View Full Version : Rai & Markus' "Drake" HD camera
October 25th, 2004, 11:51 AM
Important moderator notice:
This thread originally was part of a much larger thread to build
our own HD camera's. For various reasons it has been split off
into its own thread. Please keep in mind that some of the earlier
conversations may read a bit strange due to this split (some
parts may exist in the old thread, etc.). You can find the old
thread in the following place:
Thank you for your consideration, back to our regular broadcast:
My name is Markus Rupprecht. I'm the producer/director to the fantasy movie project Rai refered to. The camera creation team agreed not to participate in these hundrets of "soon we will have this and that" threads but wanted to complete our development, get a prototype runing and do some serious work with it before spreading the news. So that is why he didn't tell a whole lot. On saturday and sunday we had a first time shooting, both indoor and outdoor scenes, day and night with the camera and are curently processing the material.
We build the camera to match film workflow, so sound is recorded seperatly on DAT, we don't have a sync connection. The camera produces a "beep" sound and records red frames like a movie camera does. Scince the technical informations are stored in the file header of the recorded file we don't use a slate. A small detail, but it did increase the workflow very nicly.
We use a CMOS sensor that is capable of producing very natural images, no video look at all. I'm stunned by the perormence. It's cinema. The price we have to pay is that the cmos sensor needs a lot of light. Comparable to 25 ASA film material.
The night scenes were shot with a 1:0,95 f-stop lens with open aparture. It's a hell of DOF. And a hell of focus pulling. But we managed complex dolly and crane moves.
But this is definitly nothing for unskilled hobby filmers.
We also developed a rather complex software, not only controlling the sensor and recording the data to hard drives. The software is capable of eliminating issues like fixed pattern noise and gain noise. It does that already almost perfect, what is left in terms of noise is changed to random noise, that looks like film grain, so not too bad.
Scince we record uncompressed it is possible to run several algoryhms over the material to enhance the quality and to adress certain cmos behaviours in certain light conditions. That is done in the "processing" phase that is currently running in the room beside the one I'm sitting in now.
But this is just technological stuff. What we now have is a instrument to tell stories. To have a indi budget and shoot cinema quality with all the benefits that digitally recorded material has. And the best is: it works.
Unfortunatly I will now leave you allone with those informations. Please be patient, soon we will release some images.
October 25th, 2004, 12:54 PM
Is this camera shoulder-mountable, or is it thethered to another device (PC somewhere)?
What CMOS chip/framegrabber combo are you using?
How much lag do you have in your preview, and do you use a viewfinder, or again, is everything externalized on another PC on-set?
October 25th, 2004, 02:02 PM
sync is indeed no problem in terms of "keeping sync" but you need a starting point. That's why you use a slate - to record a "starting information" on your recording devices.
As I said, we will post detailed informations soon so we won't answer all questions directly but yes, the camera is portable, it can be mounted to a tripod or steadycam or whatever, it can be oberated with batery. It has a 17" viewfinder and the software does allow realtime preview in full 720p resolution, so no lag.
We did timing critical pans and focus things and it all worked well. We don't use special grabber cards, just regular PC parts and buses. Well, we searched a while to find good components that are not power consuming and had to to even build some internal devices. I'm no engeneer, so Rai or somebody else from our crew will propably answer specific questions in the future. But let me tell you there won't be something like the "10 steps to build your own HD camera" tutorial. It's complex like hell.
We tried several industry camera heads. They are all good in theory. And when you have perfect daylight, life is beautiful. But it wehen it gets darker you get into trouble. So between throwing pieces together that can record 24 frames per second and turn the thing into something that can record 24 frames in cinema quality and that allows a film workflow did cost us more than half a year with five people, all pro's. When all the problems in the beginning showed up I decided not to do the self development of such a camera and instead to buy the new JVC 720p camcorder. Well, it sucked more than the problems we had to face. You know them, this forum ist full of problem descriptions.
So my advice to anybody trying to do something similar is: Better be serious in what you want and what you want to do with it. And that should be making great films, right?.
October 25th, 2004, 03:29 PM
How long were your longest takes?
If you're not going more than 1 minute, you'll never notice any drift. It'll typically take a long time (at least 10-20 minutes) for drift to occur, depending on how accurate the camera's fps is.
BTW, how did you get by with no framegrabber? Doesn't the camera have to hook up to the computer somehow?
Also the ProCam 3560 seems to be rating at ISO320-400 at 0db gain. According to Altasens, ISO1600 is at 18db.
Still curious what camera head you guys are using.
