Aapo Lettinen

that is probably one of the largest reasons why Alexa has been so popular: it does not have much of a look of its own so it is acceptable for everybody. A major selling point for the producers too (economical, reliable and does not have a distinctive look which someone could dislike. Otherwise said, everything a producer could want from a camera)

if you open a LUT file with a text editor it can look like this (just a cube file I exported quickly from Resolve to test it. just a small part of the file, not going to post it all to flood the thread heh) : TITLE "Generated by Resolve" LUT_3D_SIZE 33 0 0 0 0.00114443 0 0.000396735 0.00216678 0 0.000595102 0.0037995 0 0.000610361 0.0184024 0 0.000518807 0.0331426 0 0.00030518 0.0492103 0 3.0518e-05 0.0692149 0 0 0.0937819 0 0 0.121599 0 0 0.153338 0 0 0.18735 0 0 0.224537 0 0 0.265293 0 0 0.308919 0 0 0.355138 0 0 0.402823 0 0 0.454932 0 0 0.514519 0 0 0.581064 0 0 0.651179 0 0 0.722255 0 0 0.782116 0 0 0.817014 0 0 0.841856 0 0 0.861295 0 0 0.877424 0 0 0.891371 0 0 0.903655 0 0 0.914351 0 0 0.922576 0 0 0.929686 0 0 0.936126 0 0 0.000350958 0 0.000823987 0.00172427 0 0.000717174 0.00294499 0 0.000579843 0.00596628 0 0.000442512 0.0198215 0 0.000335698 0.0344396 0 0.000183108 0.0501869 0 1.5259e-05 0.0700236 0 0 0.0939956 0 0 0.121798 0 0 0.153033 0 0 0.186725 0 0 0.223636 0 0 0.264591 0 0 0.308354 0 0 0.354574 0 0 0.402396 0 0 0.454795 0 0 0.514519 0 0 0.581384 0 0 0.651865 0 0 0.723339 0 0 0.783032 0 0 0.818021 0 0 0.842954 0 0 0.862455 0 0 0.87866 0 0 0.892729 0 0 0.905196 0 0 0.91606 0 0 0.924544 0 0 0.931777 0 0 0.938552 0 0 0.000640879 0 0.000686656 0.0021973 0 0.000503548 0.00361639 0 0.000335698 0.00764477 0 0.000198367 0.0213321 0 0.000137331 0.0355993 0 7.62951e-05 0.0514382 0 0 0.0710613 0 0 0.0949721 0 0 0.122652 0 0 0.153948 0 0 0.187564 0 0 0.224414 0 0 0.265034 0 0 0.309056 0 0 0.355566 0 0 0.403677 0 0 0.456275 0 0 0.516335 0 0 0.583642 0 0 0.654978 0 0 0.727108 0 0 0.785672 0 0 0.820508 0 0 0.845457 0 0 0.86508 0 0 0.88127 0 0 0.895354 0 0 0.907958 0 0 0.919066 0 0 0.927916 0 0 0.93521 0 0 0.942123 0 0 0.000885023 0.00340276 0.000350958 0.00263981 0.00294499 0.000213626 0.00422675 0.00300603 7.62951e-05 0.00939956 0.00320439 0 0.0228885 0.00323491 0 0.0372015 0.00325017 0 0.0530098 0.00283818 0 0.0728618 0.00202945 0 0.09691 0.000823987 0 0.124346 0 0 0.155825 0 0 0.189517 0 0 0.22652 0 0 0.267414 0 0 0.311299 0 0 0.357961 0 0 0.406531 0 0 0.459846 0 0 0.520455 0 0 0.588495 0 0 0.660685 0 0 0.733913 0 0 0.790722 0 0 0.824918 0 0 0.849515 0 0 0.869062 0 0 0.885374 0 0 0.899519 0 0 0.912108 0 0 0.923339 0 0 0.93225 0 0 0.939971 0 0 0.94696 0 0 0.00111391 0.0162966 0.000259403 0.0030518 0.0153658 0.00015259 0.00483711 0.0149844 6.10361e-05 0.0111086 0.0147707 0 0.0248112 0.0145266 0 0.0390936 0.0140993 0 0.0552071 0.0133211 0 0.0749828 0.0121462 0 0.0993515 0.0102693 0 0.126772 0.00743114 0 0.158679 0.00402838 0 0.192523 0.000823987 0 0.229801 0 0 0.270848 0 0 0.315221 0 0 0.362158 0 0 0.411002 0 0 0.465034 0 0 0.526635 0 0 0.595773 0 0 0.669154 0 0 0.743252 0 0 0.797848 0 0 0.831266 0 0 0.855116 0 0 0.874144 0 0 0.890394 0 0 0.904662 0 0 0.917464 0 0 0.928679 0 0 0.937652 0 0 0.945357 0 0 0.952667 0 0 0.00131228 0.0295109 0.000137331 0.00349432 0.0282444 7.62951e-05 0.00546273 0.0273899 3.0518e-05 0.0124971 0.0267338 0 0.026566 0.0263371 0 0.0413825 0.0256962 0 0.0577249 0.0246281 0 0.0778363 0.022858 0 0.10251 0.0203555 0 0.129961 0.0166018 0 0.161975 0.011841 0 0.196551 0.00679026 0 0.233951 0.00233463 0 0.275578 9.15541e-05 0 0.320455 0 0 0.368093 0 0 0.417639 0 0 0.47219 0 0 0.535088 0 0 0.605646 0 0 0.681254 0 0 0.755505 0 0 0.806912 0 0 0.83827 0 0 0.86157 0 0 0.880461 0 0 0.896437 0 0 0.910674 0 0 0.92343 0 0 0.934691 0 0 0.943755 0 0 0.951614 0 0 0.958968 0 0 0.00143435 0.0436408 0 0.00392157 0.0419776 0 0.00613413 0.0410315 0 0.013962 0.0403601 0 0.0282597 0.0400549 0 0.0434119 0.0394598 0 0.0606241 0.0386511 0 0.081178 0.0366827 0 0.10631 0.0332952 0 0.133715 0.0277409 0 0.165698 0.0203555 0 0.201129 0.0130617 0 0.239597 0.0053254 0 0.281498 0.000366217 0 0.327016 0 0 0.375509 0 0 0.42652 0 0 0.482124 0 0 0.546273 0 0 0.618509 0 0 0.695766 0 0 0.77023 0 0 0.816281 0 0 0.846159 0 0 0.868925 0 0 0.88748 0 0 0.903334 0 0 0.917266 0 0 0.92987 0 0 0.941207 0 0 0.950439 0 0 0.958343 0 0 0.965866 0 0 0.00144961 0.0614939 0 0.00440986 0.0604257 0 0.00704967 0.0598611 0 0.0150912 0.0599069 0 0.0300603 0.0601358 0 0.0456092 0.0601053 0 0.0634165 0.0596017 0 0.0847944 0.0584573 0 0.110216 0.0560311 0 0.138079 0.0511482 0 0.170336 0.043595 0 0.20621 0.0336461 0 0.245716 0.0207523 0 0.28896 0.00799573 0 0.335195 0.000366217 0 One does not write the values by hand to the LUT file so it is not "complex" to make them and it is easy enough for anybody to try it

