that you would normally put in front of your lens, like this; http://www.teamworkphoto.com/images/schneider/centerfilter.jpg .
                      A center filter as used at the aperture is not used to limit vignetting or fall off, but is used to increase speed in exposure adjustments when controlling the aperture with a motor.

                      “It is the way it is and that version of the lens”

                      You can get the exact lens without a center filter. In fact, very very few of these lenses exist with center filters, and from most Xenons I have had in my hand, sold, or have seen sold, only one or two had a center filter. It is a *very* specialized feature for industrial use under quick light changes. It was also *only* used in Schneider Xenons that were built in motorized blocks, so either a seller was trying to dupe people by removing the entire motor assembly and selling it as a ‘non’ motorized version of the lens or the buyer would have had to do that himself, in which case he’d be well aware of the center filter. You can find info about the use of a spot filter as Schneider calls it on their website (http://www.schneideroptics.com/info/faq/industrial.htm#qu14).

                      If you want to use a center filter with such a lens for imaging purposes, it’s silly. If you want to compare that, then by all means, go ahead. My issue is not that you use a center filter with a lens, my issue is with you calling it a superior version and simply being completely unaware and ignorant about what you’re using, then continuing to discuss this with me as if this is some kind of secret knowledge. It is well known, even under hobbyists, that the center of the lens is the sharpest part of the lens. With a center filter, you’re forcing the lens to only use the edges of the optical elements.

                      “The good thing from this discussion and the article is, that it can illuminate this aspect to people looking for that lens in the future.”

                      And the bad thing is, is that it becomes pseudo-advertisement in the form of an article, coming to conclusions under flawed circumstances. When you write articles and post them on the web, you have to be aware of what happens to your article. In this case, people will draw conclusions based on yours. They assume your testing was done correctly, will look at the images and compare them. For example, this can be seen in the post we’re commenting on right now; “A tiny bit less shapr than the SLRmagic”.

                      If you would test something properly, that wouldn’t be an issue as people are free to make their own comparisons using the images you provide to them.

                      In this case, your images are flawed and tell a non-existing story. Be aware of what you write and what you publish.

                    • Matt, I agree that the “optical superior” was misleading and removed it from the article. I just explain that there are two different versions and which version was used for the shooting.

                      Regarding sharpness: It is valid to comment on sharpness even under these circumstances. You propably overestimate the decrease in image quality of the new sensor used in the GH3. For example you can see that the Nokton is significantly sharper at F1.4 compared to 0.95 although the F1.4 shot was taken with ISO 800 and the F0.95 shot with ISO 500. I just say you should not use it as a scientific reference, but you can get an idea. And even an Angenieux biased reader will see that the loss of sharpness and contrast of the P.Angenieux at open aperture is not caused by the higher ISO. But the reader will also see from the ISO that the transmission of P.Angeniex at F0.95 seems to be significantly lower as it required ISO 1000 already compared to the ISO 800 of the Nokton F1.4 shot that was taken only 4 minutes befor.

                    • Matt

                      Thank you for removing that bit then.

                      “It is valid to comment on sharpness even under these circumstances. ”

                      It’s completely impossible to compare sharpness with such varying ISOs. Look the following two shots:

                      http://i.imgur.com/A4GGVSR.jpg

                      http://i.imgur.com/05ngEnn.jpg

                      Do you not see the difference in sharpness? Look at the trees. Look at how sharply defined the leaves and branches are. ISO changes this big, even on the GH3, have a huge impact on both contrast and sharpness. The difference in aperture that was used in these shots should have made the lens *sharper* as it went from close to wide open to a middle-range aperture, no diffraction is taking place in these shots.

                      “But the reader will also see from the ISO that the transmission of P.Angeniex at F0.95 seems to be significantly lower as it required ISO 1000 already compared to the ISO 800 of the Nokton F1.4 shot that was taken only 4 minutes befor.”

                      You said yourself that you’re shooting under RAPIDLY CHANGING light with a SUNSET. Anyone who has ever shot during a sunst KNOWS that the light changes during the last hour of a sunset extremely fast. The light is going from bright to COMPLETELY NON-EXISTING. The difference of 4 full minutes can cost you one to two FULL STOPS of light during a sunset shoot. Then you go on to attribute this to the lens’ transmission? 4 minutes is a very long time during a sunset shoot. It’s a shame Flickr strips the images of time & date EXIF, because I’d be worried to see how long the shoot took in it’s entire length. You can see the difference in ambient light if you look at the artificial light in the background and compare it with the ambient light.

                      Light changes during a sunset. You cannot compare lens transmission when light changes rapidly, only when you shoot them at identical times.

                      Lastly, you mention ISO 1000, but I see ISO 1250.

                    • Matt, your shots do not contain EXIF data, so it is not possible to determine the ISO difference, not possible to see which camera (sensor) was used and not possible to see the exposure time (e.g. regarding motion blur in the leaves). And the shots were taken at night, where noise at higher ISO in the dark areas has a completely different impact than in my pictures that still have quite good light. In the context of these photos you talk about ISO performance of the GH3 although I am quite sure that those photos where NOT taken with the GH3 and that you have currently no idea about the ISO performance of that camera. When something is flawed, than THIS.

                      You simply continue to ignore what the pictures (and crops) in my article as well as the explanations demonstrate and you only rattle down your personal position. You may have some special knowledge about Schneider lenses but that is not enough to twist all facts of my article. There are so many other things that you screw up that it makes no sense anymore to continue the discussion with you.

                      For example you write that my pictures contain no EXIFs (of course they have EXIFs when you navigate in Flicr to that corresponding menu or open the “original size”. Flickr only removes EXIF from the images they scale down on their server) but you said also (in a meanwhile moderated post) they have varying exposure times (which is wrong – all shots were taken at 1/125s in order to avoid motion blur of the model). You say also you see ISO 1250 (from pictures without EXIF?) and seem to relate that still to the F0.95 comparison. None of the F0.95 shots had ISO 1250, only the F1.4 shot of the Angenieux required already ISO 1250.

                      You talk about rapidly changing light conditions during sunset. This is correct for the last few minutes before the sun disappears. But my article clearly says that the pictures were taken short AFTER sunset. In that period the lighting conditions do not change that quickly. The main impact on the ISO differences was caused by the different lens transmission. And of course the main difference in sharpness comes from the different lens performance especially at open aperture and not from the differences in ISO performance between 400 and 1000) of the sensor of the GH3 with its 16 MP (that compare to 64 MP on a fullframe sensor in terms of line resolution).

                    • admin

                      Thanks Helge! You made it very clear. Now let’s move on and not stick into endless discussions!