October 26th, 2004, 02:17 AM
For all german speakers among you: Under www.hackermovies.com you will find the first article of a series describing the development process of our HD cam. It's called "DRAKE". So if you don't speak german, go for the key word DRAKE, in the article you will find some early camera images, already done with a mobile prototype.
And to move a little away from all these boring technical discussions I'd like to introduce some beauty to the topic. The first processed shot from our premiere weekend.
It's divX compressed 720p, aprx. 8MB in size. Make shure you got enough screen resolution, watching it downscaled is no fun.
Here is the download link:
It's shot with a 25mm 2/3" lens f-stop: 0,95, about 4 meters of dolly / crane move and 3 focus points. It took just 3 takes to get it perfect. So I guess the damn thing is usable. Enjoy the pictures as we do. We've build the indi viper!
October 26th, 2004, 03:45 AM
Here is a pretty poor english translation,:
October 26th, 2004, 04:38 AM
Damn it's really a shame that I missed the last "DRAKE tech meeting" in september! :-(
Congratulations for finishing your first (test) shot! Looks really great.. however I still noticed (esp. in the background) this uniform pattern (is this still from the bayer filter?!). I already noticed it in the earlier frame grabs I've seen from DRAKE - afaik Steffen Hacker noticed this too. ;)
Nevertheless it sounds pretty darn cool and perhaps I can convince you one day to build one more :-))
October 26th, 2004, 05:52 AM
Hey, how are you getting such large F-stops? Are these custom-made lenses?
I do not know of any 2/3" lenses from Angenieux, Zeiss, Optima, Cooke, etc. that open up that wide.
I'm trying to make some sense of this whole thing here and there's no information about what camera head, how you've input into the computer, what lens you're using, etc.
Please don't blame me for being confused, again, I'm just trying to make some sense out of this whole thing, and the details still seem quite sparse.
I'd be very curious to know what the specs on your machine (digitizing computer) were.
Thanks for uploading the clips though, greatly appreciated :)
October 26th, 2004, 06:16 AM
Just took a look at your movie there, it looks very nice, but I think I may have spotted one problem that I think may be giving you that "ASA 25" rating you're talking about.
Now I have no idea about the technical details concerning your camera because you haven't listed which camera head you're using, so there's no real way of telling. But Juan could probably back me up on this one just by watching the film.
Those clipped over-exposed highlights should not be happening. You're getting a very harsh clip which tells me that you may be using the top 8-bits of the image, or that this is a "10-bit" camera, but there's so much noise in the lower two bits that it's essentially an 8 or 7-bit camera (like the Micron chip we first experimented with here).
With the Altasens for instance, you can come away with a VERY dark image, that meaning that the white chip on a Macbeth is exposed at IRE 10, (yes, ten!), and still have a perfect image (there would be some noise of course, but it's equivalent to using a Canon DSLR like the D60-that's my point of reference-at ISO400 or 500, so nothing to worry about). That's because the Altasens for the 4095 values it encodes at 12-bit has greater than 11-bits of real-data (only the bottom 3 values are noise-that's from the head engineer over at Altasens).
So you may be only getting ASA 25, but it's not because the chip itself is insensitive to light, but you may not be capable of taking advantage of the lower bits due to noise and other artifacting or non-linearity (again, not knowing what the chip or camera head is, there is no way of telling).
BTW, how did you get around rolling shutter "skew"?
Another thing, that "uniform" pattern, that's not a CMOS issue, are you using a ground-glass for 35mm DOF simulation? The "uniform" pattern looks too out-of-focus to be fixed pattern noise.
October 26th, 2004, 06:45 AM
Hm are we talking about the same noise here? Maybe you have some post processing activated in your divx decoder filter ;-) ... I made these two captures from the avi and upsized them 200%:
This sort of "ordered dither" pattern is clearly visible in all bright but not near-100%-white areas of the video - and it is fixed and not moving! (can't be from some ground glass)
October 26th, 2004, 07:22 AM
Okay, nope, you're right, that looks like something from a de-bayer filter, like some of the pixels didnt' get interpolated or something odd like that.
Actually on my CRT monitor here, I didn't see that stuff, but stuff like that can get lost on CRT's. Are you on an LCD?
Also are you guys using variable gradients, or some other form of bayer demosaicing?
