Personal View site logo
SLR Magic 2x ANAMORPHIC lens
  • 804 Replies sorted by
  • @slrmagic If you don't want to do 1.5x, then at least stick with the established 1.33x standard. "1.35x" is incompatible with modern digital standards. The 16:9 format is 1920x1080 and the new "21:9" cinema format is 2560x1080, which is a horizontal stretch of precisely 1.33x. A stretched 1.35x adapter would produce a 2592x1080 frame, which should technically then be cropped to 2560x1080. If you instead fudge it to 2560x1080, the pixels will not turn out square. It took decades to finally enact world-wide square-pixel standards, please don't muck it up now!

  • @plasmasmp I am using the LA7200. From the shape of the ovals on the LA7200 I think I just gave away part of the specs of our prototype lens!

    But it is a good finding. Shape of ovals become increasingly dramatic as FL increases. The prototype lenses I am using are on the wide end (at least wider than 50mm). It is entirely possible even if we have 1.5X with wide angle we will:

    1) Not have obvious anamorphic oval bokeh 2) If we have 2 solutions the oval shapes will not match due to wide vs longer focal lengths even with the same squeeze factor.

    If the ovals will not be obvious when we provide a wide anamorphic solution we can conclude 1.35X is better than 1.5X as it is not worth the post processing time with extra work if there is no obvious effect. But I read somewhere a suggestion is if there no oval bokeh then users can just crop wide angle lens to 2.4:1 ratio and there is no need for anamorphic lens as the wide aspect does not seem to be the concern.

    If we have to give a rating in a particular order for anamorphic feel. What would it be"

    a) Horizontal Flares b) Oval Bokeh c) Wide aspect

    I have to admit it is a bit discouraging to read here that anamorphic is about Horizontal flares that can be added in post and Wide Angle aspect that can be cropped anyway (stimulating a telephoto anamorphic lens aspect ratio). It seems like oval bokeh is the only factor that cannot be done in post. But this kind of defeats what we have discussed in the past 5 pages of making a wide anamorphic as it will reduce/remove anamorphic oval effect.

  • 1.35x is the right way to go. You would be the only other company other than Hawk offering a solution without resolution loss. $3000 is not a lot to ask for a high quality lens without chromatic aberration or vignette on the far edges of the lens.

  • @JosiahSjostrom, I have been up all night hunting for anamorphic sample videos online. Although technical data are omitted often there is a wealth of information out there on anamorphic on DSLRs. I started noticing the trends and the " complete anamorphic look " you are referring to. Aside from the proportions ( occasionally 16x9 stretched X2 ) the bokeh becomes much more impressive with higher stretch factors and generally the visual elements are more pronounced in X2 squeezed footage. That said I still believe the X2 squeeze and 16x9 image are a bad match. Either we have to crop down to 2.66:1 or better 2.35:1 by killing masses of innocent pixels or watch the footage in 3.55:1 thinking WTF throughout the film. GH2 + 4:3 patch armed with the X2 squeeze is the ultimate champion in my mind at this point. Thank you Vitaliy again. My both GH2s are being used as a B-Roll this week. I'm not sure if I could put the 4:3 patch on GH1s. I will post some tests with Kowa-X2 B&H tonight if GH1 accepts the patch.

  • why the heck do so many people vote for 2x, what is it that you HAVE to shoot at such a ratio?

    EDIT: I'd love 1.35x but you have to do what people want I guess

  • http://www.easypolls.net/poll.html?p=50477872e4b0ce84916c0ecc

    1.35X for 2.4:1 has a low vote now. It is the ratio we want to have 2.4:1 without cropping.

    We see 1.5X being more popular but 1.5X and 2X is already available and many are emailing us saying 1.35X can make this new anamorphic lens the only one supporting 2.4:1 without cropping.

    From our point of view we do not see HD DSLR or APS-C HD and mFT HD doing 4:3 mode anytime soon.

    1) All Plasma/LED TV's are going from 4:3 (do they still exist?) to 16:9 so for consumer purposes HD mode will be in 16:9 mode to accommodate consumer flat screen TV playback.

    2) Many Notebooks are moving to 16:9 format too to support the video modes

    3) DSLR will always shoot in 3:2 and it makes sense the full width of the sensor is used so it should not be cropped (well, who knows...the GH2 with multi aspect ratio and mFT cameras with 1:1 mode proved it is an possible offering.) But 16:9 will stick around.

    With all this in mind it is safe to bet 16:9 IS going to be the new video standard for quite some time. 1.35X is probably the new format and we have to accept the new anamorphic look so we get all details and not have to crop and do the extra work. I hope people who stress 1.35X is the right way to go can discuss on the forum rather than just sending us private emails with their thoughts and concerns!

