Home of Professional Cinematography since 1996

style="margin-bottom: 0"> 

class="style5" Viper 'Magenta' Filter

>Published : 21st November 2005

>When one does not want the green cast of Viper/filmstream, what types of filters have been used to correct for the lack of white balance ?

>And what's the density required, meaning how much stop does such a filter lose - it seems like it must be close to 1 stop loss.

>I'm currently investigating the Viper for a project, and we don't want our dailies 'green', instead we want a more accurate representation of what we're going to end up with.

>I'm also looking into monitors with end-result LUT's, but it seems with the Viper that an optical filter may still be a requirement.

>Any help greatly appreciated.

>Mark Doering-Powell
LA based DP


class="Paragraph">>... and we don't want our dailies 'green', instead we want a more >accurate representation of what we're going to end up with.

>I'm also looking into monitors with end-result LUT's, but it seems with >the Viper that an optical filter may still be a requirement.

Mark,

>Have you considered a suitable magenta lighting gel over the face of the dailies monitor[s]? Won't affect what's recorded helluva lot easier [read: "cost effective"] than finding monitors with LUT’s.

>Sometimes simple's best.

>Cheers,

>Clive Woodward


class="Paragraph">> I'm currently investigating the Viper for a project, and we don't want our >dailies 'green', instead we want a more accurate representation of what >we're going to end up with.

>I use 35 Magenta and 5 red. I would price out having a filter made to go into the filter wheel.

>Dave Stump ASC
VFX Supervisor/DP
LA, Calif.


>David Stump, ASC, has done a lot of work with magenta filters and the Viper.

>Jeff "not a pain in the asp" Kreines


>Woodward, Clive wrote :

class="Paragraph">> Have you considered a suitable magenta lighting gel over the face of >the dailies monitor[s]? Won't affect what's recorded helluva lot easier >[read: "cost effective"] than finding monitors with LUT’s.

>Yes, you're right about trouble finding/prepping HDLink LUT'ted monitors. But I do want to affect what's recorded - and let all subsequent dailies/DVD's reflect a look that's closer to where we're headed. Otherwise, some people watching from afar may not understand that its all to be colour corrected, and as usual, people get used to the way it looks on the Avid for several weeks. Ugh!

>Plus, the green cast isn't really helping (I know its flat and correctable, but from what I've learned its just as well if not better to 'white balance' it with a filter specifically when colour correcting in an RGB environment).

>Mark Doering-Powell
LA based DP


>Dear Mark,

>You are heading towards a classic case when shooting with the Viper... Actually, if I had to give you an advise I would probably tell you to not use a magenta filter. These filters do not work perfectly and the results depend a lot on what you shoot. For example a scene where you have lots of natural greenish-blueish shades (trees, ocean, etc) the magenta will destroy the greens and blues. However it might give interesting results on skin tones.

>Bear in mind that an optical filter limits the amount of lights that strikes the Viper's CCD and because of that you loose some of the precision in the picture.

>Another alternative is to use LUT's to display the green images with a white-balanced look. Again this has some drawbacks as if the LUT’s are not correctly computed you will display images that will never exist after the post production process.

>Also do not forget that the Viper reacts a bit differently depending on the light conditions. What is important is to make sure that you do not underexpose or overexpose because in post you will not have the precious highlights the Viper is capable of capturing.

>I'll be in LA on April 14th, if you want to meet there and discuss this, let me know. I might share with you our experience with the Viper on several projects.

>Dan Tatut
CEO
CHROME Imaging
40 Rue de la Coulouvrenire
WWW: http://www.chrome-imaging.com


>When I was looking into doing a feature on the Viper, the plan was to get the dailies colour-corrected just like with film dailies when making the Digital Betacam downconversion for editing. I was going to shoot a grey scale at the head of scenes, etc. just like with film. A dailies colourist was going to correct the footage.

>David Mullen, ASC
Los Angeles


>In our 3cP system a conversion from Viper to normal images is an automatic via Viper preset. Our monitor is calibrated to tele-cine of your choice .

>Next step is a colour timing with printing lights and curves and preset profiles "imitating" film stocks (Fuji and Kodak), processing (normal, push bleach) and then sending images and meta-data to telecine to the monitor where colourist makes transfer.