                    • Matt

                      Matt, your shots do not contain EXIF data, so it is not possible to determine the ISO difference, not possible to see which camera (sensor) was used and not possible to see the exposure time (e.g. regarding motion blur in the leaves). And the shots were taken at night, where noise at higher ISO in the dark areas has a completely different impact than in my pictures that still have quite good light. In the context of these photos you talk about ISO performance of the GH3 although I am quite sure that those photos where NOT taken with the GH3 and that you have currently no idea about the ISO performance of that camera. When something is flawed, than THIS.

                      You simply continue to ignore what the pictures (and crops) in my article as well as the explanations demonstrate and you only rattle down your personal position. You may have some special knowledge about Schneider lenses but that is not enough to twist all facts of my article. There are so many other things that you screw up that it makes no sense anymore to continue the discussion with you.

                      For example you write that my pictures contain no EXIFs (of course they have EXIFs when you navigate in Flicr to that corresponding menu or open the “original size”. Flickr only removes EXIF from the images they scale down on their server) but you said also (in a meanwhile moderated post) they have varying exposure times (which is wrong – all shots were taken at 1/125s in order to avoid motion blur of the model). You say also you see ISO 1250 (from pictures without EXIF?) and seem to relate that still to the F0.95 comparison. None of the F0.95 shots had ISO 1250, only the F1.4 shot of the Angenieux required already ISO 1250.

                      You talk about rapidly changing light conditions during sunset. This is correct for the last few minutes before the sun disappears. But my article clearly says that the pictures were taken short AFTER sunset. In that period the lighting conditions do not change that quickly. The main impact on the ISO differences was caused by the different lens transmission. And of course the main difference in sharpness comes from the different lens performance especially at open aperture and not from the differences in ISO performance between 400 and 1000) of the sensor of the GH3 with its 16 MP (that compare to 64 MP on a fullframe sensor in terms of line resolution).

                      Helge, these are two shots taken with a Panasonic GH3, one at ISO 400, one at ISO 1250. Shutterspeed is 1/125th. Noise has effect on the entire sensor, not just on ‘dark parts’. That is not how sensors work. It becomes more visible on a black surface, but the same noise is present in the entire image. I have no clue why you’d even think I’d use shots from a different camera, that is simply insinuating once more and it’s simply rude.

                      “You simply continue to ignore what the pictures (and crops) in my article as well as the explanations demonstrate and you only rattle down your personal position. You may have some special knowledge about Schneider lenses but that is not enough to twist all facts of my article. There are so many other things that you screw up that it makes no sense anymore to continue the discussion with you.”

                      Then reply to those things with arguments instead of simply stating that I am incorrect over and over. This is a discussion, not a yes / no battle. I have presented my arguments, if you ignore them there simply isn’t a discussion. If I said something incorrect, bring it up instead of saying I am incorrect. No one else commenting here except you and Andrew thinks the test is valid, and everyone with a right mind wouldn’t compare sharpness from ISO 400 and ISO 1250, or lens transmission in a rapidly changing moment of light.

                      “For example you write that my pictures contain no EXIFs”

                      I did not write that at all. Please read before you reply. I said Flickr removes the >>TIME & DATE EXIF<<. It removes the exact time the image was taken and replaces it with the time it was uploaded. This is how Flickr operates. I was indeed incorrect about the exposure times, my apologies for that, but keep in mind that while 1/125th gives a decent apparent sharpness, it is not enough to test sharpnes on a moving subject either.
                      Whether you used ISO 1250 for the f1.4 comparisons is not what I am discussing, I am discussing your sharpness comparison when using 1250, as even on 1250 you comment on the sharpness of the lens, despite that being a pointless comparison.

                      The golden hour as it is called is often shorter and more distinct in light changes then anything that comes before it. The light goes from dim to complete black in a short amount of time, yet you seem to think this will not influence the amount of light for some reason. Once again, the difference in ambient light can be visibly seen in your test shots, so this is not up for debate.

                      I don't know why you'd even bring megapixels into this, but if you'd want to make that comparison: You'd be more correct comparing sharpness from ISO 400 to ISO 1250 if you HAD 64mp, as less data would be lost due to noise. On a smaller 16mp sensor a lot of data is the actual noise of the image.

                      In the end, what you are claiming is that there is no loss in detail going from ISO 400 to ISO 1250. That is ludicrous, and you know it.

              • Andrew

                The photo shows the center spot ND was hacked off the lens as there are no screw threads on the center ND filter. The test was based on stock condition rather than modified condition. For example, sensor benchmarks are also done in their stock condition without having the AA filter selected by the manufacturer hacked off the camera sensor to test for maximum potential of a sensor inside a particular camera.

                You said:
                “The design of the center filter has changed over time with retooling of the line, but all that has changed is the filter’s fading on the edge of the dot and the size of it.”

                The sample you showed had very similar circles of confusion to be significant even after your modification of the lens.

                You said:
                “Secondly, on the lens you were using, the center filter is completely uncoated. You’re adding two more air to glass surfaces to a lens, uncoated, which will greatly harm contrast and sharpness.”

                Difference in contrast was also not significant in your samples.

                You said:
                “Notice the difference in aperture. Focus did not need to be changed, as optical formula stays identical.”

                Are you sure you did not need to change the focus? When dissembling the lens and reassembling the lens to install the ND filter to the position intended by Schneider-Kreuznach the focus and optical alignment would change slightly. This means you were holding the ND glass in front of the lens rather than installing it back into its original position? This changes everything in terms of sharpness comparison.

                Final word, the comparisons with and without the ND filter hacked off in your samples did not show any significant difference in the characteristics of the circles of “CoC” or bokeh swirl or sharpness.

                • Matt

                  Dear Andrew,

                  “ND was hacked off”

                  The filter is locked between the aperture and front lens group. They screw out and are calibrated with shims, hence why there are no ‘screw’ threads, as you wouldn’t find on any center filter in any CCTV lens. That is not how they’re designed. Just like lens elements do not ‘screw’ into anything, they’re held into place inside their barrel with spacers. As someone like you who works with lenses, you should / would be well aware of that.

                  It is NOT an ND filter, they serve completely different purposes and are unrelated.

                  “The test was based on stock condition”

                  Incorrect. There is no such thing as ‘stock’ condition with the Schneider 25mm f0.95. Please read what I have written before. You can order this lens WITH or WITHOUT center filter. It is simply an add-on option when you purchase this lens, whether they will install it or not. You would install one when you’re planning on using this in difficult light situations. You still can, nowadays, ask Schneider to install a center filter. If you don’t specify anything, it won’t come with a center filter installed, so if you want to discuss the whole ‘stock’ thing then no, normally it does not come with a center filter installed. This is a specialized option, only added for motorized aperture control in CCTV cameras.

