I sometimes need a nudge from a reader or Ikonoskop owner/user to get me out of my writer’s block.
I’ve seen some confusion and I have been getting some questions about the on-board histogram.
And I’ve also experienced some people who seem to think that shooting in raw means you don’t have to think about exposure. Well, let me assure you, the importance of exposure is as important as it has ever been.
Ikonoskop has always had this link on their site to an Adobe PDF about raw capture, linear gamma and exposure. This document is a very interesting read and if you take the time to let it sink in, you’ll notice that the on-board histogram makes a lot of sense. I consider it as my primary tool for setting exposure.
I’ve noticed that not everyone knows how to read a histogram.
What I aim for is to have the highlights as close as possible to the right wall of the histogram, but without touching it, so without overexposing.
The example below shows you what I try to avoid. In the bottom right corner you see the graph touching the right wall. In this situation I would lower exposure. Just enough to keep it from touching that right side.
I’m sorry for the poor quality of this image (it’s a hug blow-up), but I don’t have the gear to be able to shoot something better through the dII viewfinder (if someone can, please do and send me an image).
And remember, you can only see this histogram through the viewfinder.
This does NOT mean:
As you can see, the histogram has almost no read-out in the middle, because the scene has pratically no color. But if you would overexpose a colorful scene by more than 2 stops, you’ll start noticing the same kind of histogram, where color information is getting lost. the graph will become flatter in the middle and the end result, will look like … well … not pretty.
Although I’m still amazed, every day, by how much information in the shadows the dII captures, too much underexposure will lead to blocking and banding of the shadows (if you want to crank ‘m up in post).
If you want to capture a scene that holds more than 11 stops of dynamic range you’ll have to decide what to drop, shadows or highlights. And this, to me, is where the false color function comes in. I’ll look at which highlights get pink and decide if it’s OK that they’ll be completely white in the resulting image and I try to figure out if there’s a risk of causing aliasing on the edges of the highlights.
The on-board histogram is in my opinion an essential tool. It’s just a shame that it can only be consulted through the viewfinder. And it’s a pity that it’s in this half transparent white, ‘cause in very bright conditions, where reading overexposure is crucial, it’s quite painful to figure out how far you can go on the highlight end of the histogram. I don’t mind an overlay, but it should be opaque and preferrably not white. And I would be thrilled if somehow in the near future, there would be a way to get it out of the viewfinder and into some other monitoring solution.
I always welcome comments, in fact, they make it more interesting to keep writing. So please, feel free to participate.
A couple of weeks ago Ikonoskop released firmware 1.27 (and 1.28 on March 3rd with bug fix for 24 and 30 fps metadata). This is a significant update in my opinion and as important as 1.24 was in March 2012.
A lot of whishes from users from all over the globe have been granted in this version. Let’s look at some major improvements.
1. Information Read-Out
Additionally in the viewfinder and side OLED, a constant information read-out for card number, take number, db setting and shutter angle. Especially the db setting is a crucial addition in my opinion. I’ve seen too many times that users weren’t aware of the fact that the camera wasn’t set at 0db.
Hopefully in the next version we’ll have this information read-out as an on and off switchable overlay in the HD-SDI out.
2. Sensor Tune
There’s now an on-board pixel mapping function to take care of ‘bad’ pixels. You already know ‘the camera loves you’, but it’s also a bit of a liar in this firmware. It says the camera should be cold when starting the sensor tuning, but that’s a mistake. The camera should be warm (let it warm up for about 20 to 30 minutes). The instructions in the user guide are correct.
3. New gamma curve for viewfinder and HD-SDI out
Cleaner, pretty accurate image in viewfinder and HD-SDI out. It’s time to start up tests, recording directly from the HD-SDI again.
A lot of the solarization we used to get in low light situations has been improved and the brightness is now pretty accurate and in line with the exposure of the raw material.
