RED alternative, a true indie cam possible?

[tonyp12]

guess I can tell you what I have in mind as I have applied for a patent
and any copies of this idea would just finically benefit me anyway.


Fact:
Memory (both ram and harddrives) have gotten cheaper and bigger the last years.

Only reasonable thing to do with digital video a year ago was to compress it,
though adding hardware cost and editing problems.

Now days you could store 12bit (4:4:4) RAW bayer patter cmos data directly
to relative cheap ddr2 ram.

What if you created a pci-e 16x card that have 16 slots for ddr2
and one 2k high-end cmos sensor.

4GB x16 slots = 64GB

1080p Calculation:

2 073 600 pixels at 12bits each = 2.97mb
* 60 frames = 178.2mb/second

That shows that you could have 6 minutes of continues filming.

The next step from here is to move it to a more permanent storage:

A internal 1 terra byte 3.5" HD,
that dumps the 64gb ddr2 ram between film takes.
or a docking station that have eight 1 tb drives in raid-5 configuration.



So you would need some type of board that have hardrive ports, usb/firewire ports, hdmi, cpu and operating system.
That have already been created and is called a computer motherboard.

http://www.via.com.tw/en/products/mainboards/motherboards.jsp?motherboard_id=550

This 17cm x 17cm board have everything a digital camera could need.
and the very important pci-e 16x slot for the cmos module.
It would cost a fortune to trying to create something like that from scratch.

End-user can replace mobo or cmos module as technology improves.
With an open source Linux operating system and open source camera software,
this would allow anyone to tinker with it and make changes.
A true hobbyist camera.


The problem is to locate a source for the cmos.
4K cmos are in the works but will only be available to the elite with an initial $250k membership fee.

This 2K one could be a good start.
http://www.altasens.com/ProductFlyer...oductFlyer.pdf

What cmos would you like to see be used?
What would you want the camera to have?

 

[Cüneyt Kaya]

scarlet is 5k-8k USD (i guess) and can do 2k RAW....red is revolutionery because of their price tag....to get their cams to the hands of many, what would be your price tag and could it be used in handhd systems?

 

[tonyp12]

We have a target price tag of around $5000-$7000, not including 35mm lenses.

That will include:
Motherboard with 2gb of ram, cmos module with 64GB of RAM, 1tb Harddrive, aluminum camera body, battery, 7" LCD screen and 5.1 microphone.

It will be handheld, but probably 50% bigger than a RED.

If it was not for the $250k initial fee we probably would go with DALSA 4M70 (Meritage) Global Shutter CMOS Sensor
Resolution: 2400 x 1800, Pixel Size: 6um x 6um

At $700/each not bad.

 

[Cüneyt Kaya]

sounds cool----

but what about support...
red has the best customer support i ever saw, hard to beat.

 

[chuckt]

The problem is to locate a source for the cmos.
4K cmos are in the works but will only be available to the elite with an initial $250k membership fee.

This 2K one could be a good start.
http://www.altasens.com/ProductFlyer...oductFlyer.pdf

What cmos would you like to see be used?
What would you want the camera to have?

First, your link does not work.
Second. Very good idea,
Third: The $250K membership fee can be avoided.
If everything else is ready, I might be able to get you the 4K sensor without the $250K.

 

[Dylan Reeve]

For that sort of application, I think you'd find a custom board to be much better than a stock ITX solution.

Also, something like an SSD which already offers a SATA interface is a lot easier to work with - I believe you might find problems addressing 16x 4GB modules.

Also, there's no ready-made interfacing option for a CMOS imaging chip, so whatever you do there is going to have to be a custom hardware solution.

It's surely possible, but I think that unless you're pretty adept with electrical engineering, low-level programming and imaging, it's going to be a pretty daunting process.

 

[Brent J. Craig]

Nice try (troll) but you'll never compete with Red. Combine genuine passion with a couple of billion in backing and you have an unbeatable combination. Add to that a carefully groomed, fanatical community and you have a revolution. (Although specs may change at any moment.)