Most of the film emulation conversation is about creating a magic button which automatically corrects hasty lighting and exposure, low budget and bad cinematography and outputs the glamorous "film look" of the 70's and 80's. Everyone can see what is the problem with that approach when discussing about a look of a feature film instead of just learning to do one's job properly so that one does not need to rely on the engineers and post fixes to make content watchable? btw. , Look Up Tables (LUTs) are just a way of transforming values to another values in a predictable manner. They are not specific to cinematography, they are used in all types of signal processing in everyday applications. Most of the stuff they are used for does not involve any digital imaging or anything audiovisual whatsoever. They are just used whenever they are the most practical way of transforming values to other values in computing. LUTs can be complex or very simple depending on how many values need to be mapped to different values. The simplest Look Up Table could look basically like this, mapping input values from 0 to 20 to output values of from 0 to 20: 0 = 0 1 = 0 2 = 1 3 = 2 4 = 5 5 = 5 6 =6 7 = 7 8 = 9 9 = 10 10 = 10 11 = 11 12 = 12 13 = 13 14 = 15 15 = 16 16 = 17 17 = 17 18 = 19 19 = 20 20 = 20 One can add more levels of precision and more dimensions of mapping the input values to the output values ( for example 100, 200, 015 = 101, 220, 022 ; 101, 200, 015 = 108, 221, 019 mapping the 3 dimensional input values to 3 dimensional output values to map for example a color value to another color value) but it does not change the working principle of it. It is just a table which contains information what type of output value a certain input value will generate. Making a 3D LUT out of Resolve takes something like 15 seconds. I actually use the option quite often to transfer basic colour grading information from Resolve to Premiere Pro if there is a technical reason why this is the most practical way of working. Grading the clip in Resolve, outputting the 3D LUT, then importing that LUT to the clip in Premiere timeline using for example Lumetri to import the lut to the clip. That transforms the basic grade (colour mapping information) between the programs but does not transfer anything else like the previously mentioned halation effect or anything similar. a LUT just maps input values to output values and nothing more