October 26th, 2004, 07:54 AM
I knew that everybody would jump on the fixed pattern noise thing. Oh, well. We know that it's there and we found a way to eliminate it. thing is that for every gamma curve you apply during shooting you have to define a huge set of paramters to get enough data so the software is able to filter the noise out. We have to do it ones for all time to come, but the calibration module is a fresh feature, so for the weekend we couldn't feed in all the data we would have needed for a perfect picture. I decided to use a gamma curve on this shot where the paramters were not perfectly balanced. Well, after all it was a test. I was prepared to kick the entire material into the garbage. It turned out better than expected. And after all the quality of a movie shot is not only defined by technical issues on pixel level. Please, all you engeneers out there, keep in mind that those machines are ought to produce art, not sets of clean signals. Anyway I uploaded one of the test clips with already a rather good FPN removal in place. You'll notice the difference.
Find it here: www.drachenfeder.com/int/dach_after.avi
We need a little more time and more and more test runs. But the methods are proofed to be allright.
The cispnes in the shot is the way we want it to look. The original camera picture has more information in the dark parts than it seems. So, just be patient. We do our work and when we have the perfect set of comaprison shots we post them.
October 26th, 2004, 08:41 AM
@Jason ... yes I'm lucky to have an Apple 23" CinemaDisplay which sports a res of 1920 x 1200 - very nice tool for HD monitoring. :-)
@Markus - sorry... it wasn't meant as a critique in any way.. i just wanted to know what I was seeing there.. just tried to understand, how your cam works and what type of filtering you have to apply... any chance to see some real "raw" (totally unprocessed) data from the cam? Ok sorry again.. guess we just have to be patient until you've got time to post more details. 8-)
October 26th, 2004, 08:45 AM
PS: The new vid looks much better pattern wise.. however on a high-res display (TFT) there is still some slight pattern visible!
However: the image looks really nice and much more filmlike than any other HD cam in that "price range" (if you can say that already... taking into amount all the work that went into the project)
October 26th, 2004, 03:37 PM
thank you for the footage markus: question - are you using a35 mm adapter type device. with rai's help and original drawings we are working hard on the ocillating adapter prototype.
October 28th, 2004, 12:10 PM
<<<-- Originally posted by Richard Mellor : thank you for the footage markus: question - are you using a35 mm adapter type device. with rai's help and original drawings we are working hard on the ocillating adapter prototype. -->>>
Rai build already an oscillating adapter that has the quality of the P&S adapter. But we found a better optical solution for the DOF problem without a ground glas. Part of it is really to understand the workflow to handle lenses and the light in a scene in a way that allows shots like the one you have seen. Focus pulling, for example is even more important when you open up the aparture. That's what you need to do, to compensate a smaller focal plane and / or smaller distance to the object.
You all know the physics behind the DOF, so I don't need to explain. Thing is, it's not only engeneer thoughts that led to our development It's pretty much cameraman experience and engeneer skills. We startet our quest by searching and buying the best ground glasses available.
But even with the best setups we lost 2 f-stops through all the extra glas. Plus, when you move the GG you loose sharpnes on a subpixel level. Now, 2 f-stops means you need 4 times more light in a scene. Imagine a night scene at a campfire. 4 times more light - there's a difference. And what good is it when you build a relativly cheap ground glas adapter just to end up attaching SLR lenses worth thousands of $? Plus the expensive light you need for your scenes? Browse through your camera bags. How many sharp lensen do you own that are as fast as 1.4 or better: 1.2? 50mm... big deal. But any wide angle among them? Any 20mm? 28mm? 80mm, any 135mm? That would be a set of lenses you could work with. Try to find a fast set of those and you rethink everything.
That's why we concentrated on developing a workflow that allows the camera operator to control critial DOF situations you get when you open up the aparture. That means you need to see your picture in full detail all the time (realtime preview with no lag in full resolution for one thing) and you need fine and total control over the Lens position (3 axis moving and panning head) and a intelligent focus device, helping the operator to pull focus as easy as possible.
Combine that together with the fastest lenses you can get and you can skip the ground glas thing entirely and expensive law suits with the P&S or the movietube guys.
October 28th, 2004, 07:34 PM
Are you getting exactly the same level of DOF as with 35mm out of your camera? If so how? What size is your CCD? You mentioned that you were using a fixed 25mm lens from a 2/3" camera. Yor lens had a great f/stop of 0.95. Is this how your getting your DOF? Seems like a pretty rare lens. Were you able to find any other lens to complete a full set (zoom or fixed) that had similar stops and if so what were they and how much did they cost? The reason why I ask is even 2/3" shooters are looking for better DOF then what their cameras traditionally offer them and are able to get it with the pro version of the Mini35 adapter - the Pro35. How does your results compare DOF wise?