See our 3cp at glance page on www.gammaanddensity.com

Phillip Cotter, Video Projects Manager
Gamma and Density Co.
1041 N. Formosa Ave., Formosa Building, Suite #009
W.Hollywood, USA,90046


>Hi Mark,

>here's my experience with the viper:

>"Mark Doering-Powell" wrote :

class="Paragraph">> ...what types of filters have been used, and what's the density required

>MotionFX in the UK made some special filters for the use on the viper, we used them as our basic/main filters and added CC filters depending on the colour temperature on set. So we tried to keep the camera almost 'balanced' all time, except in those shots where light levels or light-colorimetry told us to keep the original look - so I don't recommend to use one filter all the time, balance for each scene/location and think carefully if you have enough light to do that.

>The advantage of a balanced image is you use as many bits as possible in each colour-channel (RGB 10bit log) to give as much information to the post.

>This procedure didn't take to much time with a portable WFM and a small greyscale or with an experienced eye at the waveform monitor...

class="Paragraph">> ... and we don't want our dailies 'green' …

>Once you use filters your dailies aren't that green anymore...

>Your recording system (stwo/directors friend...) should allow you to apply a LUT only for a dailies playout if a LUT is required, that's what we used when no filtering was used in camera to capture for off-line

class="Paragraph">> I'm also looking into monitors with end-result LUT's

>Who decides the 'end-result'? Do you want to have that discussion on set? Every shot could be different and require a little different LUT.

>For on set work maybe a simple adjustment like on the HDLink Utility (via USB) could simulate basic set-ups on an Apple Cinema Display, maybe that's enough?

>"David Stump" wrote :

class="Paragraph">> I would price out having a filter made to go into the filter wheel

>I think it was 2001 or 2002 when we asked Thomson to integrate some colour filters into the viper... Hello - anybody out there?

>"Dan Tatut" wrote :

class="Paragraph">> I would probably tell you to not use a magenta filter

>As I said : it depends on your setup / location / lighting - but if you want to be sure, test it for yourself! Over all: In 70% of all shots a filter optimised the results from my point of view

>Hope that helps...

>Cheers

>+++ Florian Rettich +++
+++ D.I.T. (US) / D.F.T. (UK) +++
+++ Munich, Germany +++


>On the subject of putting custom filters in one of the Viper's filter wheels, installation of filters in those locations must be done by Thomson service personnel. Because the filter wheels are located inside the optical front section of the camera, anyone else will void the camera warranty by opening the enclosure that contains the optical block, preprocessors, and filter wheels.

>Just a couple of observations. Even if you lose a stop with a filter, there is usually much more sensitivity than most folks can use; in fact we usually use some ND filters anyway to achieve normal film style depth of field. But usually pre-production testing will determine a specific end result, unless your goal is to assure everyone on set that the exact desired end result is possible, the HD monitoring output, a colour corrected output of the camera, is available even in the FilmStream mode to alleviate the green monitor issues on a second HD 4:2:2 monitor. Of course if the "greens" are distracting on the primary working monitor, then the Magenta and Red filters would be necessary.

>GEORGE C. PALMER


>Florian Rettich wrote :

class="Paragraph">> I think it was 2001 or 2002 when we asked Thomson to integrate some >colour filters into the viper...Hello - anybody out there?

>It was only NAB 2002 that the Viper and FilmStream was introduced. Concerning the magenta filter in the filter wheel Thomson soon replied that it's way too expensive to have the filters manufactured, considering they have a couple of thousands of these filter wheels in other Thomson cameras and only about 30 Vipers (now after three years !). The same would apply to the question why not have a different prism block to get rid of the green tint at all, or at least most of it (I know some colour balancing would still be required).

>Btw. we could apply a real-time LUT to correct for the green tint even in the very first alpha-software-version of our "director's friend", before there was any viewing HD output on the Viper.

>Mark Doering-Powell wrote :

class="Paragraph">> But I do want to affect what's recorded - and let all subsequent >dailies/DVD's reflect a look that's closer to where we're headed.

>Maybe you should also consider shooting RGB with video processing. Allows you to really determine what's recorded but still 10 bit. Or think about HDStream. Gives you log data, but 3.2K and 5.6K white balance presets, saves storage during recording (even allows for HD-CAM tape recording) or if you go to hard disks and know how to do some smart filtering from YUV422 to RGB444, you can get somehow close to original 444 capturing. It always depends on ...