                  “The sample you showed had very similar circles of confusion to be significant even after your modification of the lens.”

                  Similar, yes, identical, far from it. When you have a center filter installed, the CoC becomes a donut (o), as the aperture determines the shape of your CoC. You could install a heart shaped filter in here and you’d get heart shaped CoC. It’s as simple as that. Without the filter it is simply a round CoC which becomes brighter towards the edge due to it’s spherical correction. The difference would be more clear with bright lights in the darkness instead of overlayed CoC

                  “Difference in contrast was also not significant in your samples.”

                  If you work with lenses, then you’d be aware that adding two air to glass surfaces will ALWAYS reduce contrast. It’s simply how optics work. If you’re going to try and make a point that adding two uncoated air to glass surfaces in the middle of a double gauss lens has no effect on contrast, you’re simply ignorant of basic lens design, as it’s undeniable by fact. If you look at the 100% crop of the yellow bars, you can very clearly see the difference in contrast between the moss.

                  In the end, contrast is always situation dependant. On a gray day with little contrast in the original scene, very little lenses would have difficulty with it. The difference would show better under bright or straight light, as can be seen in this test:

                  http://i.imgur.com/dPYYz7p.jpg

                  Are you going to ignore the difference in contrast and CoC here as well? I figured it’d be clear enough for you to see in which situation the previous test was made and see how much difference there still was for such a situation. This time, from tripod, whilst changing the filter by keeping the lens on the camera, to insure absolutely zero change in anything. Same ISO, same shutterspeed.

                  You work with lenses on a regular basis, apparently. I should NOT have to explain to you how drastic adding two air to glass spaces, uncoated, to a high speed lens will be. You cannot, with a straight face, try to say it will not change the contrast. You should be able to determine that without *any* tests, or any images. It’s simply the way contrast in lens design works.

                  “alignment would change slightly.”

                  With proper shims, no, it will not. If you know how to assemble and reassemble lenses, this will never be an issue. On top of that, double guass lenses are influenced very little by element spacing in the middle of the lens due to their design. Any change on a high quality lens like this will be insignificant, but comparing sharpness was never the goal.

                  “This means you were holding the ND glass in front of the lens rather than installing it back into its original position? ”

                  Of course not, that would be pointless, as having a center filter (NOT ‘ND glass’) in front of your lens would be pointless for CCTV usage. If you understand why center filters are offered for CCTV lenses, you would be well aware of that.

                  “Final word, the comparisons with and without the ND filter hacked off in your samples did not show any significant difference in the characteristics of the circles of “CoC” or bokeh swirl or sharpness.”

                  Then what can I say? If you can just flat out deny the difference when they are so clear? Does it change how incorrect on every other point you are? No. Whether you prefer it with or without is not the debate. Whether you create a flawed test by using a CCTV center filter inside your lens is what we’re talking about.

                  Andrew, your knowledge on lenses disappointing me, and your attitude is both rude and unnecessary. I’m happy to see new brands releasing interesting lenses, I’m not happy with easy advertising through flawed tests. You have a good lens on your hands, and you’d be able to do a proper test regardless, instead of having to shoot your lens on ISO 400 and another on ISO 1000 to skew the results.

                  I’m looking forward to future lenses, but I’m not looking forward to further insinuations about my knowledge on lenses or denying lens design 101.

                  • Andrew

                    Even with proper shims, disassembly and reassembly of a lens affects the lens. I bought an Iscorama from a photographer recently. He told me he had always been using the lens. When I received the lens the image from the Iscorama was horrible. I reconfirmed with the seller again if he had always been using the lens and he confirmed he did. A few months later I learnt he sent the lens for disassembly, cleaning, and reassembly prior to selling the lens online. What happened is even with proper “shims” and “spacers” and screw threads on the optics inside the Iscorama, the lens is no longer performing the way it should in terms of sharpness and focus accuracy. The lens need to be re-calibrated professionally again.

                    • Matt

                      Dear Andrew,

                      The difference in wearing out a lens’ threads and optical construction is not one caused by assembly and disassemble, it’s one caused by deterioration of material, small knocks against the barrel & simply age. The fact that after re-assembly focus had not changed, as well as when shooting from a tripod, makes it clear that there was no error in this procedure. I’m not an amateur when it comes to this.

                      It saddens me to see that you choose which parts of my post to reply to, and leave everything else completely un-commented. You claim that the center filter would not have any noticeable effect on contrast and the CoF, yet my example above clearly illustrates this. I’ve spent a lot of time today assembling this lens to make these tests shots to show you you’re wrong, not for the sake of me being right, but for the sake of making you aware of how this lens works, yet you completely ignore it and change subjects.

                      You keep posting with assumptions that I’m apparently unfamiliar with lens construction as you’ve written before, and making straw man and ad hominem arguments, and it’s tiring. You write to me about what I write, not what you assume or think about me. This will be my last reply, as I’m tired of you dodging my comments in every way you see fit. Goodluck with the lenses and all that.

                  • Andrew

                    Matt,

                    You said:
                    “Just like lens elements do not ‘screw’ into anything, they’re held into place inside their barrel with spacers. As someone like you who works with lenses, you should / would be well aware of that.”

                    As someone who works with lenses should/ would be well aware it is entirely possible for elements to be screwed into place. It is a common practice for lens making. The famous Iscorama lenses use this practice as a starter for your “lens design 101”.

                    • Matt

                      Lenses are made out of glass and have no ‘threads’ ever, that should be a given. As soon as metal contraptions are made to hold them into place, they’re not merely lens elements anymore.

                  • Andrew

                    You said:
                    “It saddens me to see that you choose which parts of my post to reply to, and leave everything else completely un-commented.”

                    It saddens me too. I typed very long posts responding to your entire message and the post did not show up.

                  • Andrew

                    Here is an example of how lens elements have screw threads installed to screw into place rather than being held by shims and spacers for Iscorama lenses made in Germany. Placing lens elements in place with shims and spacers would affect focus and IQ upon reinstallation and requires re-cailbration.

                    http://www.abload.de/img/20121123_170036-1mhur2.jpg

                    • Matt

                      Like I said before, lens elements do *not* have screws. What you link to me is a lens element held together in a metal frame. Of course these have metal screws and are important to the physical construction and spacing of the lens, but to call the complete construction a lens element is somewhat clueless and it surprises me that you’d even think we’d be talking about a complete contraption when we’re talking about elements. Lastly, these are often the worst to re-assemble as you can easily over tighten or screw them in too loosely and loose the exact pin-point ffd of the lens.

                    • John

                      It is what it appears to be. The element has screws and it is screwed into where it should be. Interesting to know that is even possible!