October 28th, 2004, 11:34 PM
Jason wrote: "I do not know of any 2/3" lenses from Angenieux, Zeiss, Optima, Cooke, etc. that open up that wide. (f0.95)"
There are actually TONS of 25mm C-mount f0.95 lenses. Google them. Remember that that's not T0.95, it's f0.95... I have an Angeniuex 25mm 0.95 sitting right here, and it's what I used to shoot a lot of the tests on the Sumix camera.
Speaking of the Sumix camera, the stills that Markus and Rai posted look exactly like the output of my IBIS-5. Coupled with the fact that they say it's a 2/3" sensor, it's not very sensitive, and it's inexpensive, the only sensor it COULD be is an IBIS-5. If I'm wrong, please correct me.
Furthermore, it looks like they're using essentially the same algorithm as me for de-bayering. Perhaps it's even my software? The de-zippering process is what leads to the "grid" pattern that people have commented on. As Markus mentioned, you have to recalibrate the DAC offsets for both greens for every gamma adjustment, or else you'll get extensive gridding. Looks like they figured that one out the hard way...
All in all, this is absolutely nothing new as far as I'm concerned -- I was doing this months ago. I'm just waiting for the Altasens to hit...
October 28th, 2004, 11:43 PM
Brett asked "Were you able to find any other lens to complete a full set (zoom or fixed) that had similar stops and if so what were they and how much did they cost?"
When you deal with a 2/3" sensor, you're basically dealing with the same area as 16mm film. So your best bet is to find good 16mm C-mount lenses. Video lenses will have better stops, but they aren't as nice.
I have a "megapixel rated" 8mm f1.4 Computar lens designed for factory automation, but my 10mm f1.8 Cinegon designed for 16mm cleans its clock any day.
The 25mm f0.95 is crazy -- nothing beats having to stop DOWN to get to f1.4. :) However, you can't really get a whole family of f0.95 lenses; the 25mm is that fast precisely because it's that size. As I understand it:
entrance pupil = effective focal length / aperture
So a 50mm lens with entrance diameter of 50mm will be f1.0, a 25mm lens with a diameter of 25mm will be f1.0. (At least in pure theory, and without any other glass elements...) The problem is that those two lenses will cover very different areas -- the 25mm wouldn't cover 35mm, for example, while the 50mm would. There's some interesting resources around the web about super-fast lenses. Check these babies out. (http://www.abex.co.uk/sales/optical/fast_lenses/fast.htm) Stanley Kubrick famously used a Zeiss f0.7 lens (designed for NASA) on "Barry Lyndon" to shoot film in candlelight (back before 500asa film).
October 29th, 2004, 01:55 AM
>> the only sensor it COULD be is an IBIS-5. If I'm wrong, please correct me.
It's IBIS-5A. But you're right. When you get into the topic, in the end there is really not much choice.
>> Furthermore, it looks like they're using essentially the
>> same algorithm as me for de-bayering. Perhaps it's
>> even my software?
Just a second here. So you wrote a software to debayer the images. That's absolutely neccessary, nothing new, you are right. And it's not your software. I don't know who you are and what kind of software you write but accusing someone willing to share experiences to a community right away to illilagaly having used intellectualy property is a good example why it's a better way to not diskuss every single step you do in a development in a public forum.
Yo may edit your post and remove this sentence.
Now. Did I miss any official competition here? On who is the center of the indie HD world? As I see it, as many approaches we have, as many camera concepts we will build. Ours is build to shoot feature films and that is exactly what we started. I'd say this is something new. I don't know of another project doing this. If so, please contribute the information. It would be of great help to share experiences We have one year of hard shooting labour ahead of us and we also discussed the possibility of going for another sensor (Altasens if available). The camera and software is build that way to allow it. But thing is. Today, Friday 29. october 2004 there is no Altasens you can buy in a store. Tomorrow we continue shooting. So we take what we have. And we are willing to share our experiences. If you feel the need to argue, please mail me in private: email@example.com
October 29th, 2004, 02:11 AM
Hold on Markus... You're jumping to conclusions here. Ben has been on these boards for a while, and just for the purpose of pushing these developments he bought a camera, and experimented with some debayering tools that run within After Effects. He made the beta-versions available to the board, so it doesn't look like he's accusing anyone of anything.
He's merely pointing out that what you have been sharing with the board isn't really all that much...
And if you read these threads, you'll find that there's actually a couple of people working on developing a CMOS camera for feature production. I don't think many of us here are thinking of going newgathering with it...