>But if you consider manually balancing the camera before every shot with a couple of filters, I think these are also other viable options. But only if you really want to affect what's recorded. Otherwise I would also suggest to use the viewing output of the Viper for on-set viewing and a LUT applied to the output from s.two while downconverting for offline.

>Dirk Meier
Digital Film Services
Cologne/Germany
"Director's Friend Co-Founder... Found and Lost"


>George C. Palmer wrote:

class="Paragraph">> On the subject of putting custom filters in one of the Viper's filter >wheels, installation of filters in those locations must be done by >Thomson service personnel.

>That's going to be appreciated by crews everywhere.

>Ship the camera in to get the filters swapped out...

>Jeff Kreines


class="Paragraph">> Ship the camera in to get the filters swapped out..

>I've always preferred filters in the front so that's not a problem for me.

>It was 2002 that we were talking with Thomson about 2 filters, one balanced to take the green out in Tungsten environments and one balanced to take the green out and balance the camera in daylight.

>I believe that they were talking to Schneider about this.

>As the demo I shot for Thomson for Cannes in 2002 clearly showed the advice they were giving at the time not to make any allowances at all for location colour was wrong and that a much better result could be got by colour balancing with filters so that you could use ALL the dynamic range available in ALL colours.

>Cheers

>Geoff Boyle FBKS
Director of Photography
EU Based
www.cinematography.net


>Hi mark,

>As Florian states we developed a number of specific filters for Viper for use on Silence Becomes You.

>You can read a lot more about this projects, and the use of Viper at www.digitalpraxis.net

>As Steve Roach points out using filters has a better beneficial effect than correcting in post due to maximising the available data range by aligning the RGB separations across the whole of the exposed image. Correcting only in post will always leave the absolute shadow and/or highlight detail with a cast offset that can't be corrected. But this is hidden by the toe and shoulder of the final film/print LUT images...

>If you want to know more about how to shoot with Viper feel free to contact me directly, or on list. We obviously have a great deal of experience with Viper as MotionFX (my client) has three of the things and shot 25TB of data for SBY in Filmstream mode - the first full feature film to be shot this way in the world.

>There are examples of images on the website too showing the general effect of the filters.

>Florian was out DFT (Digital Film Technician) on the project so has a lot more experience than most.

>Steve Shaw
Digital Praxis Ltd
www.digitalpraxis.net


class="style7">>On the subject of putting custom filters in one of the Viper's filter wheels, >installation of filters in those locations must be done by Thomson >service personnel.

class="style7">>That's going to be appreciated by crews everywhere.

class="style7">>Ship the camera in to get the filters swapped out...

>No one in their right mind would consider changing anything in a sealed optical path in the field, unless it was a MAJOR emergency, and even then...

>Just to keep things straight, you should post a disclaimer every time you disparage any HD camera, seeing you have a vested interest in your own.

>Bob Kertesz
BlueScreen LLC


class="style7">>Just to keep things straight, you should post a disclaimer every time you >disparage any HD camera, seeing you have a vested interest in your >own.

>Gee, Bob, I think everyone on the list realizes that I have a vested interest in our digital camera. It's not exactly a big secret.

>But I can't see that praising or disparaging workflow issues on other cameras is something I should avoid. Should I also post a disclaimer when I say something nice (and I have) about other cameras?

>I don't see any one camera as ideal for all users. Just like film -- some people will want to use an Arricam and others an A-Minima. No camera can be ideal for everyone.

>Jeff Kreines


>A pundit comments :

class="style7">That's going to be appreciated by crews everywhere.

>Ship the camera in to get the filters swapped out (Magenta and Red filters installed in the Viper filter wheels...

>It is a fairly inexpensive procedure even for service personnel and the filters, normally unused in cine applications, that are removed (4-point star, 6-point star, and soft focus from one of the filter wheels, and one or two of the ND filters from the other filter wheel)would probably never have to go back in. Optimally two from each filter wheel so magentas and reds could be mixed. The procedure is routine and can actually be done fairly locally and it could be done with no future changes necessary. Most cine applications use a rod mounted matte box anyway, so the filters can be and are routinely used in the matte box. And I'm sure that those who appreciate the merits of existing premium cameras like the Viper won't be offended by such routine procedures, since much larger investments are routinely made in custom camera pieces for cameras of all types.