                    • Matt

                      John,

                      Elements do not have screws. Element mounts have screws. Big difference. Almost all lenses use at least some form of screwthread mounted lens elements ever since lenses were invented.

      • Perhaps this may help to distinguish between the Schneider-Kreuznach Xenon used in the test and the “normal” CCTV-version available as well:
        http://m.bhphotovideo.com/mobile/detail?R=46335_REG&amp;

        • Matt

          That helps absolutely nothing, as that’s simply the current release version of the Schneider lens. Schneider changes the styling of it’s lenses every few dozen years.

          • veteran of light

            don’t reply to that Heldge troll. He’s not doing anything but spamming here and writing marketing material for cheap korean brand…SLR Magic that is (with their infamous quality issues).

            Even a baby photographer understands bokeh is dependant on the distance from the subject in focus, yet ONLY with SLR magic he did step one feet closer to the model (thus achieving much smoother bokeh!!!). And said he had camera on tripod.

            What a joke. Heldge is a marketing troll, steer away and don’t feed him.

            • The camera had a fixed position on a tripod and was NOT moved. If you see different portions of the background, this is caused by slight variations in the lenses focal lengths of about 1-2 mm.

              Of course the model varied it’s position slightly and that was commented in the article. She was standing a bit closer in the beginning when the T0.95 shots of the HyperPrime were taken. You are right when you say that this can influence the background blur and the sizes of the circels of confusion as well to a certain degree – but never in that amount that the lenses differ in their characteristics and not the shape of the circles of confusion.

              • Please explain your version of circles of confusion. Your references to your circles of confusion don’t match any understanding I can find on the net.

                • I mean the circles that you get from punctual lightsources in the out-of-focus area of the background. Please see also http://en.wikipedia.org/wiki/Circle_of_confusion

                  The CoC in the center of the SLR Magic HyperPrime CINE T0.95 fullsize image have a size of about 60 pixel, in the Voigtlander Nokton F0.95 fullsize image about 59 pixels. That is no significant difference at all.

                  The impact of the geometrically larger aperture of the HyperPrime CINE is explained below as answer to another comment of Matt regarding the CoF (how he seems to abbreviate them).

      • Andrew

        You said:
        “remove the filter yourself and you get bokeh better then all the other lenses tested here.”

        Removing the filter by hacking the lens will not change the busy bokeh characteristic of the Schneider-Kreuznach Xenon 25mm f/0.95. The link you showed in the other post already proved my point. The bokeh sample you shared is not better than the famous P.Angenieux Paris 25mm f/0.95, The Voightlander Nokton 25mm f/0.95, or the SLR Magic HyperPrime CINE 25mm T0.95 unless you consider busy bokeh as “better”. It may be interesting at best. However, bokeh characteristic may not be part of the design considerations for the Schneider-Kreuznach Xenon 25mm f/0.95

        You said:
        “I have personally tested quite a few 25mm f0.95 lenses, and the Schneider is by far the best both in sharpness and bokeh.”

        Just curious, which were the few 25mm f/0.95 lenses in your tests for sharpness and bokeh?

        • Matt

          You do not need to ‘hack’ the lens to remove the filter. If you don’t think a center filter changes the CoF, then I have lost all trust in your capability in any field regarding lenses.

          My point was never to say the Schneider was better in bokeh then the Voigt or SLR Magic. My point was that the test is flawed. Stop making strawman arguments and read before you reply, because right now this is becoming bothersome. I’m quite sure the Voigt and SLRmagic will be better in regards to bokeh, but that doesn’t mean your test using a center filter is any less flawed. If you don’t want to discuss the issues at hand but instead just want to talk about how your lens is better then mine, then you’re completely missing the point and I have no interest in discussing this any further.

          At the very least I’d have expected a friendly reply from someone who is trying to run a lens business in a niche market, but apparently that is too much to ask.

          • Andrew

            Matt,

            It is considered a “hack” when you do something to a lens that voids the warranty. When you remove the center spot ND filter from the Schneider-Kreuznach lens that is inside the lens you void the warranty. Just like removing the AA filter from a camera sensor would void the camera warranty. As I said, you have “hacked” the lens to take the filter out and in the process the lens no longer performs the way it was intended by the manufacturer. If you send the lens back to Schneider-Kreuznach for filter removal (if that is part of their service) then that will be a different story.

            • Matt

              Andrew,

              I’m sorry, but you have no clue what you’re talking about when it comes to this.

              Your analogy would only be correct if an AA filter in your camera is an optional purchase like a center filter is on a f0.95 Xenon.

              There was plenty of hacking done to *your* copy of the f0.95 Xenon, as they were *NEVER* made with a center filter outside of motorized blocks, as per Schneider literature. The whole reason of using a spot filter is for using motorized aperture control.

              The lens is intended to work WITH and WITHOUT a center filter by the manufacturer. The filter does NOT change the optical formula.

              I have already replied to your previous comment with this silly analogy that is completely off-point and clueless about what a center filter is.

          • Andrew

            All tests can be flawed if you want to see it that way. Helge Hackbarth already said the test did never claim to fullfill scientific laboratory conditions. If Helge Hackbarth planned to do the test that way he might be better off use a mannequin rather than a model and use a controlled light condition in a studio which would defeat a real life test. Some people like to do real life situation tests. Some prefer a control test of charts and fixed objects. No right or wrong here. Just different tastes.

            • Matt

              Real life tests can be done without changing ISO, shutter speed or rapid changing light conditions.

              If you’re going to compare anything, then variables need to be the same. Real life test, lab test, studio test. Doesn’t matter. Variables need to be identical, or there is no comparison to be made.

              You cannot, with a straight face, tell me that you think it’s a good idea to compare one lens shot on ISO 400 and one lens on ISO 1000, and then say one is ‘sharper’.

              Or are you going to tell me that determining which lens is sharper when one is shot on ISO 400 and one is shot on ISO 1000 is a good comparison? Because if so, I have lost absolutely all trust in your ability to play any role in any lens-based company.

              Be honest to yourself instead of trying to crudely talk your way out of a flawed test, because this is just pathetic.

              • Andrew

                The test can never be perfect. Lenses all have different optical behaviour depending on the camera sensor used. A lens may work great on one sensor and may not be as good on another depending on the orientation of the micro lenses relative to the direction of the light rays from the lens.

                http://www.stevehuffphoto.com/2011/10/28/sony-nex-7-with-wide-angle-leica-lenses-a-quick-test-and-gxr-comparison/

                • Matt

                  Sure, if you want to be anal. But there’s a difference between tiny variables and variables that completely change the outcome of the test. Plus, sensors are rather uniform in the current m43rds line up, so that wouldn’t even be an issue.

                  It’s also completely unrelated to do something silly as shooting a test where you vary ISO between lenses.