October 29th, 2004, 02:22 AM
Hmmm... so it's IBIS5?! That means it's "just" 1280 x 1024 effective pixel res in b/w ... and because of the bayer color filter it's effective resolution is reduced even further. :-/
And of course it's still 2/3" (same as the Altasens) - so without an 35mm "adapter" you still have the same DOF basically as with almost any ENG cam...
The only real benefit I can see here is that you can record uncompressed/raw and have more control over the image (colorwise etc.) ...
October 29th, 2004, 02:32 AM
Thank you Markus,
you say it: You and your crew will continue shoot with this camera. We will do a lot of things together, but the point is: This camera work.
Ben, yes it is a IBIS5A (not IBIS5), but it is not a SUMIX, not USB2.0 or Cameralink, and it work with global shutter, and have a quarz clock for exacly 24fps, so there is no audio time shift. And there are some other special features.
Details you will found on a new webside.
October 29th, 2004, 02:51 AM
yes, and no...
I am not the man who write the software, but i know, you can select a lot of debayer codecs on the fly. But its just for the viewfinder. We write the RAW data to disk and thats why you can select also different debayer codecs in the post.
We have more control over the image, because our camera head have a different hardware inside. You must know, the Sumix have a color problem at high speed, because the SUMIX use the A/D unit on the IBIS5 Chip. But our hardware use a external A/D unit with some other nice features.
But we build the other hardware not only for this IBIS5A. We work on a solution with a real full size 35mm CMOS sensor and 1024p. But this is another story.
October 29th, 2004, 03:04 AM
Sounds interesting Rai! :-)
One more question: how do you capture the raw data? It's written to HDDs afaik... is this an extra unit (including a mini pc)? And how do you power it?!
Thanks! I'm curious as anyone else here is I guess...
October 29th, 2004, 03:37 AM
we are getting there. Let me write my articles. We explain what problems we faced and how we solved them. As I said, the camera is mobile. It's a box that runs with batteries. Just be patient. I don't have much spare time right now. Besides: if you still got connections to this HD cinema we should think about a little test screening...
October 29th, 2004, 05:37 AM
Markus, of course I understand that you're in the middle of shooting right now etc. and don't have much time on your hands - I sent you an email just a few days ago (and the rest of your team). Currently I have quite some spare time (thankfully ;o) again and also could help out with things like software dev. etc. (if help is still needed) and/or would like to visit your shoot (if you're ok with that) - and have a first hand peak at the cam :-) ... and of course the test screening would be an option too! (and I still got my own 720p projector ;o)
So don't hesitate to contact me!
PS: I already guessed it should be portable.. with "only" 720p (and 8bit as you said sometime) and perhaps losless compression the datarate should be managable with something that still remains portable. However with "real" HD (i.e. 1080p or even more) this should become more of a problem...
October 30th, 2004, 01:55 AM
Yes, Markus shoot with IBIS5A. But not IBIS5 and not with Sumix.
Have you ever see a video like this:
i think, next week Markus will send more
Show me a altasens clip. Show me a working camera with altasens. Okay, you can wait till the year nnnn. But all we wanted was a working camera NOW!!! And it work.
I say it again: There are worlds between Sumix and our Camera hardware. Sumix use a bad A/D, this produced dark pictures and bad colors and can not handle global shutter. We know this, because we tested also the Sumix camera (And others and others and others).
We found a german manufacter and developed together some special details for film making. And one of the next steps will be real big sensors.
The other side ist IBIS5A and DOF. We go a new way. We use ultra fast lesens to produce DOF near 35mm. You must know, with ultra better F-stops you have the same like bigger sensor sizes. DOF use F-stops not T-stops. But this part is optical engeneering and here is not the place for this.
At last. We build a camera design. Case, case-parts and also a software design.
For the big DOF sensitive, we build follow focus systems and more. We working on the web side. You will found all details there.
October 30th, 2004, 04:23 AM
Definitely impressive footage, especially considering that this is just the start. And you seem to have a team that knows what they're doing and have a firm grasp on what it takes to develop all these pieces into a usable tool.
Which leads to the next question: do you have plans to sell these camera's in the near future? I think many of us would e interested...
October 30th, 2004, 05:28 AM
<<<-- Originally posted by Barend Onneweer :
... the next question: do you have plans to sell these camera's in the near future? I think many of us would e interested... -->>>
Its not only my company, but i can say yes. And i add a quotation from Markus: "...be shure, we did it indie style..."
October 30th, 2004, 05:28 AM
We're not criticizing your move to purchase the camera when you did. We're talking to the people who are just about to purchase a camera AND then planning on upgrading to the Altasens. A little bit of waiting and they can have the Altasens and skip some of the pain and frustration of using an inferior camera chip.