4. Black Level
As suspected, putting the black level back in the metadata has resolved the magenta issue in the shadows in the decoders that were suffering from this. Ikonoskop took the black level out of the metadata some while back, because some decoders seemed to lose details in the shadows because of the black level. So Ikonoskop decided to give that power of setting the black level to the user. Unfortunately, depending on the decoder that was used, a lot of users didn’t seem to be able to get rid of the magenta in the shadows this seemed to cause.
So, in my opinion, having the black level back in the metadata and having the level tweaked, is a major improvement for easiness of the post workflow.
If you’re transferring via the the ExpressCard Reader, make sure to have the new driver installed. Otherwise you’ll have the old metadata and thus magenta in the shadows. This also means you have a choice. If you want to work with the old metadata and don’t want to change all your LUT’s and grades again, just use the old driver.
5. False Colour Function
I’m pretty proud of this one. In January during a phone conversation with Peter Gustafsson, as we were discussing the new LUT for the gamma curve in the viewfinder and HD-SDI out, I got this idea that it might be handy and quick to have this kind of false colour function based on the LUT. A couple of weeks later, they already had it implemented in the beta version for 1.27.
You activate it by turning the step/select wheel on the left side down. Overexposed pixels become magenta, underexposed become green.
The on-board histogram is more accurate, but I understand that I’m quite alone in my love and trust for this read-out.
My first time in Berlin and I can’t say I’ve seen much of the city except for some bars, 3 restaurants, a hotel and the inside of a theater. But it has been an interesting and fun weekend.
Markus Seitz - PHOTO: SVENJA KLÜH
I was there for a workshop (very well) organized by Digitales Kino, a platform created by Markus Seitz and Sebastian Böhm for making workflows in digital cinema more transparent for filmmakers through the organization of workshops.
The day was organized around 4 major topics:
PHOTO: SVENJA KLÜH
The main comparison was projected in 2K (DCP), and I found the three cameras held up fine without much surprises. But of course there was this main difference in esthetics by the 16mm lenses on the dII and its typically powderey, film-like look (yeah yeah - I am an Ikonoskop fanboy after all, but you already knew that). Maybe we should settle on calling it the CCD-look.
A lot of people seemed surprised at the quality of the structure, its richness and depth, of the dII image. I found the C300 had the least pleasing image in the interior and the RED the least pleasing in the exterior shots. I was under the impression that there wasn’t that much difference in dynamic range between the RED and the Ikonoskop for instance (but data sheets and a bunch of figures will surely prove me wrong).
PHOTO: SVENJA KLÜH
There was of course a test that showed the difference between a global and rolling shutter and you will not be surprised to learn that the dII won that one, even when stabilized.
We learned from the night shots and testing of high ISO settings that there isn’t much difference in what we can get out of the Ikonoskop and RED (keeping noise acceptable), but were amazed by how much we apparently could push the Ikonoskop more in post when working with a DCP. The Canon C300 is completely out of this league when it comes to low light situations. A Berlin street only lit by some streetlights in the rain becomes a daylight scene on the beach, with the C300 (“Hier kommt die Sonne”). It’s amazing how light sensitive this camera is.
PHOTO: SVENJA KLÜH
During Lukas Eisenhauer’s & Sebastian Höglund’s presentation of the camera there were some good questions by critical attendants about the memory cards. Some find the size they come in (80GB and 160GB) too small for the price Ikonoskop is asking.
So Lukas explained that at the time of the development of the camera they had to make their own memory card as there wasn’t one on the market fast enough to handle the necessary data rate. He said they don’t get much demand for bigger cards (maybe price related ;-) and that there are even users who prefer the 80GB cards (while looking at me). He got some flack for making the analogy with 16mm film reels and although nostalgia can be nice I think the person from the audience was right. It’s not because the dII is conceived as a replacement of 16mm film cameras that you have to copy all the limitations of film as well. But is it a limitation?