                  Stop changing the subject and come out for your mistakes.

          • Andrew

            Matt, you were annoyed by people not having an idea of what they are talking about.

            OK…sorry. Maybe I could have started off with a more friendly tone but it was not easy after reading the long thread of arguments you had with Helge Hackbarth. I cannot blame you for having facts wrong due to having less exposure to technical aspects of lens assembly but I am annoyed as well when I see people not having an idea of what they are talking about as well whether intentional or unintentional.

            We made this lens so that there is a cheap affordable fast lens option for mFT users. It is not the best we can do but it is the best we can do at an affordable price. We sent some samples (both pre-production and production) out to testers. Testers are free to complain or praise about or lens. But it just seems whenever we get positive feedback for a lens we made people think it is advertising. We limit the sales of our HyperPrime LM 50mm T0.95 lens to Hong Kong only because of similar things that happened. We hope we do not have to do it for the HyperPrime CINE 25mm T0.95 as well.

            Helge Hackbarth was in Hong Kong for vacation and spent 2 out of his 4 days stay (leaving his friend in the hotel to wait) in Hong Kong to do the test. As a disclaimer, we only had a preproduction lens to lend him and he has no affiliation with us. Lenses he has from our brand were all bought from us. To have interesting content to share with everyone we should at least give him a “thank you for the test” rather than jumping in to complain how flawed the test was or complain he is a troll. This is also why I started jumping in making comments as someone is not being treated fairly for the work being shared with everyone.

            • Matt

              Andrew,

              “I cannot blame you for having facts wrong due to having less exposure to technical aspects of lens assembly”

              Which is your assumption based on what? From the start of your involvement in this discussion I’ve been confused about your ‘level’ of knowledge, calling center filters ND filters, saying they cannot be removed, not understanding the consequences and denying the effect of adding two uncoated air to glass spaces in a lens. You have made a lot of baseless and incorrect statements, about the most simplest of things in lens design, but I’ve refrained myself from speculating about your knowledge, yet every time you reply, you go back to saying I have the facts wrong, without any proof whatsoever.

              Once again, I have no complaints about your lens whatsoever, it looks like a fine lens. for a good price.I have an issue with the misrepresentation of the Schneider lens and the follow-ups coming from Helge and you making inaccurate claims about this lens and it’s center filter, as well as other users claiming this is the ‘cine’ version of said lens, and the flaws in this tests which lead to incorrect assumptions or conclusions.

              I have never called anyone a troll, so I would be happy if you could keep your straw man arguments with you. Whether he is affiliated or not is not to the point. To the point is simply that if you publish articles such as these, people will take them in as they are; comparisons. You compare sharpness, you compare bokeh, you compare contrast and so forth.

              Whilst admitting that things like sharpness could not be compared, sharpness was still compared, and so forth. This test could have, by following some extremely basic rules and testing principles, be a great article that would have promoted your lens greatly, more than likely. Nothing about real life tests mean that it’s impossible to keep the same exposure settings or ISO, or use the same focus distance for each lens. Now however, it’s simply a flawed test which should not have been published in it’s current state. It will only cause incorrect assumptions and conclusions, it’s as simple as that.

              If you REALLY think comparing sharpness from ISO 400 to a shot made on ISO 1000, then by all means continue, but don’t expect anyone to take you serious.

              • Nick

                to be fair, technically Veteran of light called Helge troll but the two of you were ganging up on him.

                • Matt

                  What veteran of light said is what he said, I am unaffiliated with him nor do I approve of his offensive attitude, even if we share the same opinion regarding the test.

                  I cannot control someone else’s behaviour.

  • Mister_Roboto

    SLR Magic has really been on a roll, and given pretty good quality for the money… now only if they’d change their name to something less dumb 😉

    • adaptor-or-die

      SLRM = slurm

      • Mister_Roboto

        All hail the hypnotoad.

    • +1
      Mirrorlessmagic :D.

      • Twaddler Belafonte

        Pretty sure David Copperfield already owns the rights to that name.

    • Bob B.

      How about View-Camera Fancy!?

  • kavat

    Voigt killer! The SLR 35mm t0.95 is already very good but this 25mm is trully impressive!Nice 3d rendering @ f0.95. Good work SLR MAGIC. My next lens!

  • Juan R B

    Pirelli racetrack tyres are great on Daewoo Leganza.

    • Anonymous

      Hey Juan, you got it wrong.

    • Nothing, absolutely nothing, is good on a Daewoo Leganza. Solid gold paint, pirelli, anything above “yuk” level is not good on a Daewoo Leganza. Yes, I did drive one. Once.

      • ISO 1638400

        Ah, but rebrand Daewoo as a Chevrolet (US) or Holden (Australia) vehicle and they can’t make them fast enough. The majority of the buying market is none the wiser. Thus the Leganza, Lanos, Lacetti, Kalos and who knows what lives on! Just goes to show, you can’t kill a bad car off, especially if it’s a Daewoo.

        • The one I drove was before GM sank money into them. It was like driving a car where no nuts had been put on any bolts. To get round a corner the steering wheel had to be turned almost the street before the one that was wanted.

    • c0ldc0ne

      If you pronounce “Juan R B” quickly, it sounds like “wannabe”.

  • PannyBoom

    should i get Nokton 17.5mm f/0.95 or SLR magic 25mm f/.095, 35mm or 50mm which is the best focal length to start with?

    • matt jones

      In terms of ability to focus at f0.95 the 17.5mm is a lot easier than 25 😉

  • Bob

    True that. With more depth of field in a wider angle lens it is easier to focus. It defeats the purpose of getting a 0.95 lens for the GH3. I would say 25mm is the widest you should use for 0.95 for a cinematic look.

    • MarcoSartoriPhoto

      True, BUT maybe he’s not interested in thin DOF. I own a couple of f0.95 lenses (50mm SLR AND 17.5mm Voigty) and I use them for low light portraits and landscaping.

      • Bob

        With reported strong field curvature of 17.5mm 0.95 the Oly 17mm might be better for landscape needs

        • true homer

          Curvature can be fixed, extreme softness can’t

          • Bob

            You cannot fix curvature. When a far away subject is in focus with near subjects or image plane is curved you cannot fix it. You are confused between distortion that can be fixed with field curvature that cannot be fixed.

        • MarcoSartoriPhoto

          Bob, you are right, but Voigty is better when you need to take photos at night: I used Oly 17mm f1.8 for the whole day, today, but I switched to the 17.5mm f0.95 from 7pm ’till 10. Stopped down to f1.2 and f1.4. I don’t use super fast lenses because I like paper thin DOF (though sometimes it can be funny), but simply because I love to take pics in dim light (without tripods), and I thought PannyBoom would use it in the same way.