>Separately, on the subject of using a different prism block to get rid of the green tint, the prism block is not the source of the green tint; that arises from the nature of the unprocessed FilmStream image and its Dual Link transport stream. So a different prism block would not correct the green look, but it would sure limit the flexibility of use of the Viper for RGB 4:4:4 or any other of three processed mode options. But since the comment was based on the economic constraints, yes it would be difficult to implement it even if desirable, due to the economies of producing low volumes of a premium "niche" camera.

>GEORGE C. PALMER


>A few questions:

>(1/) If this green issue is such a problem, why did the Viper go into production with the issue unresolved? Isn't it just a normal by-product of the process?

>(2/) Why change the colour balance if this is the way the camera is to be used? Shouldn't we be focusing on monitoring a processed signal instead of changing the signal itself?

>(3/) Isn't this, in reality, yet another attack on the immensely unpopular colour green? Who are we to judge one colour over the others and find it wanting? When will this extreme "anti-greenism" end? I suspect this forum is simply populated by close-minded magenta-ists and red-ists who are looking out for their own selfish purposes?

>(4/) Okay, so is the green shift really a problem?

>Art Adams, DP [film|hidef|video]
San Francisco Bay Area - "Silicon Valley"
Local resources : http://www.artadams.net/local


>1/) It was part of the design philosophy not to manipulate the pictures as they leave the camera, this is how they come.

>2/) There is a processed picture for monitoring, unfortunately, or fortunately depending on your point of view, this doesn't show the full range of the recorded signal. Being single channel it can't.

>3/) "When will this extreme "anti-greenism" end?" Never! I hate the stuff. I'm always getting the green reduced in TK.

>4/) No.

>Cheers

>Geoff Boyle FBKS
Director of Photography
EU Based
www.cinematography.net


>Dear all "anti-greenism" extremists,

>The Viper can be used in production, no questions asked. The real problem with the Viper is not the green tint on the images. For those who are interested in seeing a properly colour corrected Viper image, we will post some on our web site this week. Otherwise just go to Digital Praxis (Steve Shaw's website) and see it by yourself. Green is NOT A PROBLEM!

>The Viper has some other issues that a DP must be aware of :

>1/) Its FilmStream mode is imitating the film's look by capturing pictures in 10bit log with a dynamic range that is higher than a regular video, but still it's not film. Conclusion: see the Viper's Filmstream images a different type of media an rely on a good colourist with a good knowledge of what 10bit log is and how to white balance the green tint! Moreover the system used for the colour grading should have a 32-bit float colour pipeline to give maximum freedom (avoid as much as possible the 14/16bit systems because at some point they have to clip the values).

>2/) The Viper has no recording mechanism (unless the production has the budget to rent the new Venom recording device), therefore you have to record directly on disks. In any case this generates terabytes of data (even for a short feature film). This means that the management of the data has to be understood in advance by the DP, the production and the post-production guys!

>3/) If you are not using the Venom (pretty expensive) and record directly to disk, then it means that the Viper is connected to the capture system through SDI cables. In this case you must make sure that the double SDI signal is correctly synchronized (does not happen with Thomson's break-out box). Otherwise some pixels might be wrong in luminance and chrominance and it's pretty impossible to remove them manually (for those who had this problem, just drop us an email, we developed a dedicated tool to correctly restore those pixels)

>Besides these points, the Viper is probably the best quality camera around. Seriously, even a CineAlta with a log curve does not stand the comparison.... just try!

>Dan Tatut
CEO
CHROME Imaging


class="style7">>Just to keep things straight, you should post a disclaimer every time you >disparage any HD camera, seeing you have a vested interest in your >own.

class="style7">>Gee, Bob, I think everyone on the list realizes that I have a vested >interest in our digital camera.

>Gee, Jeff, I don't think that's true.

>And even if it were, in over six years participating on the list, I have yet to see a single manufacturer's or sponsor's representative talk trash about their competitors. Not one. Except you.

>Bob Kertesz
BlueScreen LLC


>Art Adams wrote:

class="style7">> (4/) Okay, so is the green shift really a problem?

>I'm a jury of one that is still out on this question.

>As I mentioned here and on my blog last week, I just shot a couple of spots with the Viper in FilmStream mode. George P. did a marvellous job as my tech and tape op.

>I created a de-viping process in Digital Fusion that simply implements the standard log2lin and colour balancing math that Thomson shared with me. I just today concluded a day of client-supervised grading on the material, and am very happy with the results.