It’s true, I prefer the 80GB cards. I find 15’57” ideal for what I do (fiction, documentary, interviews) and never had a complaint from any of my rental clients. The transfer process is easier with smaller cards: it doesn’t take as long to fill one, so transfer can start earlier and transfer doesn’t take so long. And I feel safer, ‘cause if a card should die, I only lose maximum 80GB of data.
I own 11 80GB cards that are rented out all the time. They are used by all kinds of filmmakers. They sometimes come back scratched and with obvious signs of having fallen on not so soft surfaces. After a year of rental, they are still working as expected. I did have a card failiing once during rental. The data was not lost, but the card didn’t perform anymore. Problem was I didn’t update the firmware on the card properly. When it came back I recovered, transferred, deleted, rebuild the card and it still works fine as to date.
I wonder if this would be possible with an of the shelf SSD. Would they be this dureable? Would they survive being taken in and out of the camera hundreds of times? Would they survive constant transportation and occasional abuse?
So there was some great back and forth discussion with the attendants that continued into the next topic about workflow. People worry about the huge amount of data that is involved with this camera and how this affects and slows down workflow.
My workflow speed? To transfer 15’57” from Card Reader to computer it takes me about 9’10”. Extracting audio, another minute. Conforming in Resolve about 2’. Rendering to ProRes 4444 takes about 8’22”. Major key in getting this speed is Thunderbolt hard drives. I can recommend the Promise Pegasus R4, R6 and J4 and to fill them, you can use Western Digital Black Caviars. Of course, if budget is no issue, fill them with SSD’s.
PHOTO: SVENJA KLÜH
The topic that was the most interesting to me was the one about DCP creation. Very informative! It’s amazing if you hear the stories from these guys how frustrating this last link in the chain can be.
And they ended with a big suprise: upscaled 4K Ikonoskop footage vs. native 4K Red One MX footage. It made me smile and reminded me of of the second post I wrote on this blog. It made me realize that this camera and sensor still has a lot of wonderful years ahead. But I’ll let another attendant of the workshop speak, Sirio Magnabosco (on Vimeo):
The most wonderful part of the weekend was of course meeting all these wonderful people passionate about filmmaking. After the workshop we continued discussing camera stuff in a restaurant where we briefly met up with Andrew Reid, who showed us his Blackmagic Design Cinema Camera. When it got late we finally got fed up with talking about cameras (much to the relief of Markus’ girlfriend Svenja, I believe :-) and started discussing filmmaking. One of the topics was your nation’s most influencial directors.
Well OK, Sweden won, but that’s just because Lukas Eisenhauer is like a walking talking film encyclopedia.
And then there was Bob Ross!
When I got up the following morning and looked out of the hotel room I saw it had been snowing all night and my mind wondered off to the discussions, laughter and many ideas from the past two nights.
What is it that makes some people so passionate about filmmaking?
Maybe we want to distill the essence of the human condition and capture it in our art, I thought to myself.
Somewhat like making perfume. Moving images is our fragrance of choice.
Last December somehow turned out to be Filmschool Month for me.
Patrick Geeraerts, head of the editing department of RITS in Brussels, asked me to deliver a lecture on the Ikonoskop for some sixty 2nd year students (directors, screenwriters, multi-cam directors, …). I asked what the scope for this class would be and he replied: “Try to broaden their horizon!”
I don’t know if I succeeded in broadening there horizon but the lecture went somewhat like this: 1923 - Kodak - 16mm film -WWII - S16 - Eclair - Aaton - Ikonoskop SP-16 - Ikonoskop A-Cam dII. And of course on the matter of 16mm not only as an economic but also an esthetic choice. About workflow and more importantly about how filming is all about preparation and less about improvisation (some will find this debatable).
I also let Lennert De Taeye, film student at KASK in Ghent and by now experienced A-Cam dII operator, use a camera at his school for the purpose of shooting some acting exercices.