    Kind rgds., Andrew

  • @slrmagic - "I am suggesting $1200-$3000 here because so far from the whole thread people are suggesting they want a large anamorphic (larger coverage than the Isco adapter) for use with wide angle lenses as well."

    Yes, and here's what I see as the happy medium:

    1. 1.5x (1.33x is too mild; 2x is too wide)

    2. 24mm MFT (35mm APS-C) image circle

    3. LOMO square-front design, rear-element focus

    4. Plastic shell with 15mm rails mount

    With a 1.5x squeeze, you offer an affordable, interchangeable alternative to the highly-prized ISCO anamorphics. With 24mm MFT coverage, you support the popular Nokton 25mm f0.95. With the LOMO square-front design, you minimize the size and weight of glass you need to fabricate. The square front can also interface well with hoods and matte boxes. With the simple rear-element focus mechanism, you eliminate the need for close-up diopters. With the plastic shell, you minimize cost and weight.

  • @slrmagic I have other samples from the la7200 with similar bokeh. I dont know where the still is, but the background of my twitter is from the la7200 + a 50mm nikkor https://twitter.com/eyepatch_films

    long focal length + MFD + wide open produces the the most elongated bokeh. The sample above truly is 1.33x. The bolex is 1.33 and slowly gradual to 1.5x at 2m.

    EDIT: found it. attached is la7200 + nikkor 50mm mfd f1.8 lights creating bokeh are are infinity

    415687_802354253839_710697539_o.jpg
    1920 x 811 - 307K
  • @plasmasmp Your's is a good example of what 1.5X ovals look like. I am still confused how you refer to 1.33X. You convert it to 2.35:1 ratio but your ovals are from 1.5X lens.

    As we all know 1.35X gives the 2.4:1 ratio but it does not give the obvious oval shape as some suggested. I have attached a sample of 1.33X for comparison to show the point some others are pointing out "the anamorphic oval bokeh feel". T

    Lens used is one of our prototype lenses. Confidential for now =) Btw the photo is just a photo I took this evening and is not meant to be artistic at all! Just for reference for 1.33X Anamorphic bokeh.

    Kind rgds., Andrew

    P.S. I typed prototype incorrectly but I am unable to change the uploaded photo for some reasons >.<

    P1180344_anamorphic.jpg
    2549 x 1080 - 241K
  • Iscoramas are 1.5x squeeze. Various diopters on the la7200 can have that effect

  • @plasmasmp You have to calculate for the relative effect from a 1.35x, then you are down to 1.15-1.20x calculated in the same manner with a different overall squeeze factor.

  • @RRRR by looking at circles and correcting the aspect ratio. The same thing happens when you use the double element .5 diopter on the Isco 36. I have both right here. I would be happy to test an SLRmagic anamorphic on the Scarlet as well.

  • @sicovdplas

    That's not realistic, you're asking too much for too little money.

  • @plasmasmp

    I wonder how you come to such a conclusion? How can 1.5x suddenly become 1.33x?

  • @slrmagic I also think you are right to doubt 1.33x/1.35x.. If there is no / little anamorphic charachteristics you can easily crop any normal lens to desired aspect ratio.

    People keep asking for really cheap adapters and if somebody can combine that with good quality = great, but a cheap and bad adapter is not of much use to anyone. Might be fun for a little while and then become a dust-gatherer / dumped price on second hand market.

  • 1.33x bokeh. This is from the bolex 8/19/1.5x at .5 meters. At this focus distance, the squeeze factor is 1.33x

    LOOK AT ATTACHED FILE. We all should want a 1.35x lens!

    1.33.jpg
    1920 x 820 - 215K
  • @slrmagic I want to make clear that I have nothing against squarefronts. More difficult to use with mattebox perhaps but it makes little difference if the lens itself is worth it.

    I don´t think the IQ differences between lomo round- and squarefronts are because of the different shapes.

    I think also it would be sensible to look into the idea from @johnnym of prisms to see if such a lens can be done cost-effectively and with high quality.

    And regarding price there are requests for it to be cheap but everyone who cannot afford to spend 1.200 - 3000 usd on a lens (or up to 5000 on a set) still most likely will be in the position to rent such a lens /set if they are widely available (which this price wouldn´t prohibit).

  • 2.35 or 2.66 ratio please!

    I would pay 500-1k for a lens that has GOOD sharpness and also a cinematic softness.

  • @RRRR @jackdoerner

    Square fronts have advantage over round fronts. The same round front lens need to be made larger than a square front equivalent. Square front has less vignetting than a round front of the same size (If Diameter of round front is same as width or height of a square front). Quality have to do with design and materials used.