>However, in our world where “over-brights” are preserved, respected, and utilized, the fact that some extreme highlight areas had a magenta cast did rear its ugly head.

>Had I the time, I would have figured out something to render colourless any channel that had blown out, but this was not something I could do on the schedule of this project.

>I imagine the de-greening filter would have helped us out, although I would have wanted to shoot comparison tests to make sure we didn't much up our colour sensitivity. We certainly piled enough ND on the Viper to survive the light loss.

>But even then, we would have white balanced in post to some extent, and in doing so would have sent some channels off into the stratosphere.

>I imagine single-sensor raw-shooters will have a similar issue, and I further imagine that in the debayering of those images that enwhitening of unevenly clipped channels will be SOP -- similarly to how I see Adobe Camera RAW artificially whitening blown areas in my DSLR stills.

>I'll post to post when I have more to share about my Vipeline. For now Art, my opinion is "not unless you have some blown-out areas in your shots, and then, only if you plan to do more than just grade the material."

>Stu Maschwitz

>Director, The O, SF/LA
http://prolost.blogspot.com
http://www.theorphanage.com


>Stu Maschwitz wrote :

class="style7">> I imagine single-sensor raw-shooters will have a similar issue, and I >further imagine that in the debayering of those images that enwhitening >of unevenly clipped channels will be SOP -- similarly to how I see >Adobe Camera RAW artificially whitening blown areas in my DSLR >stills.

>It appears that Panavision has not gone this route with their Genesis. When you go into 'raw' viewing mode, it indeed goes flatter, and shows you the whole range, but no green cast. I hate that green when judging the image. We're told that all 3 'stripes' on Panavision's non-bayer sensor are roughly equal in RGB sensitivity.

>Seems to make a lot of sense if you're going to make your own sensor from scratch and know you're going to process the image right away.

>Love the Viper though. Using it next week.

>Mark Doering-Powell


class="style7">>(1/) If this green issue is such a problem, why did the Viper go into >production with the issue unresolved?

>It's not a problem, it's simply one way of achieving high dynamic range out of a very sensitive sensor. But while we are on the subject...If I walk on to any set, anywhere in the known universe and ask if something, (anything) is a "problem", that thing immediately becomes a matter of grave concern. My professional advice is; stop saying "problem"!

>Put the word out of your on set dictionary and you will live 5 to 10 years longer, I guarantee it.

>Green sensitivity is a technical choice that emulates human vision. Technically, it buys a lot of dynamic range. Yes, it's part of the Viper process. Just imagine you are viewing the digital equivalent of an interpositive of your footage that just happens to be on a green base.

>Use the LUT, it works. Curse the world that refuses to invent 10 bit monitoring solutions, and then move on.

>(2) Why change the colour balance if this is the way the camera is to be used? --Shouldn't we be focusing on monitoring a processed signal instead of changing the signal itself?

>Very well said. We should also focus on monitoring PERIOD. We should be focused on some way of realistically viewing what we are shooting, as we are shooting it, and the discussions of workflow all seem to excuse the manufacturing community from solving this problem. We are stuck with consumer grade flat panel monitors that are not durable enough to stand up to the rigors of production, and bottle monitors that are stuck in the age of phosphorus. We expend Herculean effort to build incredible new cameras, and what do we view the output on? Unverifiable, unreliable, un-calibratable (newly invented word)
relics.

>(3/) Isn't this, in reality, yet another attack on the immensely unpopular colour green?

>I happen to like magenta. It's nice.

>(4/) Okay, so is the green shift really a problem?

>No. Relax and enjoy freedom of choice. Vanilla... Chocolate...
Magenta...choose. Then live with your choice. That's how life works.

>Dave Stump ASC
VFX Supervisor/DP
LA, Calif.


>Just a couple of more observations. Because the Viper has 4 operating modes, and three are entirely normal processed, only the FilmStream mode seems to violate the sensibilities of some because it has a greenish look. It is not a "shift". Exactly like an unprocessed negative and its uncorrected IP which also has no "shift"; that IP, if uncorrected or ungraded, however, will not be pleasing or useful to anyone until it is graded.