yes I am mainly talking about the noise structure and other imperfections like micro dust, slight mechanical instability and different MTF response to the finer details. Everything else like color reproduction goes unnoticed for 99.9% of the audience and they will forget about it in 30 seconds so it is not the main area of interest when making any comparison between the formats. Additionally, like Stuart said, the color reproduction of the current digital cameras is fine. If there is anything fundamentally wrong with the colors then it has been a intentional choice by the filmmakers or alternatively they cheaped on the budget somewhere.

it is not about taking digital image and trying to add a varying amount of "film look" over it. The whole point of the "film look" is having a more or less imperfect image and then appreciating it as is or even enhancing the effect. The whole idea of digital image capture is to make "as real as possible" representation of the reality and adding imperfections over digital image just makes it crappy digital image. It is like looking the reality through a dirty and dusty window, trying to see what is going on outside. Most of the audience would rather have the window cleaned so that they could just see better what is going on... Personally I only add "film grain" or other imperfections on a digital image if there is VFX or other stuff which needs to be hidden so that the audience is not distracted by it. For that use the "film emulation" works pretty well and helps selling the VFX much better so that the audience can concentrate on the story. Otherwise I rather appreciate both image capturing methods as is, trying to take advantage of their differences rather than trying to hide them (trying to hide the differences just leads the end result being mediocre and dull, being half something half nothing like a raw image which is not graded yet)

10 - 11 years ago the digital fans where always starting the conversation by claiming that "film is dead because Red is 4K" and so on. Ever since the HDCAM era it has been this absurd competition or "Format Wars" as some might call it. One of the sad things for purists is that the "film look" they admire which is generally the stuff shot between 1940's and 1980's is non existent nowadays even if you DO shoot on film and with the exactly same lighting style, set design and the same cameras and lenses. That is because those old film stocks and processing are not available anymore and all the admirable imperfections generated by them are lost forever. Same set design and lighting style are challenging to reach because of the current budgets and schedules. That is a horribly depressing outcome for people who are clinging to the past, trying to recreate the magic of the silver screen exactly as it was when they were still young. One would need to recreate the whole world of the 60's or 70's to reach even a small bit of the true feeling of being there again, in that dark movie theater in a small town long long time ago. Not saying that there is anything wrong with it, but it is just a little bit sad I think. People trying to recreate a world which was forever lost a long time ago and which is almost impossible to understand by the newer generations.

I have learned enough in the past year developing my own Crystal Sync systems to know that most people are only interested in the film in a very theoretical and nostalgic level. Part of that POV is that they tend to outright reject any new tools or ideas developed after maybe 2005 or so which may seriously affect the survival of film as a shooting medium by my opinion. Amateur or not, but you HAVE TO be able to update cameras and develop new accessories for existing film cameras whenever needed to keep them alive. Even being able to develop completely new cameras like David S. has presented. It is allowed to S16 or U16 modify film cameras so why the heck the electronics cannot be updated too? one cannot just rely on the availability of the decades old Tobin modifications which pop up on eBay every couple of months, one at a time if ever. People complaining about the availability of Crystal Sync cameras and then rejecting any newly made option which is offered? I don't get it. Constant evolution and adaptation is the key to survival for any industry and that is one of the pitfalls of the whole film originating image capture. The way I see it, film enthusiasts tend to be somehow "mind-locked" and cannot easily think out of the box when on the other hand, video/digital persons are very adaptive and can immediately update and finesse their tools whenever needed. (images are from the Konvas 15EPSS crystal modification I am working on at the moment. Just to show F.W. and others that I am definitely using SMD parts whenever it is needed... )