  • Ulli

    they all look good, depending on your taste for the rendering of the oof area. The SLR Magic has the best price/performance ratio, but the two old lenses might appeal because of the non standard bokeh.

    • Bob

      I quite like the swirly bokeh of the Angenieux. Maybe it vignetting will be better with 16:9 crop? The low contrast is a bummer though. But for filming the Hyperprime looks like the best option so far.

      • Matt

        I don’t know about the Angenieux, but the Schneider does not vignette on the GH3 when filming in 16:9.

  • rUY

    Is there a way to see the Exif of the images? I am curious of T0.95. Actually, I doubt of that for a long while.

    • Bob

      The EXIF would not be helpful due to natural light conditions that is changing continuously due to position of the sun and clouds relative to the position of the sun. I might have worded it more complicated than it needs to be but im sure some would know what I mean. The wide open shots of the T0.95 has more bokeh than the three f/0.95 varients. Due to round bokeh we can confirm they were all shot wide open. This confirms T0.95 from thinner depth of field in the wide open shots.

      • rUY

        The distance between the model and the camera is not the same for 2 images. so, the Dof is not really the same. if they are using tripod and everything remain constant, then, the SLR is not really 25mm lens at all or the Voigtlander should be less than 25mm.

        • Nick

          rUY you are absolutely right. Manufacturers such as Leica, as an example, calls their 51.6mm and 52.3mm lenses a 50. Canon, as another example, calls their 54mm lenses a 50 as well. This affects Dof as well. Different testing equiptment yields different results so specs may vary. Benchmark software testing sensors can give a top result for one test and a worse result with another test so there is really no fixed standard.

    • Bob

      The size of the circles of confusion are much larger for the HyperPrime, Second is Voight, Tie on third and fourth is Schneider and Angenieux. I guess this confirms entry pupil size confirming T0.95

      • Surely the size of the circles of confusion would be the same for all lenses, as the size is determined by standard visualisation of an image and determined by sensor size related to the image size and viewing distance.
        With these images being a maximum of 1600 x 1200 and for screen display the issue of a circle of confusion for sharpness would not really be relevant.

        • JimD, you seem to have a different understanding of circles of confusion (CoC). Please see the explanation I gave some answers above.

          The CoC of the punctual lightsources in the center of the out-of-focus region are about 60 pixels in diameter in the fullsize image of the HyperPrime CINE and about 59 pixels for the Nokton.

          • The circle of confusion is the area outside the area of true focus and determines the visualisation of DOF. As the area out of focus moves further from the point of true focus (in either direction) the CoC also increases. Being visualised as the scale of blur increasing as distance increases.
            However for people this is a perceived change and the visual result is determined by the size of the viewed image and viewing distance. To all intents and purposes images on web sites display no real DoF as the images are the incorrect size to see the effect of CoC.
            Yes, they may have sharp objects of focus and they may have blur for out of focus areas. But they do not represent the CoC in action at all. Restricting them to 1600 x 1200 size does nothing for the visualisation of CoC and therefore the real visualisation of the DOF .

            • That’s why I also added 100% crops to the article showing the out-of-focus blur and the circels of confusion of the punctual background highlights with actual pixels.

              I’m afraid, you still did not read the Wiki article about the definition of the CoC.

              • There is nothing in the article that contradicts what I have said here or above. Circles of confusion refers to all out of focus areas not just bright lights, light from windows or even the trees. Circle of confusion would be seen more as the fuzzy area around the light rather than the light itself, if you are using a light as a reference. The shape of the resulting light may well be out of round but that has little to do with circles of confusion and more likely to be other properties of the lens or coating due to angles of incidence.

  • Matt

    Horribly skewed test due to using a CCTV version of the Schneider lens with a centre filter in place. Not only does it actually limit it’s speed, but it’s not meant to increase image quality, it’s meant to increase speed in exposure adjustment on auto exposure CCTV cameras. You can take it out, or you can use a non-CCTV version of this lens which has absolutely beautiful bokeh.

    Secondly, the ISO changes from shot to shot! How can you compare shots if one shot is ISO 1000 and the other is 400? Of course the 400 shot is going to look sharper!

    I’m sorry, but if people are going to ‘publish’ test results then at least think things through before you actually perform a test.

    • true homer

      Deja vu

    • Bob

      the lens in the test is the Cinema version. ND is not removable.

      The CCTV version does not have built-in ND filter and has a different build. Compare the lens used in the test with this CCTV version and you know what I mean.

      http://www.avsupply.com/Schneider-Optics/xenon-21-012101.php#.UTHxUPoazCQ

      • Matt

        This is complete bull shit. The center filter is DESIGNED for CCTV usage. It’s a black dot in the middle of the lens that will increase light change with small aperture changes.

        There IS no cinema version of this lens in C-mount whatsoever, simply CCTV lenses with and without center filter, and those build in motor blocks.

        The filter is not part of the optical formula, and both are identical bar the filter.

        What you linked me to is simply the modern release of this lens. Schneider re-tools their lenses every dozen years.

        • Matt

          Accidental double reply since bullshit is apparently filtered / makes the comment go to moderation, my bad.

          I don’t mean to be offensive, but when people promote products and talk complete nonsense about the competition, I’m not too happy and it just doesn’t feel good or correct.

      • Matt

        This is simply incorrect. The center filter is DESIGNED for CCTV usage. It’s a black dot in the middle of the lens that will increase light change with small aperture changes.

        There IS no cinema version of this lens in C-mount whatsoever, simply CCTV lenses with and without center filter, and those build in motor blocks.

        The filter is not part of the optical formula, and both are identical bar the filter.

        What you linked me to is simply the modern release of this lens. Schneider re-tools their lenses every dozen years.

        • David

          I dont think Schneider just retool their lenses and add ND with no changes and still call the lens 0.95. If that is true their lens offering would no longer be interesting for the NEX and NX. They just joined micro four thirds and they will retool again? complete BS that is why your comments need moderation.

          • Matt

            What are you talking about? Schneider retools the physical appearance of the lens and the coating. The ND filter was ALWAYS an optional choice when you purchased this lens.

            The Xenon 25mm f0.95 is NOT a lens designed for Micro 4/3rds or Nex. It is an industrial CCTV lens where a center filter might be needed.

            Please, don’t reply if you don’t know what you’re talking about in the slightest. f0.95 says nothing about how much light reaches the sensor, and you should know that. It is basic lens knowledge.

    • Matt, you are wrong (again). Please find my answer to your earlier post.

      • Matt

        I am most certainly not. Please see my reply to your reply.

  • John

    Also, the model changes position, so focus distance changes, and the distance of the camera seems slightly closer in the SLR Magic test.