>The green tint we see in the FilmStream could be correlated to the natural caste of a film emulsion, which requires grading for finishing, but in the case of the FilmStream, the caste is created by the nature of the mathematical contributions of the R, G, and B signals when they are encoded onto the Dual Link transfer signal. In addition, when we attempt to correlate the green look of an unprocessed FilmStream Dual Link output to a processed RGB signal, we must understand that the latter is not only colour corrected but also exists in a linear colour space.

>The FilmStream output is not only unprocessed but it exists in a Log colour space which requires processing, grading for finishing. So even though the green caste is natural to the output of the camera it was not a design specification, only a result of creating an unprocessed Log RGB signal which could be treated exactly the same as a film IP for those who are comfortable with the work flow of film. And if one chooses to use a magenta filter that's fine, but it isn't necessary if exposure is guided by using a 320-400 EI lighting reference. So the green is not a problem, but a natural artefact of producing a camera capable of superior dynamic range and colour depth. It can be used as is using a film shooting mindset paradigm, or a magenta filter can be used to mitigate the green; if carefully implemented the same result can be obtained. For everyone else, the 4:4:4 RGB processed mode might be a better cup of tea.

>GEORGE C. PALMER


>"Dan Tatut" wrote:

class="style7">> In this case you must make sure that the double SDI signal is >correctly synchronized (does not happen with Thomson's break-out >box).

>Dan - could you share some details on that 'problem'?

>Why should there be a timing problem using two cables with the same length and a equalizer in the break-out-box?

>Cheers

>+++ Florian Rettich +++
+++ D.I.T. (US) / D.F.T. (UK) +++
+++ Munich, Germany +++


>Message :

>Here at SLF Green is the new Coral ! It's called Jade, manufactured by Harrison & Harrison, designed by SLF for our clients !

>Art Wrote :

class="style7">>Isn't this, in reality, yet another attack on the immensely unpopular >colour green? Who are we to judge one colour over the others and find >it wanting? When will this extreme "anti-greenism" end?

>I agree with Art, one colour is no better than the other. Your project and budget dictates your camera. Your creative juices colour it ! Your technical ability and know how makes one acquire the best image possible with the tools you have been supplied with for the job.

>George Wrote :

class="style7">>On the subject of putting custom filters in one of the Viper's filter wheels, >installation of filters in those locations must be done by Thomson >service personnel.

>The point of changing filters and using a Matte Box. Most rental company's don't like to change filters for short rentals ! Considering that they already have Matte boxes and filters in there fleet which will compensate for this bias already.

>Geoff Wrote :

>It was 2002 that we were talking with Thomson about 2 filters, one >balanced to take the green out in Tungsten environments and one >balanced to take the green out and balance the camera in daylight.

>I believe that they were talking to Schneider about this.

>Currently Schneider only has CCM 10-20 & 30 + CCG10 Listed in Matte Box Sized filters, not rear mounted sizes.

>Other manufactures will or can produce a wider comprehensive range of CC's, in 0.05 or even 0.025 increments. But again in Matte Box sizes.

>I can not comment on one camera over another, only the wider the variety the more use there is for a variety of tools to compensate and eliminate a wide variety of problems and dilemmas. Thus keeping your back ground contacts hard at work helping you to resolve these things......

>Carey Duffy
South London Filter Ltd
http://www.camerafilters.co.uk


>Hi Florian!

>By looking at your company name, I guess you work with Dirk Meier (BTW, very good guy!)

>Well, since the Viper is an electronic camera working with a CCD, you have all the problems of a CCD-based camera. But basically if you do not use a break-out box (which syncs the 2 SDI streams) and the camera is connected only through SDI cables to the capture system (we observed that with the guys in Italy at Digit One running the Viper connected to Director's Friend), and if the cables are not exactly the same in length and type (or they are not shielded). You will have sync problems when the signal arrives in the capture system (whatever it is).

>We developed a tool for our post-production system (Matrix, check our website : http://www.chrome-imaging.com for more), that detects these very specific pixels and restores them to give you a clean image (samples included in the email).

>This can be avoided with devices that resync the signal... but it happens in production that you do not have such device at the shooting location... (it happened a couple of times)

>For example when we were involved in the shooting of "The Slug", a short feature shot with the Viper, the camera what mounted on a rig and connected to the Director's Friend through cables. There was no sync device around and we could not get one easily. The break-out box cable was too short for the rig and we had to connect through SDI cables only. Of course the problem occurred (actually it was quite invisible on the DF monitor, but popped up at the end of the day when we imported the images in the colour grading system).