Also, if I may ask, how much did it cost to machine your enclosure? I'm looking at going to a machine-shop for my enclosure, and was wondering how much it might cost (ballpark figure).
October 30th, 2004, 06:02 AM
<<<-- Originally posted by Jason Rodriguez : ,
.... A little bit of waiting and they can have the Altasens and skip some of the pain and frustration of using an inferior camera chip.
Look at the sample video, and you understand why we far away from frustration.
I say it again: After testing the SUMIX camera we send it back to SUMIX. Like Ben said, it is absolutely useless for film making. Also every other camera with rolling shutter.
...upgrading to the Altasens? My comany had ordered a altasens evalution kit. But after all test with our camera, like this sample video from Markus, i see not a reason why we change our camera. Yes, we will test Altasens out, but its not the only chip we will test.
And if you see the Altasens price, you will have a reason for frustration. A little bit more and there are Sensors near the new ARRI range (with full 35mm size).
October 30th, 2004, 06:08 AM
More shooting experience, more clips.
it's an extreme DOF shot. Today divx is not my friend and I didn't find the best settings so I also uploaded a still image as a bmp file
And we did the other night some shots. So here are two critical night shots to show lowlight situations:
(the "effect" comes from the hot air of the fireplace you saw in the shot before)
The shots are not color corrected. The chip tends to green. Well darkness is really the enemy of IBIS 5-A. But anyway the test proofed that it's usable with a decent A/D hardware. The noise is tollerable under the given circumstences. The scene is illuminated with two 800W Arri Spots from the roofs, each 4 meters away, turned half way to flood, one 200W spot for the doorframe, turned to flood plus the original torch.
The Chip was set to 3db gain and we applied a cine spline gamma curve to pronounce the shadow parts. I don't like those harsh contours to seperate foreground from backround so we tried to establish different grey tones in the scene. I think it worked well. Don't know if I would have been brave enought to try this with some Kodak Vision material...
Last night we found a very good extreme lowlight setting and I will do some test shots tonight, like somebody will hold a candle in his hand and nothing else. We will see how that will come out. Also we will do a small action scene tomorrow with lots of movement and motion blur. The 24 frames clocked along with the shutter settings allow a very nice motion blur, that we saw already in the tests.
Any other ideas to define critical shots to try? (Preferably ones I can use for real scince the shooting shedule is tight right now :)
*Relaxed waiting for Altasens
October 30th, 2004, 11:49 AM
Hello Markus, Just downloaded your shots. Looks good. Later I will get them over to the HD system. Yesterday I showed your footage to a group of filmmakers and some thought that this was a clip from a trailer. I did a DVD of those two clips and the one with the lady looks amazing.
Of course the lines were visible, but I'm sure that is not a normal thing. What is it by the way! It reminds me when I use rear net filters and the back focus catches it or if light strikes the net grid on rear lens filters.
October 30th, 2004, 12:31 PM
The "lines" are one of the flaws of this chip. It's connected to the color filter. So if you don't be careful you have them in the image! Fortunalty there is a way to filter them out with mathematics. I couldn't believe that it't possible at all without lossing "quality" but it works. Currently we have a test setup with a sliding bar in the camera control software, so if they show up you can control certain parameters while recording. Scince the raw data goes through a development process afterwards you can apply all kinds of those filters before storing the raw data as an uncompressed avi to the editing PC. So basicaly all shots have lines, that's important to know. In the shot with the lady there was an aditional problem with the bright areas. In pure white you need a differnent algorhythm then in dark parts. When there is time I "render" a celaned up image from the converter software.
We found out that it's best to think about the contrasts of your shots while shooting and to apply the right gamma curves. If you do too much in post the problems may show up again. Luckily the gamma path function got finished so now we can define all kinds of setups. Like I said with the nightshot. I wanted to pronounce the dark parts and changed the gama curve to a point were everything I wanted was in place. No we can do the color corection and apply the "look" (like the bronze look in the "take1" shot) without having to squeeze the gamma curve too much.
I belive that the "online" gamma curve adjustment is in general the best method, indipendend on the image sensor used. So I asked the programmers to make it usable in the recording software and controllable on the camera screen while shooting. And it works...
October 30th, 2004, 03:57 PM
I see that you are talking about software, is this software somthing you are developing I see taht it is propietary of the cmos interface, could you be more specific about your seup and software availavility.
