Slashdot Mirror

A lens passes light, in some sense it is a filter. The filter passes through information. Typically a lens is qualified in either lines-per-inch of megapixel. How can you check this out yourself? Just make a photo of a subject far away and tell if you can make a distinction between one row of pixels and the adjacent row. Typically the light blends because your sensor is better than the lens. unless you have some very nice (read: $$$) prime lens.

But in another subject. Why is this really news? The Elphel Eyesis 4pi designs are online for 5 years. http://wiki.elphel.com/index.p...

Elphel makes open source hardware/software cameras.

World first? Videocam?

Well it's indeed the world first open source and cinematic video cam.
Elphel is only the other open source cam but it's not cinematic.

Ok the heading is a little cheesy, but then after a fantastic uptake the crowdfunding
campaign has run out of breath a little and we need to generate some new buzz :)

Things did not change much (to the better, at least) - http://blog.elphel.com/2013/10/fpga-is-for-freedom/

Using their previous Eyesis camera Elphel did the backpack thing last year. Now since wikipedia already states they are the original designer of the high resolution hardware Google used in Europe, it is not really surprising Elphel is already ahead with development of new stuff. While last year a 360 camera + Inertial measurement unit + backpack was offered, the current development offers a 120Mpix 4pi camera where the resolution is equal in every direction, so no fisheye on the top! ...and this is all done with open hardware.
While the camera of Google is indeed a new one, I wonder how it compares :-)

Is open hardware really that big a problem? It's not like opening a Fab is cheap.

The hardware behind Apertus is actually Elphel. They make "open source" cameras based on existing chips. The video codec is performed by an FPGA (loaded with GPL-licensed firmware made by Elphel).

Here is more about the hardware.

Is open hardware really that big a problem? It's not like opening a Fab is cheap.

The hardware behind Apertus is actually Elphel. They make "open source" cameras based on existing chips. The video codec is performed by an FPGA (loaded with GPL-licensed firmware made by Elphel).

Here is more about the hardware.

Further development on future models will come to a standstill as the developers engage in fierce, unyielding debates about minutia.

Let's just ignore the fact they've been doing this for ages now.

Further development on future models will come to a standstill as the developers engage in fierce, unyielding debates about minutia.

Let's just ignore the fact they've been doing this for ages now.

The homepage for the Apertus project is cinema.elphel.com.

Details about the JP4 codec can be found here: http://wiki.elphel.com/index.php?title=JP4

Because not all film production goes to the cinema / tv.

We make films usually distributed to around 100-200 people. We record, edit, encode, duplicate and print all the titles with our own equipment.

We also run a film festival once a year attended by about 400 people, the films of which are mostly made by the same size of operation, some with the exact same equipment (literally).

The added irony of your comment is that if we had plenty of money we'd be using a Red or 35mm which don't have such restrictions.

I found this yesterday too: Open Source Hardware HD camera we're going to take a long look at that system. Our cinema is having 3d projectors installed in one of the screens soon, it would be great to produce stuff for it locally.

http://www3.elphel.com/

Elphel, Inc. was started in 2001 to provide high performance cameras based on free software and hardware designs. Freedom of the users of Elphel products is our top priority - we value and protect it with the GNU General Public License that covers all the Elphel software and hardware designs.

This freedom extends from the convenience of the out-of-the-box usage of the cameras with the intuitive GUI to the possibility to modify any parts of them. It protects user right to create and distribute derivative products based on our designs, products that may be suitable for the applications we had never thought about ourselves.

This thread is not complete with mentioning the Elphel. Some instructions. If this roadmap entry is up to date, you'll have to choose a specific elphel model. I have no experience with either.

This thread is not complete with mentioning the Elphel. Some instructions. If this roadmap entry is up to date, you'll have to choose a specific elphel model. I have no experience with either.

This thread is not complete with mentioning the Elphel. Some instructions. If this roadmap entry is up to date, you'll have to choose a specific elphel model. I have no experience with either.

Elphel cameras aren't much less expensive that $1000... but they are 5 mpix color network attached cameras with fully open source/free software hardware and software.. They are really very good.

You are right - you can not copyright schematics - redraw it preserving the netlist and it will not be covered by the original copyright.

But we are comfortable with the GPL for our products (http://www.fsf.org/resources/hw/cameras) - hardware board is more dead than alive if you take away the FPGA code (GPL-ed) and the software (GPL-ed, based on GNU/Linux). Our hardware evolved in parallel with the code that was ported/modified to run on the newer boards.

So theoretically - yes, it is possible to make a copy of the hardware itself not being bound by the GPL terms, but the result will not be really useful. Building a non-free derivative you will not be able to use our FPGA (http://wiki.elphel.com/index.php?title=FAQ) code - it is all compiled at once (you may say "statically linked") so it is not possible to add proprietary module to the existent code.

And working with the free hardware for more than 6 years I can confirm that in this area demand is much higher than the offer, that keeps us in business.

I have made a similar argument, I want a FPGA on the HT bus. A Fully Programmable Gate Array would allow software developers to create hardware "programs" to accelerate their products. For example, Photoshop and Premiere could re-program the FPGA to accelerate processor intensive tasks. Codecs could be implemented on the chip for realtime encoding (for example http://www3.elphel.com/en/products is a camera that does OGG Theora in a FFPGA in realtime at 640x480x90fps). And game writers could optimize any part of their applications they wish. Sure FPGA's are not as fast as a non-programmable processor, but they are infinately more flexable, and if it's sitting on the HT bus with direct access to system RAM, you would see some amazing uses. Accelerated encryption/decryption, email filtering and virus detection, or any other process that is sequential in nature and done repeatedly.

These guys have a Linux powered network camera that uses Ogg Theora. The stream is a good bit thinner than MJPEG, has a nice high resolution and the entire thing is customisable. Bit pricy maybe, and not sure what it's night time performance is like. Might be worth a look!

http://linuxdevices.com/articles/AT3888835064.html

The cameras, w/lenses and add-ons will cost more than your Mac Mini, but these are capable of 1280x1024@30fps w/Ogg Theora encoding.

http://www.elphel.com/

-Charles

You should take a look at the Elphel 333 fpga security cameras. They can do real-time encoding in the free Theora video format at HD resolutions, and provide the stream over ethernet.

The cameras don't have sound, so you'd have to use the mac mini to handle the audio, and the image quality isn't as good as one of the "prosumer" HDV cameras. On the other hand, by doing the compression in hardware you don't have any resource problems like you would transcoding an HDV or component HD feed, and can concentrate on just decoding the stream. :)

Best, you'll be supporting free multimedia instead of the MPEG patent holders.

There's an article describing the camera if you want more details.

http://www.cordin.com/product%20sheets/510_550.htm
I used to work for this company. They use rotating mirror and sensors (or film) statically mounted around. As somebody already mentioned light is very precious resource for those speeds, and cameras are ususaly not designed to work with this small exposure time. Friend of mine developed digital network sensor that can capture ultra fast events (10 ns exposure time, best if used with intensifier, of course). It's actually has been featured on slashdot before (linux inside, sure).