>If you are interested in some sample images with the problem and the corrected version, I can send you some, but they are quite big. Let me know if you have enough space in your mail box..

>Dan Tatut
CEO
CHROME Imaging


>George C. Palmer wrote :

class="style7">>...And if one chooses to use a magenta filter that's fine, but it isn't >necessary if exposure is guided by using a 320-400 EI lighting >reference.

>Judging from this thread, and more private replies than I've ever gotten, it seems that we're coming down on two sides of the issue of Viper green.

>Greenies say the greens great. Its just like orange neg (although in recent years some mask densities have been reduced there too !) And they do not have an issue with possibility of magenta highlights (render them away in post somehow ?), nor with pushing the red/blue channels up in post which can add some blue noise on rare occasion. Greenies tend to be post people for the most part, really smart ones that can juggle several thousand pixels at once.

>Magentites (aka anti-green extremists) say you get a better image/signal to colour correct with the red/magenta filtration (whether full correction or partial, or sometimes none at all depending on the lighting) since we don't need the green 'mask' and it can tint your highlights after you correct-correct. Magentites tend to be people who spend time on set, with clients, etc. and have seen a therapist due to bad choices in correction-correction. Some may have been previously fired for green dailies ! Others have consulted for Thomson and yet still filter away the green. Some post/vfx pros come down on this side too.

>Recently David has joined Switzerland in saying its not a problem, but just a choice you live with, and so it is of course. Paparazzi have photographed David using his own filtration recipe as well, so he's not as neutral as it seems.

>Barring a UN mandate, I'm still on the side of filtering towards the intended look when possible, as the green cast is not a "necessity" to the workflow, only to the sensor's unprocessed workings. If not for anything other than to better judge the total image while also checking the dynamic range on-set. Instead of toggling back and forth to the viewing output - which you'll do anyways - but it seems to make sense to capture a correction cast that's CLOSE to where you'd like to be, and give you more range to correct-correct later. And then I can still make it green if I want to.

>Mark Doering-Powell
LA based DP


>MDP

>Your powers of observation and persuasion are astute. However you can put this argument to bed with 2 words.

>Test It

>B. Sean Fairburn
Director/ Cinematographer
Role Model Productions LLC


class="style7">>It appears that Panavision has not gone this route with their Genesis. >When you go into 'raw' viewing mode, it indeed goes flatter, and shows >you the whole range, but no green cast"

>I might be wrong, but I believe the Genesis has only a processed RGB Dual Link output...Can anyone clarify that?

>After a couple of conversations I discovered why the custom Magenta and Red filters have proven so elusive in the camera filter wheels. The very tight thickness specification required of filter wheel filters to maintain consistent back focus has deterred filter manufacturers from responding to the need, perhaps because of the limited sales volume potential. The technical fact is that the thickness tolerance for a filter in the filterwheel is completely different from a filter in a matte box....the very tight tolerance for the filter thickness (glass length) is required to maintain back focus when switching between filters. So it is easy to acquire filters for use in front of the lens, where the thickness does not matter, but it is more difficult (so far impossible) to get them for the filter wheel. If any filter manufacturers out there are interested in providing these filters, please contact me offline, and I will put you in touch with folks who can implement them.

>GEORGE C. PALMER


class="style7">>If I walk on to any set, anywhere in the known universe and ask if >something, (anything) is a "problem", that thing immediately becomes >a matter of grave concern. My professional advice is: stop saying >"problem"!

>Better advice than this has never been given on CML. Well said, Dave.

>Bob Kertesz
BlueScreen LLC


>"George C. Palmer" wrote:

class="style7">> So it is easy to acquire filters for use in front of the lens, where the >thickness does not matter, but it is more difficult (so far impossible)

>I heard that, too. But by looking at the Schneider Kit of filters for the F900 I would assume it is possible to make magenta filters for the viper (in a good quality) as well but with less profit/high price because of not so many sold vipers - am I right?

>So it's not so much a problem to produce it...

>Please correct me if I'm wrong.

>+++ Florian Rettich +++
+++ D.I.T. (US) / D.F.T. (UK) +++
+++ Munich, Germany +++


>It not a problem to manufacture the filters; it' an opportunity to make a meagre profit selling them.

>Max H. Penner
Paradise F.X. Corp.