October 30th, 2004, 05:22 PM
markus ,rai the clips you have provided are beautiful .can't wait for all the detail's of the camera . the clip of the woman looks 3d
October 30th, 2004, 06:21 PM
Just curious, what is the bit-depth that you're recording the RAW images to?
I'm wondering because when we've been recording 12-bit images off the Altasens, there's no need for on-set color-correction and application of gamma curves to preserve anything for post since there's so much information in the camera picture (at 12-bits only). This is the same methodoloy that the Kinetta is using, everything on the hard-drive is non-gamma corrected and the complete RAW information from the chip. So whenever you want access to that base information, then you have it, there's no need for the application of any final color-correction on-set. Just shoot and go, and then in post figure out what exactly you want with gamma curves, etc.
So again, is your post path a complete 12-bit system, or is there some downsampling somewhere along the line? In fact you guys still haven't told us how you're hooking up to the camera, whether it's GigE, CameraLink, USB, Firewire (400 or 800), etc.
October 31st, 2004, 03:42 AM
I would love to see pictures made with the altasens. Could you also post some material?
But 75MB/s is just not practical. We will shoot a 200 minute feature. Let's say we are good and keep a shooting ratio of 1:5. That's 1000 minutes of raw material. That is a lot of disk space. Too much for now. Especially if you backup everything for safety reasons. Time is working for us and soon we will spell TB as easy as we do GB but the time is just not there.
We discussed it back and forth and decided to go for the 8 bit depth in recording. Shure if you can record a 12 bit image and higher resolution you are a happy man. But then you end up with this hight budget viper stuff. How to handle the data? Super quality but a pain to operate. So what Rai and I are describing is not what stands on our wishlist for christmas. It's what we use today. From the practical standpoint of a indie film production. And that is not studio work or comercials but long feature with original locations and all that. Outdoor in the middle of nowhere. It has benefits when you exchange the "film cardrige" in the camera only ones per hour and a battery pack operates it for 3 hours. Those high datarates mean parts in your "portable computer" that consume lots of power. That is really a practical point to think about.
Another practical problem is noise. We needed a while to design a case that alows sound recording on set. No fast laptop without hearable fan. If you work with Laptop you can put it a few meters away. O.K. what about crane shots or steadycam? Any handcamera becomes a challenge. We treid it, it sucks. Start to optimize your system for noise and everything will change.
So all in all that is really an advice I now can give by looking back on the path we've been going. Don't see one technological element isolated. Not the DOF, not the resolution, not the bit depth of the image. When everything does not lead to something that you can put on a normal tripod with the workflow of a super16 camera engeneers might be happy, but cameraman and directors wont.
But I do hope realy much that one of the "high end" self build HD cameras will come to a point where someone will shoot a great film with. Just to proof that art and creativity beats economical aspects 2:1
October 31st, 2004, 06:14 AM
If you need the recording/storage to be on-camera, the limitations you describe are very relevant. You'd need a low-noise, low-powerconsumption, portable solution. So you end up shooting 720p24 at 8-bits. Which can still deliver great image quality.
But the other approach could be to separate the storage from the camera. Considering we'll be having "video assist" with playback on-set - we might as well position the recording device with the video monitors on the "video assist cart".
That would allow for a bigger (soundproofed) box, with a SATA raid on board. You'd need bigger batteries, but you'd probably be able to use car batteries on the cart (in case there's no wallsocket around...). The camera head would be connected using GigE - which allows for long cables.
Now you're recording 1080p24 at 12 bits. And in return, you don't need to be as critical about exposure and filtering on-set - which could be a huge time-saver.
And all this would obviously not be at the cost level of a Viper + HDCAM SR deck.
October 31st, 2004, 01:15 PM
I would have to assume a laptop would work at this level? Am I wrong?The datarate you're talking about is only 21 MB/sec, which a 7200rpm laptop drive should be able to handle.
However, I'd urge you against both 720p and 8bit. There are no good 720p cams out there that I'm aware of -- at least not on the level of the Altasens. If you do go with the Altasens, you'd probably want to see the whole FOV, which I guess necessitates 1080p. *
8bit is a nightmare I cannot wait to give up. I hate being locked into one "look" with only superficial image controls. The color tools available in your capture software are probably not as nice as the ones in your compositing app, so why not wait to do the color correction? With 10bit, you can shoot clean and finesse later...
That puts you at 59.3MB/sec (1080p @ 24fps, 10bit). I'm looking into RAID'ing a few mini hard drives together to acheive that datarate with minimal battery usage. That's what the Kinetta does -- they use the same hard drive line as the iPod.