But I’ll let him speak for himself:
“In December the ikonoskop was used to shoot several acting exercises of third year film students at KASK . The school provided a RED 0ne (first generation sensor), but I’m not very fond of the image this camera produces. Luckily Joachim was so kind to provide the Ikonoskop A-Cam dII! After using it last year on a project I started to like the organic feel of the images from the Ikonoskop.
The first day I worked as a cinematographer. We shot everything, except for one scene, on a dolly, with the Vocas shoulder rig. This rig and the Ikonoskop work really well together, the counter weight on the Vocas rig is very effective to stabilize the camera movement.
The second day the Ikonoskop was equipped with a TVLogic monitor. The compact size of the Ikonoskop was excellent for the type of shooting: very close on the ground scenes.
The third day the Ikonoskop stayed on the camera tripod and rails. As director’s monitor we used a JVC monitor, the SDI-output gives a really nice image on the JVC we used.
Last day was another shoulder cam day, this time with a very compact rig (the Vocas was unfortunately not availble).
The data was transferred, and backups were made, on set with the express card reader to standard USB 3 drives (=fast!).
The little camera did catch quite some attention. But everyone seem to like the image on the monitor. (Some even surprised that this little camera could produce such a great image).
We shot with a set of Zeiss Superspeeds. Timecode generated from the Aaton Cantar. Resolve 9 and audio autosync is great! Little disadvantage you need to resync timecode every time you change the battery.
Too handle the magenta issue I set the raw settings in Resolve 9 to BMD Film. Add contrast and saturation to bring back the image as shot. XML for round tripping between resolve 9 and FCP 7.
Looking forward to see the film’s projected in the school cinema!”
- Lennert De Taeye
The latest version of DaVinci Resolve, 9.1, now includes CinemaDNG Input Device Transform (IDT) for grading Raw images in the ACES colour space.
This gives in my opinion, out of the box, the most accurate colour rendition of the Ikonoskop footage. Or as Jesse Borkowski said on Twitter: “I know! It’s like I got a new camera!”
To set it up in your Resolve projects, here’s a nice tutorial by Jesse:
If you were using the IkonoVinci.lut, be sure to deactivate it when using ACES.
In a prior post about Resolve, I wrote about my quick fix to get rid of magenta in the shadows and I mentioned it’s only one of several ways to get rid of it. All you want is a good starting point for your grading.
It’s in the grading that lies the power of the uncompressed raw CinemaDNG of course. That’s why you bought or use the camera in the first place. As Ikonoskop has always mentioned: the dII does the capturing, your computer does the processing afterwards. There’s no processing of the image in camera. No curves, no rolloff, no LUT. There are many things you have to be aware of to get the most out of your Ikonoskop footage. After all, it’s all about luminance and chrominance. And with the dII, it’s you who’s in control, not the camera. You not only need to be maybe the director, DP, editor and colorist, you’re also the guy or girl in the lab developing your film.
That doesn’t make it easy to get quick results at first of course. It needs studying and a lot of trial and error. But after a while you get better at it, you’ll make your own presets and LUT’s you can trust and most of all, you will have control over your images like you’ve never had before.
So if you see footage out there with whites that aren’t really white, with harsh highlights, with a magenta or green or blue or whatever cast in the shadows, and it’s difficult to assume it was an artistic choice, don’t go blaming the camera or the sensor. They’re quite alright actually. It’s all in the hands of the one who’s responsible for the grading.
Just watch this to see how nice Ikonoskop footage can look:
But anyhow, I started out by mentioning those other ways to get a good starting point in Resolve, in addition to my IkonoVinci.lut. Apparently not everyone finds it so easy to install this LUT. And I really suck at making tutorials.
So I’ll leave the making of tutorials to people more capable and this leads me, quite seamlessly, to another way to get a good starting point without magenta in the shadows. It’s written buy a new Ikonoskop owner, Jesse Borkowski, and it’s a great tutorial. I’m not quite sure about using the Linear Gamma setting, but that’s just my opinion.
