Thu Oct 31, 2019 2:41 pm
Ok. To give you a little idea of where I'm coming from, I am looking for a phone watch with an 3840+ wide sensor. At the moment I have only found up to 8mp versions (though I thought I saw a 12/13 mp verson, but can't see that. That may give reasonable 2k.
-
Now the brain disease is clearing up (on and off a bit too) I am thinking of.all these things to try. But having wisdom, I know I don't want to over reach and can't effectively do much on my own anymore. I also have a big project coming on steam shortly that will require my attention.
-
So what can be done. Welcome into my little secret war stratergy room, complete with red like skulls with a throwing dagger through it (because I don't like skulls iconology being used. I'm more an Unreal player rather than a Doom/Quake player myself).
-
I claim exclusivity to IP not already existing or at least able to be protected, prior to my coming up with it (this is to have a basis to overturn any external claims on things I come up with, and then interfering. Not too uncommon. This establishes an incident of prior art, but not necessarily an original incident in each part):
-
One, is to crowdfund the development of a seperate code base of services for different camera software, including open source camera software. This is seperate the code base from the opensource codebase and it's license. The plugins are general services the other software can use. I came up with this stratergy for my own open OS licensing proposal, to enable commercial entities to have produce some profit to fund business progression and focus, through some exclusivity. It is pretty much like plugins and code for the old Netscape browser, are not in anyway part of the code, but just services that the browser could call. As part of this effort, an upgrade of the open camera software code to handle these services and definitions for compression, image and sound transformations, IO and user interfaces. The software would read an external file or the users can enter things directly. In this, the software does not necessarily have any codecs of transformation or user interface format, the user can load in what they want, and use any competitors stuff. So, a competitor could package the camera structure service with it's compression, transformation and UI services to make a seperate commercial product. However, some of these enhancements would be exclusive lock mechanisms that only allow certain product (anybody could make their own alternstive code version to get around that. So, it is basically a specific end product version exclusivity for a specific brands product. I had a whole working out how such a licensing system would work to foster commercial development of open source products, with retriculating features back into the common pool.
I have tried to get open camera software people to implement definable features to allow others to easily plug in new ui, codecs, and improve filesystem performance, or allow a plugin service, but to no aveil, too much like not easy Street.
-
Development of these UI, transforms, compression, and recording services and data descriptions. All this keeps away from other's patents. Only patent owned, licensed or patent free technologies ship with the packages. These are safe margins the court should support.
-
Development of cineform open source wavelet codec for phones and x86 OS's.
-
Development of cdng into a metadata format with variable compression in a container with old and new codecs,(maybe cineform).
-
Extensive deterministic realtime energy/performance coding for cpu you and fpga etc to maximum performance through native coding.
-
JavaScript web code version of open camera software, where the performance critical sections are native code to maintain optimum performance. So basically, all the bits that should not really matter to high performance demands are written in this transportable coding format. An implementation for a new architecture would use this codec unless native code was available. The performance sections would also have stand in default transportable code (GPU and fpga etc too) until somebody writes the proper native code. This means, a product on a new architecture, can quickly get up and running without new coding, by transferring a JavaScript engine, and running the code.
-
Variable distributed processing functionality. In this, the encoding of compression can be distributed between all available computational and other resources in parallel (a scheme can be defined on how to most efficiently to do that for the compression and or content used). This can be throttled back for whatever reason, such as: heat, power drain, other functionality on system or to make room for other functionality, or otherwise as desired. In this way, as an example, a frame or group of pictures, maybe done on the internal photo or video hardware encoder, while the next frame or gop is done on the CPU and the one after that on a GPU, and another on a dsp etc etc. But, it is likely that each computational resource could handle a different amount of frames compared to each other, so the amounts of each can vary, and a que can.be used. But, they might work on different parts of a gop or frame, at the same time, or with each other on common parts, or any other combination. In this way, the user can obtain maximum performance for their device, allowing higher frame rates, resolutions and characteristics.
-
I have been looking at an alternative way to encode Bayer data into a way which is not Bayer.
-
As can be seen above, each step is a progressive step that can be done one by one (except the services, cineform or new XDNG etc can be done as their own seperate projects or instead of the other projects). Lot's of options. A cineform codec registered into Android with filesystem streaming, gets you a long way, but somebody has to still write propper filming software to use it. Adjusting an existing open source camera app to use external services, like a codec, is a reasonable thing and allows simple system unregistered code to act as a compression codec. With these sorts of things, industry can supercharge efforts in camera development.
aIf you are not truthfully progressive, maybe you shouldn't say anything
bTruthful side topics in-line with or related to, the discussion accepted
cOften people deceive themselves so much they do not understand, even when the truth is explained to them