    Kind rgds., Andrew

  • @johnnym

    Yes and No. I was typing fast but I did not type what I did by mistake. Let me try again as there is a common misconception when using a lens. Lenses create an image circle. The image circle can maybe be as large as a fullframe sensor. However, it is possible the intended use of the lens is NOT for the entire image circle. The fact that it has the coverage does not mean it is designed for that size. It is possible it has a large coverage but it is meant for use on 4/3 image sensor only. One example is the LA7200 was designed for 1/3" image sensor but we are using it here for 4/3 and often times APS-C and Full Frame sensor as well. The side effects are softness as many noticed already or vignetting. If the lens is designed for a larger coverage it would cost more. For example the Isco lenses. Some already know you cannot use wide angle lenses with the Isco. To use with wide angle you need an even larger adapter. If we are designing one for mFT it needs to have acceptable boarder performance and no vignetting. We cannot have a lens that covers mFT then is soft or have extreme field curvature for example. This is why it will cost a lot.

    I guess we have many tech guys here so an example is we can use patch cable Cat 5e cable for gigabit ethernet but it is not the same as using Cat 6 cable for gigabit ethernet and Cat 6 will cost more although the can both be used for Gigabit ethernet. A comparison is the LA7200 or cheaper anamorphic designed for small image sensors are Cat 5e cables but we are planning on designing an affordable Cat 6 cable and not an affordable Cat 5e or 5 cable. Did I just make any sense?

    Further, I did some tests and understand why many are suggesting to make a 1.5X or 2X adapter. I have a 1.33X and I am not getting the anamorphic ovals many want. I have to say I am also getting lots of emails (they did not vote in the poll for some reasons) that we should ignore all the requests for 1.5x or 2X because the new format now is 16:9 for HD video and 1.35X will give the true 2.4:1. It kind of put us in an akward position and I hope people can continue the discussion here on why 1.35x to give 2.4:1 is more correct for this age rather than sticking to 1.5x and 2x that is already available on the market. 1.35X will give the wide anamorphic look but we would not expect to see the obvious ovals.

    I am suggesting $1200-$3000 here because so far from the whole thread people are suggesting they want a large anamorphic (larger coverage than the Isco adapter) for use with wide angle lenses as well. Larger coverage means more complex design and more exotic materials with an exponential raise in manufacturing cost. With that regard if such a lens exists $3000 is not a high asking price. If there is an Isco adapter that can be used with wide angle lenses I am sure you will not say the price is close to the Isco. If it is still affordable with such specifications that is another story. We can then make compromise to reduce the cost by having smaller coverage and not support wide angle or have smaller max aperture etc. We open this thread so we find the balance between needs/wants with offering/price consideration.

    Can someone with a GH2 post some 1.33X, 1.5X and 2X anamorphic stills in 4:3 and 16:9 that can display the anamorphic oval effect for everyone to see? It seems many of the samples we see online are already edited in post by cropping or de-squeeze etc and something original would help everyone to decide what is best if we can have a new option and possibly a new format?

    Kind rgds., Andrew

  • @slrmagic

    If you will pursue the idea of an anamorphic adapter, you're welcome to study the prisms. Saying that prisms will give you vignetting, because that's what you saw in the video, is a short-sighted remark in my opinion. You could equally develop an anamorphic adapter with prisms that covers more of the sensor. I don't understand why you made this remark. I'll take it you were probably in a hurry, cause it doesn't make sense. I also have a Kowa 16H. It also needs a 50mm taking lens at least to cover the GH2's sensor, which is effectively a 100mm in 35mm terms.

    So if you're looking into an anamorphic attachment, don't write off prisms just yet.

    As for price, to ask 3000$ for such an attachment, it would have to be exceptional quality, cause for that price you can have an Isco. And most users on this forum wouldn't be able to buy it, i guess, so why ask here? I think even 1200$ is high. You could also design a set of lenses that competes with Hawk, but we know what price range that is. I thought the whole idea was to develop something of good quality at reasonable cost.

  • @jackdoerner, I´m not sure how much age plays into the equation but judging from the components it seems like the squarefronts were one adapter design + excisting taking lenses / optical blocks. It doesn´t seem like the taking lenses were designed for the adapter, rather the other way around. For the round-fronts the designs seem a lot more inter-connected. I´m sure all these things make a difference.

    The squarefronts are cool though, but insanely hard to get in OK condition.

  • @slrmagic I do hope you continue development of this lens. I would love to have a dedicated anamorphic lens for m43. I would much prefer that over another adapter. If 40mm is as wide as you can go - I would be more than happy with that! It's a better option than anything else out there right now. 2x would be great, but I would probably buy a 1.8x too.

    Just make it as cost effectively as you can. If it's around $1200 or so - it's still better than other options out there

  • @RRRR: Oh! Silly me - I forgot completely about the LOMO squarefronts. I guess someone has done what I proposed before after all. I know the roundfronts produce better quality images, but I have never been clear whether this is due to the optical design itself, or just because they are older and/or more poorly made.

Start New Topic

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In with Google Sign In with OpenID

Sign In Register as New User

Tags in Topic

Top Posters