    • John, the camera position was fixed, but model had a slightly closer position in the T0.95 series shots of the HyperPrime. This was mentioned in the test and may suggest better sharpness in the 100% crops of her face but it does not affect quality of bokeh. The test says that these examples with the slighty varying conditions can not be used as a reference for sharpness but it can be used to determine the different lens characters and bokeh qualities.

      Propably there will follow a series of sharpness comparisons with the final production version of the HyperPrime (which was not used in the test) and the Nokton with identical lighting and a static object.

      • rUY

        yeap. I agree too. there might be a chance that it is one of those not 25mm too. the comparison does not really that meaningful to me.

      • veteran of light

        indeed it affects quality of bokeh. What do you know about photography by the way? Or just spamming here (as I said above) are we?

        Go throw your marketing elsewhere.

        P.S. Model DID NOT STEP closer (she’s still touching the rails). YOU DID STEP CLOSER for the shot. Get a clue kid.

        Admin should delete this shameless marketing. And I will wait at least for Steve Huff review before doing 2nd mistake with SLR Magic…

        • MarcoSartoriPhoto

          Veteran, what was your first mistake with SLR? Just curious.

        • Veteran, have you been there? If you claim that I changed camera position, you are simply wrong and everyone taking a closer look can verify this. A comment regarding the slightly varying position of the model was given above to your previous comment and in the article as well.

    • Eric

      Yeah the position of the camera might be the same but the model is still closer so the focus distance is indeed closer. And we all know part of the bokeh magnitude relies on focus distance. Hence why the slr magic shot’s bokeh looks smoother than it should.

      Of course that is t to say that the slr magics bokeh isn’t good. Just that the test shot is biased for slr magic

      • Bob

        Eric, what do you usually shoot?

        It may make a difference for shots one foot away from the subject but not for shots far enough for portrait shots.

        • Matt

          At this distance, it already matters a lot.

          You can tell because the CoF changes from lens to lens, which is physically / optically impossible without changing the focus.

          • David

            Fanboys are upset about the “Voight Killer”

            Of course the test is not fair. Give them what the Fanboys want to hear then it is a fair test.

            • Matt

              Voigt fan? Are you throwing strawman arguments around without having a clue?

              I do not own the Voigtlander lens nor do I care anything for which is best. For all I know the SLR Magic is indeed the better lens.

              However, everyone here is correct when they say the test is very flawed. If you really think comparing two lenses while one is shot at ISO 400 and one is shot at ISO 1000, at different focus distances, with different lighting conditions, then something’s simply wrong with you.

              You cannot make such crude comparisons and think your conclusion from those actually matter.

              • Matt, I refuse your allegation, that this test is flawed. You find my explanation already above.

                • Matt

                  And I’ll say once again, if you think testing lenses whilst one is shot on ISO 400 and one is shot at ISO 1000 and one is focused closed, one is focused farther, one is shot when it’s darker, one when it’s shot before that time, then you simply cannot compare them. End of story.

                  You’re crediting sharpness to ISO changes and bokeh changes to light changes. How much does it influence your tests? Look at the CoF of these lenses that all share the same aperture. The CoF shouldn’t change, yet it does. Your test is flawed.

          • Matt, Eric,
            the circles of confusion (I think this is what you mean with CoF) in the center have a size of about 60 pixel in the SLR Magic T0.95 fullsize image and about 59 pixels in the Voigtlander F0.95 fullsize image so they have nearly identical sizes. This proofs that the slighty varying distance of the model (who was standing about 3m away from the camera) had no significant impact on the background blur.

            You also should know already that SLR Magic CINE lenses have geometrically a larger aperture than the given T-stop. The CINE 50mm T0.95 has a geometrical aperture of about F0.92 and the 35mm T0.95 of about F0.9. I guess that the 25 T0.95 will have about F0.92 as well.

        • Ulli

          unless the slrmagic is optimized for defocused rendering, the big difference in oof looks should be blamed to the model standing closer i guess.

        • Ulli

          hmmmmm….unless the slrmagic is optimized for defocused rendering, the big difference in oof looks should be blamed to the model standing closer.

          • Matt

            “slrmagic is optimized for defocused rendering”

            Which is optically impossible. The CoF does not change.

            • Ulli

              I mean like the defocus lenses from nikon and canon, there is a certain control for back and front out of focus behavior. So it is possible to change the looks of the oof area.

      • rUY

        you are saying something that I hesitate to say, but I kinda agree with you.

  • BdV

    I visited noktor.com before, and there may also be an slrmagic.com site, but today there is no page that works. Even the slrmagic ebayshop is abandoned: 0 products. How weird is that?

  • Robbie

    nice pseudo-review ad for SLR Magic

  • Alfred

    would be nice if there is a Panasonic 25mm 1.4 m4/3 included in the test as the baseline as most people know how the 25mm 1.4 performs

    • David

      would be pointless to incluse the 25mm f/1.4 as it is a much slower lens and does not suffer from characteristics of ultra fast lenses. For example Summicrons have little to non existant CA compared to Summilux with CA. Summilux has little CA compared to Noctilux with Lots of CA and field curvature. It is only a meaningful test to compare lenses of the same speed without stopped down performance unless you want to count the number of aperture blaces when stopped down.

    • Alfred, I understand your whish as many people (including me) asked themselves exactly this question when the 25 Summilux became available: Nokton 25/0.95 or Panasonic/Leica 25/1.4.

      Speaking for me, I sold the Nokton after I got the “PanaLeica” because at F1.4 I prefered the look of the Summilux and the IQ of the Nokton degraded too much at open aperture for my taste but the CoC are no circles anymore, when the Nokton is stopped down. The AF was another advantage for the PanaLeica.

      The test intended to compare pure 0.95 designs.

  • I think the images speak for themselves. To my eye, the Hyperprime delivers as sharp an image with as pleasing a bokeh as can be expected from any currently available top-flight 25mm f/0.95 M4/3 lens.

    I am convinced the images are representative, because such slight differences in methodology (as can be expected from normal use) would not be expected to change the sharpness and bokeh in the manner and to the degree that the pictures demonstrate.

    Remember also that the Hyperprime is sold at a substantially lower price point than the other lenses in the comparison. It’s definitely a lens worth consideration by Micro Four-Thirds users shooting both stills and video. It’s especially compelling for video makers, offering built in follow-focus gearing.

  • true homer

    what happened? did matt win?

    • Andrew

      true homer, there is no winning or loosing but it seems Matt is a little bit confused with his understanding on lenses. I told him to email me to continue the discussion. Anyone else unsure can also email me for further discussion but this forum is not a suitable place for the discussion as it is already off the topic. If you wish you can also email me if you unsure about something.