* STEVE: Do we know why the FOV-protecting 720p mode on the Altasens is mono-only? There doesn't seem to be any good reason for it -- as you mentioned, as long as the software knew it was getting a strange bayer pattern, it should work fine...
October 31st, 2004, 02:11 PM
<<<There are no good 720p cams out there that I'm aware of>>>>
So what i am hearing is there are no good 720P 8-bit - 24fps @ 1/48th options on Altasens 2560 - 3560 or anyone else's cameras.
I have worked with HD at 8bit on the Sony-Varicam and other flavors. I agree not the top, but the audience never noticed a thing. My specs are not pushing the threshold and I can have a very small potable PC system built into a small battery driven package. Like others I feel for some if you put to much money into a system for filming narrative productions for HDTV etc it's not worth it. I have seen PD170 material look very good on the big screen considering the source is a very low res-horrible compression system anchored by a defective out dated 50 year old NTSC specs.
My D60-10D and Rebel images all look great at 8bit for what i need, and that is using jpeg compression. I have been privileged to work around the best imagery for may years here in LA and I have learned that the best is not always needed. So my 720p at 8bit spec will work for me ( also understanding others need better ), but now it sounds like this option does not exists. :-(
October 31st, 2004, 02:22 PM
I definitly don't want to claim 8 bit the ultimate goal. But I believe you judge it too hard calling it a nighmare.
We tried new sets of paramters and found one setting that is "pushable" best in post. I uploaded a few clips, some recorded with high contrast gamma, others with the new soft setting that preserves most contrasts. Those settings look really stable in post, the new white balance feature removed the "stripes" further. They didn't show up in color corection at all. But our viewing posibilities are limited. So if anybody viewing the clips on decent screens or beamers could tell us, if there are any visible problems still left, I would be greatfull.
Now I also would like to do a test with comparable setups to really see how much more detail the big HD resolution has compared to our 720p. I uploaded a closeup of a face and I think this is the best in terms of detail that our setup can bring. How much better is a 10bit 1080p picture? Any ideas how a setup can look like to find comparisons. Just calculating numbers seems not enough.
Anyway, here are some samples of todays new clips. One has some nice motion blur. Hope that still shows in divX
October 31st, 2004, 02:53 PM
Bravo Markus! just got your Clips- Going over to the HD system shortly. On the mac they look awesome. What is your Mbs with your camera setup?
Why is it that no one else has any clips to share?
Again great work Markus and your production team...............
October 31st, 2004, 03:33 PM
I have worked with HD at 8bit on the Sony-Varicam and other flavors.Michael, just one more word of warning: let me reiterate that the color controls available in capture applications are not comparable to the controls available in a Panasonic Varicam or your compositing software.
8bit is only acceptable if you can get 90% of the way to your image during capture. That's simply not the case with the caputre software. Markus has some great looking images, but keep in mind that the color issues are not by choice -- that chip will simply not output bright, full color images.
Anyway, whatever you do, I'm sure you'll have fun figuring this stuff out (I know I have). I wish I had some sample clips up, but my server was hacked recently, and I haven't gotten around to uploading new stuff...
October 31st, 2004, 03:40 PM
Your footage looks great. Thought i was watching Lord of the Rings for a moment :)
I want to know more about your setup.
1. What camera are you using. (link to retailer)
2. What are you capturing on and with what connection.
October 31st, 2004, 04:07 PM
Hello Ben, do you mean like a, for example this Viper raw output with out all the bits.
I am sure you have seen this. But here is a 8bit version of a raw Viper file: http://pixelmonger.com/vip1
Color timed version: http://pixelmonger.com/vip2
November 1st, 2004, 02:51 PM
well. This is not a lady with green eyes and red hair in bright sunlight but a blond lady with grey eyes in diffuse light, but whoever has access to this viper sample picture - could you mess around with this one here and give a hint on how close or far away that is in terms of detail and "post beahviour" to the viper pic?
November 1st, 2004, 03:11 PM
Markus, are you desaturating your images?
When we say that the Viper is green, (and you may already know this), but the reason it's green is because the Green CCD is more sensitive than the blue or red CCD's. But please notice how saturated the green is in the viper. That means that there's a lot of color information to extract. As of right now there is practially no color information to extract from that pic you just posted, so I'm not quite sure how to add accurate fleshtones, etc. I'll try and see what I get, but I'm not sure if you're desaturating your images first.
November 1st, 2004, 03:26 PM
it was really almost sunset, lowlight. And low colors is one of the flaws of the full frame shutter. Anyway, it's a raw image, I'm curious how far you can go in post until problems show up.