As far as I have seen, maybe at least 60% of the people on this forum are thinking just that: in almost every situation a scene "would have been better" if shot on film rather than digital. Maybe 10% of those people actually work in the film industry and maybe 1% of them are Cinematographers in more than theoretical level. the rest probably being somehow connected to some type of multimedia industry but maybe not even working full time. What they have in common is that they love to watch movies and argue about them and are interested in the technical aspects of the filmmaking in a theoretical level more than actually doing anything by themselves. basically it is the film critics telling the cinematographers how to better do their job. part of the film vs digital silliness for sure ?

what did you expect? they are the only type of religious / political subject matter allowed on the forum ?

if all the new digital movies look like crap then people should only watch the old movies shot before 2002 or so and completely ignore any newer stuff just because it is shot digitally? Nowadays I feel that most of the film vs digital arguments are about pure nostalgics vs. practicality and not about the aesthetics or "image quality" (which are just used as excuses to justify the different opinions). the other thing is that people who have no possibility to shoot their own projects on film (or who THINK they don't have that possibility) are claiming that everyone else should make the same choices they would like to do (to always shoot film whether it is practical for the project or not) and then making up all kind of excuses why film is the only right choice for all the projects out there

https://shotonwhat.com/cameras/moviecam-compact-camera I am pretty sure that all of the people on this forum have seen at least a couple of dozen of those films

they were used in many higher budget movies too in the late 80's and all of the 90's and 2000's. even in the 2010 still. I can remember for example the movies Salvador, 21 Grams, most of the LOTR movies, the Hellboy films, Tarsem films like The Fall, etc. being shot with them (just to name a few) and the Compact or SL being a steadicam/handheld camera on countless movies using Arri or Panavision as a main camera. later in the 2000's they were replaced by Arricam LT in similar use in some cases

yes the Soviets made their own cameras. most of the designs are loosely based on the Western design on the first model but then on the subsequent refinements their own design choices start to kick in and the later models are pretty unique designs. they are not "copies of the western cameras" in that sense. Their lenses are somewhat based on the similar era Western lenses (for example the Foton 37-140 zoom based on the similar range Angenieux one and the Lomo primes somewhat similar optical design than Zeiss standard speeds in most models. the Soviet lens coatings are by my opinion worse than the Western ones of the similar era and design and the chromatic aberration is different looking but these are just "part of their look" and are appreciated by most users ? there was some other Eastern Block cameras available as well like the Meopta ERK (very rare camera, I have only seen one on eBay couple of years ago and nothing since). I have seen some OCT19 modified Mitchells too so they seemingly used a limited amount of them as well. most of the Soviet cameras were just their own products like Konvas, Kinor, PSK, etc. gear. One interesting design is the Soyuz-US3N which is seemingly a partial copy of the Cinema Products XR35 (same movement etc) but is self blimped and has different body design

I had an impression that they used lots of rental gear, for example Panavision. Probably own cameras were not needed for mainstream productions

yes it sounds pretty challenging to make a completely new pressure plate. 1/20th mm accuracy would still be doable with dimensions I believe but for higher accuracy it is likely that the machinist would need to have the original plate on hand to be able to replicate it in good enough quality for it to be usable. if the seller would refund a substantial amount like 1.5 to 2k then one could maybe wait to find a new pressure plate or get one made. assuming that the camera would otherwise be in perfect condition

One of the easiest ways to protect a high power consuming dc circuit from reverse polarity is to add a diode to it which does not conduct power in normal operation but short circuits the power supply immediately if reverse polarity is applied. This creates very high current spike which blows out the fuse before other damage can happen. It is fully intentional feature and I believe the moviecams have this kind of arrangement too

I think most of the lower end stuff is done at least... just because of economics. The film scanning tends to be the biggest cost when shooting film if wanting reasonable quality output. Another thing is though that low budget indie filmmakers shooting shorts and hobby users never shoot enough film to justify a real time scanner so a frame by frame slow one would definitely work for most users. It is just that people THINK they will need a real time one in all cases (or just because it is neat) and thus reject every other viable option. It is not a projector after all... it is not meant for watching the shot footage in real time "to project it on the video screen for evaluation". To think about it further, one could easily scan a low budget indie filmmaker's daily footage overnight with a frame by frame scanner if taking into account that the said indie filmmaker does not have a fortune to spend on film and developing costs each day and the other budget of the film project seriously limits the overall footage available to them. there is other stuff though that may make it less useful to have your own scanner if trying to make a low budget feature with tight schedule. The diy frame by frame scanner is extremely useful if you are shooting small amounts of b/w film often and developing them by yourself in small batches so that you have very short b/w rolls to scan every couple of days. It makes no sense to send one 50ft or 100ft roll to 4k scanning abroad if you can scan it at home in reasonable quality with a frame by frame scanner and it takes a hour or two maximum to scan with the said unit. This is the kind of application I am developing my own scanner for because it is the most ideal use for a DIY scanner and one can get very good quality and great savings with it in that application. If shooting a feature film it makes not sense to scan anything by yourself because there is no enough time (because you need to sleep between the shooting days) and you need to concentrate on actually shooting the movie instead on handling the "lab work"

