Apertus, the Open Source HD Movie Camera
osliving writes "This article takes a tour of the hardware and software behind the innovative Apertus, a real world open source project. Led by Oscar Spierenburg and a team of international developers, the project aims to produce 'an affordable community driven free software and open hardware cinematic HD camera for a professional production environment'."
Oh silly me, I read the summary as "Led by Oscar [winner Steven] Speilburg..."
Is open hardware really that big a problem? It's not like opening a Fab is cheap.
community driven free software and open hardware
So it isn't open source, just community driven? Oh, that's right... the "community" can't agree on a term.
Didn't I read someplace that MPAA, in collusion with camera equipment manufacturers and the camera operators' unions, is looking to place patents on these devices so as to preclude competition?
Last time I checked, .mov was a container, not a CODEC.
A .mov file can use a lot of things. Quicktime 7 gives me PNG, JPEG, JPEG 2000, DV, DVCPro, Apple Pixlet, MPEG-4 and H.264 as video CODEC options. Older Quicktime versions would have offered me older CODECs too.
And what's JP4? Never heard of it. I sure hope they don't mean their camera runs on jet fuel.
Unfortunately, while the camera will have some interesting features and can do some things well, it will be hampered by an interface that only a CS grad student could decipher. Further development on future models will come to a standstill as the developers engage in fierce, unyielding debates about minutia. Eventually the camera will be forked into four different projects, with only one making it to market and carrying the same flaws as the first.
task. Research metadiscuusions
Reading through the article, I'm loving just about everything about this camera, except the most important part of all...the sensor, which is absolutely tiny. Forget about a camera for cinema, with a sensor that size you're going to be struggling to get it not to look like a webcam video. Looking at the company that makes the actual camera element, though, it looks like they also sell a model with a more reasonably sized sensor, but it can only do 5fps. If they really want to pass this thing off as a motion picture camera, they need to find a solution that will give them a big sensor at a respectable frame rate. Hopefully that will be possible in the near future, because the rest of this project looks downright awesome.
I don't get it. The crux of image acquisition is the lens and they don't include one?
I don't see how that contraption could possibly penetrate the production side of entertainment industry. What is the market for this device?
http://www.maxineudall.com/2010/02/should-economists-be-sued-for-malpractice.html
There was this startup called Red (http://www.red.com/)....
I love how the article uses crappy low-res, heavily compressed flash video to demonstrate the quality of the camera.
http://www.usa.canon.com/dlc/controller?act=GetArticleAct&articleID=3407
Canon XF300 / XF305 Camcorder. BBC approved broadcast quality camera. $7000.
SOLD.
This thing is trying to compete with the RED camera system and the 5D Mk. II. As others have said, the sensor is already behind. Everyone doing 2K on the cheap is using the 5D Mk. II as a video camera - it has a bigger, better sensor than anything anywhere in that price bracket, plus Canon's awesome lenses. The next step up is the RED system for 4K, which is just on fire right now because of its revolutionary modularity. This thing is pretty small potatoes compared to either of those two. It might be good for student filmmakers though. A school could buy a batch of them.
Or is this really the Elphel opensource/openhardware camera, and how Apertus hopes to add things around the edges. The camera is Elphel, as is the sensor and the software. The only thing that seems to be community-designed/built is the "rod" packaging, and maybe the battery rig.
The one that intrigued me was Stanford's computer science professor Marc Levoy Camera 2.0 Project.
bieber is absolutely right. What makes the Canon 5D Mark II amazing is the large sensor (even larger than 35mm motion picture film), enabling good control of depth of field.
The Mark II only captures 29 seconds of video though. Because of the sensor size, 35mm full-frame, I've thought about getting one. That or a Canon EOS 1Ds Mark III, but that's more than twice the price of the 5D Mark II.
Falcon
Should there be a Law?
So...it has the size of a medium format or even large-format camera...but the resolution of a DSLR that is 5 years out of date. Doesn't seem too impressive to me.
social sciences can never use experience to verify their statemen
A screw-mount lens ain't necessarily the one you want for production. On the other hand, there is some very nice tiny glass they make in c-mount.
This thing sounds like a video Kras. But with so many low-end video options, who needs something like that?
I can't be the only one reading the name Apertus and humming that tune, right?
The EU defines devices that can record 30 minutes or more of continuous video as "camcorders", and subjects them to special taxation. This is the reason for the 29:59 limit on SD.
Are you adequate?
Is there efficient transcoding from h.264 to WebM? The two codecs are so similar that it may be possible to transcode without decompression and the associated quality loss.
If there is efficient transcoding, that would greatly reduce the problem.
That's the same kind of sensor you get in a small compact camera like an IXUS, you'll never get anywhere near cinematic quality with that because of noise, lack of dynamic range and most of all the small sensor gives a large depth of field, so no fancy depth of field effects. For comparison a 1/2.5" sensor is about the same size as Super8 film.
For cinematic image quality (at least in terms of shallow depth of field) you need at least a 4/3rds size sensor (which is a bit larger than 16mm film), even better would be an APS or full frame sensor.