The third possibility I’d like to share with you is something you might like if you’re used to working with LOG style footage. Like ARRI’s LOG C or Sony’s S-Log. Set the color space, in the Camera Raw settings (see video above) to BMD Film and be amazed by the details you’ll see in highlights and shadows without any strange color casts, but with the all familiar washed out LOG type ‘look’. Now drop your familiar LUT’s on your footage, and I’m guessing you’ll find that the result will be pretty close to what you’re getting with other LOG footage.
To celebrate this new year, a little tribute to the past year. It’s been quite an eventful one for me and it left me utterly exhausted. But nothing some hard decisions and time with family and friends can’t resolve.
It was a very educational year in regard to the Ikonoskop A-Cam dII. I’m thankful for all the people I’ve gotten to know, online and in person, just by sharing interest and passion about this camera and filmmaking in general.
Anyhow, this reel is a selection of shots from about 50% of the projects that were shot during 2012 on my dII’s.
They have been graded in DaVinci Resolve by me, solely for the purpose of this reel. The goal is to show a lot of different styles of grading. So I even changed style in the different shots from the same project. Except at the end, when there’s a little edit of the two girls walking and talking.
A big thank you to all my collaborators and clients.
Have a great 2013!
I’ve been trying out a lot of different workflows over the past year, based on system specs, software version, firmware version, etc.
Some people probably still think I work with, what I called, the CineForm workflow. But I haven’t used it since last summer, to be honest.
When I started using a dII I was still on a Windows 32 bit system. After a couple of months of working with CinemaDNG footage I choose a Mac Mini with a Promise Pegasus 12 TB Thunderbolt raid 5 hooked up to it, to function as a media archive and was hoping that Apple would soon release a new Mac Pro with Thunderbolt capabilities. I also anticipated a Windows driver for the Ikonoskop ExpressCard Reader.
When I started testing the Mac Mini, I was somewhat amazed by the power that such a silent little block of computing ingenuity could hold. After a couple of weeks I completely switched to the OSX platform, but was still using the CineForm workflow (via Parallels), because of its speed and raw format.
When summer arrived it was quite clear that we wouldn’t be getting a new Mac Pro, nor a Windows driver for the ExpressCard Reader. So I asked a hardware/software savvy friend to build me an editing system, with support for Thunderbolt and running on OSX.
Meanwhile, Adobe had released CS6 with SpeedGrade and for the first time I was able to play some CinemaDNG footage realtime on the Mac Mini and had insanely detailed control on grading. It made me forget about CineForm raw.
When the Hackintosh was ready in September my workflow got drastically changed for a second time.
Let’s look at a very personal assessment, on the 2 systems, of the three mainly used decoder/transcoder software for ‘developing’ and grading Ikonoskop footage:
I think it’s clear looking at the results above what, for me, at this time is the best tool to do the debayering, grading and transcoding of my CinemaDNG footage. But it’s also clear that it depends on the system you’re using. So on my main editing system I’m now working in DaVince Resolve 9 Lite (and that’s free folks!). When going mobile, I’d rather use SpeedGrade.
I haven’t integrated CineForm in the test because I don’t use it anymore and I didn’t get it to work in Parallels 8/Windows 8. But I tested the GoPro Cineform Studio Premium for Windows just before the summer and I was amazed by the new functionalities that were integrated to debayer raw footage. Dynamic range was much improved and of course it was no longer a command line tool. But sadly enough, no Mac version with the same raw functionalities.
But then there’s John Hable at 19lights.com. I already wrote about his Ginger HDR in my last post, but in the mean time he has been working hard to improve his plug-in and I’m astonished by what he has accomplished in such a short time.
This is honestly spectacular. There’s still room for improvement (f.i. better CUDA support) and John is working on this. But this is very promising.