    • Matt

      There’s no discussion as Andrew refuses to reply to my arguments and insists on calling me confused instead.

      You cannot compare sharpness between two lenses when you shoot one on ISO 400 and one on ISO 1000 with different shutter speeds on a non-static object.

      Andrew and Helge are apparently fine with this and see no issue in using a grainy ISO for one when using a smooth ISO for his own lens.

      There is a huge difference in optical quality, both bokeh and contrast wise, with a Schneider without center filter:

      http://i.imgur.com/dPYYz7p.jpg

      Andrew does not see any difference in contrast and CoF here.

      Andrew doesn’t think adding a glass filter with no coating in the middle of your lens with a big black dot on it changes your bokeh or contrast.

      All he does however, is reply with “Matt is a little bit confused” and refuses to reply to any of the points made.

      He talks about that I ‘hacked’ my lens, but ignores the fact that his Schneider lens once looked like this; http://farm8.static.flickr.com/7209/7053189781_846d583947_m.jpg

      As ONLY Schneider lenses built in motorized contraptions came with this filter. He didn’t see any ‘hacking’ in remounting this lens apparently, as it requires the removal and reassembly of the rear mount of the lens to remove this motorized contraption. Yet I’m hacking my lens and have no clue what I’m talking about.

      Sorry Andrew, but you’re just trying to save face by calling me clueless ( http://en.wikipedia.org/wiki/Ad_hominem ) You can either reply to my arguments made or not reply at all, but you don’t go and insult me and insinuate that I am clueless without replying to the arguments made, then calling me a ‘little confused’.

      I’m offended by your attitude towards your target group for your OWN business. At least Helge replies to my points and doesn’t avoid the obvious or call me ‘a little bit confused’. That’s simply rude.

      Feel free to e-mail him, but you’re bound to end up with more marketing talk and avoiding the obvious flaws in this test. Form your own opinion based on what you read.

      • admin

        Matt!

        I talked with both Helge and Andrew. They epxlained me how they tested the lens and they did it correctly.

        • Matt

          Admin!

          They wrote in their original article how they tested the lens, and they did not do it correctly.

          Admin, please tell me, how does one compare sharpness when one is shot at ISO 400 and one is shot at ISO 1250?

          Mind you, that is a rhetorical question. Everyone with a right mind would know that is pointless. That you are the admin of this site does not make you anymore knowledgeable on lenses then anyone else, I’m afraid.

          • admin

            Helge answered you about the ISO story. What’s wrong with his answer?

            • Matt

              And I have replied to him.

              Once again, answer the question for yourself: Can you compare sharpness when you change ISO from 400 to 1250?

              Can you compare the amount of light that comes in from a lens, when the light is rapidly changing outside?

              Answer these questions for yourself in honesty, and you’ll understand why people call the test flawed. If someone is to show you one photo shot at ISO 400 with one lens, and one shot at ISO 1250 with another lens, and say “This one’s sharper! The other’s fuzzy.” Would you say “Ah, you are correct!” Or would you say “But that one’s shot on ISO 1250.”?

              Most people except Andrew, Helge and you seem to have answered that question for themselves quite easily.

        • Administrator

          Are you really the admin? …of this site? I’m certanly not the administrator.

          • admin

            look at my english and you will recognize me 😉

      • Andrew

        Earlier you mentioned elements are not screwed into place as well. I am sure many here also think it is not possible for lens elements to be screwed into place. I have dissembled my Iscorama lens to show you how it is possible for lens elements to be screwed into place.

        http://www.abload.de/img/20121123_170036-1mhur2.jpg

        It is possible for the elements to be screwed off. As you mentioned earlier this requires some knowledge but I have to hack into my lens to do so and focus accuracy and alignment is affected. Warranty is of course void after I do so but either way the warranty had been past since 30 years ago.

        • Matt

          And I’ve already replied to you that no sane person with any knowledge on lenses would call a mounting system for the lens element a lens element on itself. The lens element is what is held inside that piece of metal.

          It saddens me to see you keep trying to point out that I’m apparently wrong, yet miss the fundamental basics of lenses. In fact, considering your business, it’s simply shocking and it makes me wonder how much influence or know-how you actually use in ‘creating’ lenses, instead of simply doing the business side of things.

  • RFC1925

    While the comparison test was flawed, the positive side is the shortcomings prompted a very interesting discussion here. Definitely going to start checking the comments section in this site more often. Cheers!

  • I have some 17mm f0.95 Schneider Kreuznach with internal filter.
    How can you remove the filter?

  • Mark Krause

    Dear Matt, I just purchased a Schneider Variogon 1,8 / 12,5-75 CP 123, with (unfortunately) the spotfilter build in and all the motor elements still attached.
    I guess I have no problem to remove the motor components, but am afraid I do not know how to remove the spotfilter. Can you give me a hint on how to do this? I have never even opened any lens, so should I try it anyway? – Many thanks in advance! looking forward to hearing from you. 🙂

  • C-Mount

    Join the other 2,600 people for a lively discussion of C-Mount lenses.

    https://www.facebook.com/groups/cmountm43/

By continuing to use the site, you agree to the use of cookies. more information

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this.

What are Cookies?
A cookie is a small file of letters and numbers that is stored in a temporary location on your computer to allow our website to distinguish you from other users of the website. If you don't want to accept cookies, you'll still be able to browse the site and use it for research purposes. Most web browsers have cookies enabled, but at the bottom of this page you can see how to disable cookies. Please note that cookies can't harm your computer. We don't store personally identifiable information in the cookies, but we do use encrypted information gathered from them to help provide you with a good experience when you browse our website and also allow us to improve our site. You can watch a simple video from Google to find more information about cookies.

Cookies used by our Website
The 43rumors website, 43rumors.com, uses the following cookies for the collection of website usage statistics and to ensure that we can . These are anonymous and temporary. By using our website, you agree that we may place these types of cookies on your device.
Read how Google uses data when you use our partners' sites or apps: http://www.google.com/intl/en/policies/privacy/partners/
Google Analytics Cookie Usage on Websites: https://developers.google.com/analytics/devguides/collection/analyticsjs/cookie-usage?csw=1#cookiesSet Addthis cookies: http://www.addthis.com/privacy.
Disqus cookies: https://help.disqus.com/customer/portal/articles/466235-use-of-cookies.
Vimeo cookies: http://vimeo.com/privacy.
Youtube cookies: https://support.google.com/youtube/answer/171780?hl=en-GB

Disabling/Enabling Cookies
You have the ability to accept or decline cookies by modifying the settings in your browser. Please note however that by deleting our cookies or disabling future cookies you may not be able to access certain areas or features of our site. For information about how to disable cookies in your browser please visit the About Cookies website.

Close