AXIOM Alpha Prototype Hardware complete
Submitted by Sebastian on Mon, 10/07/2013 - 11:14
After receiving our PCBs for the prototype, we have now started assembling them. Doing so has required a series of complex procedures and multiple runs of hand soldering tiny components or reflow soldering on a hot plate. This has been immediately followed with connectivity testing and interface communication tests. At present, all components have been properly installed and testing/verifying electrical connectivity has indicated that everything is in good working order. On the software side of things, we have basic communication with the sensor on the Zedboard up and running and can set/get registers on the image sensor through the Zedboard's Linux shell. The next step involves getting actual image data from the sensor, however before we can do this we have to create a memory interface for reading/writing to and from DRAM, storing and retrieving the image data from inside the FPGA.
HDMI Output
To get the HDMI output up and running on the Zedboard, we have created a test pattern generator inside the FPGA. If you have a Zedboard you can give it a try here, irrespective of whether you have the image sensor hardware. Doing so will allow us to verify and adjust timing properties across a range of monitors and resolution/refresh rate combinations where necessary. The main reason for these tests is that outputting video over HDMI from the image sensor is much easier if you know that the HDMI output is already working properly across all monitors and resolutions/refresh rates, and that the only thing we really need to deal with is the actual image stream that is being output.
and the AXIOM project is certified Open Hardware by
.
8 Comments
slow mo
can the apertus camera evolve into a great slow mo camera for sports ?
+1 for sports camera
love the high frame rates the sensor offers... being modular in design we could develop modules that would suit sports...
Slow Mo - Let's hope so.
I have a footage library of 16mm high speed skydiving that does quite well. I also have jumped the Red Epic on a belly mount. I'm hesitant to put one on my head, as the lens mount is still higher off the base than I'd like. I'd do it, but I'd rather have a super low profile high speed capture device on my head. No frills, high speed acquisition is what's needed with a helmet cam.
$10K and over 100 fps at 4K... I'm in... for two!
More Progress Towards the Goal
Nicely done to the team and the fab shop. Looking forward to the next milestone!
Visionary
This is exciting. As a Linux user and amateur web programmer and filmmaker I appreciate the potential for freedom that the Axiom project offers. I would like to support this initiative.
Fantastic!
I honestly wish I was working with you guys on this. Nice work!
Most Fascinating!
I believe there is a good prospect for success with this idea. So much of the technology business depends on
smaller, more cost effective factories in Asia, particularly China. With each component of this system, somebody
somewhere can make some profit on it, probably not exorbitant. We've seen that Windows and PCs work with
a huge variety of manufacturers. We've seen Linux as an open source software succeed at certain tasks.
The HDSLR concept has interchangeable lenses, that is a first step towards this. Now we will have interchangeable
parts, from a variety of manufacturers. Those manufacturers with both reasonable quality control and pricing should
succeed, from word of mouth among users, spread through the web. The miniaturization of various parts and modules
makes this do-able. I can envision even smaller scale cameras using this concept. Little lenses, generic batteries, maybe
even combinations of modules made by the same manufacturer. You could have a module with the "floating lense" like
SONY has for good image stabilization. You could even have an outdoors, rainproof batch of components, an underwater
casing for going underwater, freeze resistant, shock resistant components.
When such a camera, or batch of modules and a connecting "rack" becomes available I'll seriously consider buying. I'll watch this with great interest. You are to be commended.
Open source alternative for
Open source alternative for all the vendor lock-in products (arri, red and so on) is desperately needed! I hope you will succeed! it's looking promising so far. could the developers please comment on here: when do you expect the first prototype to be complete (capable of capturing, displaying and storing video in real time, with still some minor bugs to work out)? can you say at all how the timeline is looking so far?
Add new comment