I could probably do the 0.5fps version of the machine vision system using the microcontroller approach and having enough time but I'm not familiar with spartans or other higher end fpga yet so high framerate cannot be done by me

I am sure a sprocketless machine vision system could be done with a infrared linear ccd/cmos and cortexM0 running at close to 300MHz but the framerate would be slow. If using the mirrorless camera as a pickup device the speed could be ok because the camera would be slow anyway but for 24fps speed would be extremely challenging and most likely would need the something like spartan6 or 7 to run the machine vision image processing and timing

A modern mirrorless camera should work well for the affordable good quality sensor solution as I mentioned if frame rate is not an issue. Can be inverted as well if shooting negative...filtration and image processing are used for that anyway and that is pretty basic for every camera. The sprocketless design can be challenging to make diy. It will most likely be very slow in any case but you may need to do some kind of diy machine vision system to run the sprocket hole counting and sprocket hole recognition to be able to make it precise enough to be usable. That would still require doing diy programming and possibly fpga or may be possible with a fast microcontroller if it is cleverly used

I have to think about the user interface when having time but I don't like the selector switches being this large (they take most of the space on the back panel) so it is still a viable idea to do a 12 fixed speed version of the camera and then the 1/1000 increment selector could be a add-on accessory. I got a viewfinder for my CP16R now and just need to finish other projects to be able to concentrate on it.

you can build custom batteries from separate 18650 cells for example. They are very affordable and you don't need much additional stuff other than the battery holders if having some sort of protection circuit for them or just being very careful. One can build a in-battery or a external charge indicator using a small microcontroller circuit. for example one which has couple of led lights which indicate the remaining battery power in percents and cuts off the power or gives a warning if the battery voltage goes too low. adding a oled display is possible as well though will be much more complicated.

from the looks of the Bescor converter I would believe it being made for something like 2A maximum current or something like that. Maybe they have specs somewhere to check it out. It is much easier to put two 14.8V batteries in series to generate somewhere around 30 volts and then use a easily available generic step-down switching converter to convert it to the voltage you want, for example the 24 volts. There is adjustable converters like this available for cheap (in eBay for example) and there is lots of different models with for example 10A or 20A or 30A maximum current handling. Prices starting from couple of bucks. Building a step up converter is not super complicated but because there is tons of different step down converters available which are perfectly suitable for this 24v Moviecam task I would use the two battery approach and the step down converter instead. Helps with the current limitation of the batteries too

basically anything which has capability to shoot raw stills in at least 5k resolution and which has fully electronic shutter option (to avoid wearing out the mechanical shutter). I used Panasonic GH4 for my tests because I have a leftover one which I rarely use for other stuff. The simplest wired remote controls are pretty easy to hack if having basic electronics knowledge. Depending on the sensor system you may need to do some basic programming to be able to trigger the camera at exactly the same frame position but using a intermittent mechanism makes this way easier so you may manage with for example a light sensor which triggers the camera after certain time period after it sees that the shutter is open and the light is coming through the aperture. needs less mechanical modifications. alternatively you can install a sensor somewhere to the projector mechanism so that it triggers the camera when shutter is in the middle of the open phase. A hall sensor or a optical one would do. This arrangement has the advantage of not needing any basic(arduino or other) programming skills but will need more mechanical work to function correctly

yes a mitchell or a bh movement would work well for this. it is much easier to improve image stability mechanically than try to handle the extensive instability caused by bad mechanics (my first prototype had vertical instability of about 1/50 of the image height which is massive. one should have something like 1/500 or 1/1000 for it to be useful. One could take a movement out of Konvas or similar low cost camera to get basic intermittent movement relatively easily. arranging the film path so that you could use a reasonable quality led light can be complicated with camera movement based designs but if machining your own pressure plate and redesigning the mechanical linking of the film transport to allow enough room for the led light, then why not. Making a basic film movement is not that hard either if just having enough time to adjust it correctly