 

[Jason Ing]

What is so unique about your IDEA that Red or Scarlet doesn't already beat now? Or in the future as they both get better and better? (look at the phenomenal progress of Red even in the last few months!!!)

It has to fill a gap or be equal or better then existing products????

 

[Jeff Coatney]

What follows is not legal advice, as I AM NOT AN ATTORNEY. But I have been in the IP and Product Development world long enough to have had my Ass unceremoniously handed to me on occasion by the Market. I pass along these thoughts as a comrade-in-arms...

guess I can tell you what I have in mind as I have applied for a patent
and any copies of this idea would just finically benefit me anyway.

The United States Patent & Trademark Office provides for a provisional Patent that issues some degree of protection for up to one year. At the end of that period, you may file an actual Patent and pay the required fees to do so. That in and of itself does not provide you with any protection under the laws of the United States* unless the patent is actually awarded to you by the PTO (Patent & Trademark Office). While your claim is being evaluated for Patent, i.e. "Patent Pending", the particulars of your idea are not accessible by the Public at large, which buys you some time to get your product to market before the PTO issues it's ruling. If in fact the PTO awards you a Patent for your product, you win only the right to legal protection against infringement. In order to win damages from a Patent Infringement lawsuit, you must first make sure the infringer is subject to the laws of the United States, you must prove the infringer violated your patent --as the burden of proof is on you, and finally, you must initially fund the lawsuit, which you may not win. Patent infringement is tried in the Civil courts typically and is not under the normal jurisdiction of the Criminal courts. Even if you are awarded damages, it will prove difficult to collect those damages as someone who's nefarious enough to infringe on your idea will no doubt be clever enough to hide their ill-gotten proceeds in such a way as to make accounting extremely difficult.

*The crux of your position is that you have asserted Patent protection in the United States, however, this forum is accessible to anyone with an Internet connection anywhere in the world. Patent protection only extends to the country in which it is filed, unless by treaty or other accord another country grants patent rights to you based on your filing in the US. At best you would have the right to legal protection in the country where the infringement occurred, but you would still be at the mercy of their laws and you would have to fund the lawsuit in that country.

Fact:
Memory (both ram and harddrives) have gotten cheaper and bigger the last years.

Only reasonable thing to do with digital video a year ago was to compress it,
though adding hardware cost and editing problems.

Now days you could store 12bit (4:4:4) RAW bayer patter cmos data directly
to relative cheap ddr2 ram.

What if you created a pci-e 16x card that have 16 slots for ddr2
and one 2k high-end cmos sensor.

4GB x16 slots = 64GB

1080p Calculation:

2 073 600 pixels at 12bits each = 2.97mb
* 60 frames = 178.2mb/second

That shows that you could have 6 minutes of continues filming.

The next step from here is to move it to a more permanent storage:

A internal 1 terra byte 3.5" HD,
that dumps the 64gb ddr2 ram between film takes.
or a docking station that have eight 1 tb drives in raid-5 configuration.

Look closer at the Phantom HD. I think you'll find they introduced into the market a similar method of capturing data. That will be provable if they have any Patents that come close and could derail your claim.

So you would need some type of board that have hardrive ports, usb/firewire ports, hdmi, cpu and operating system.
That have already been created and is called a computer motherboard.

http://www.via.com.tw/en/products/mainboards/motherboards.jsp?motherboard_id=550

This 17cm x 17cm board have everything a digital camera could need.
and the very important pci-e 16x slot for the cmos module.
It would cost a fortune to trying to create something like that from scratch.

End-user can replace mobo or cmos module as technology improves.
With an open source Linux operating system and open source camera software,
this would allow anyone to tinker with it and make changes.
A true hobbyist camera.

Since you allow the hobbyist some freedom with your product you are not, by definition, providing a solution to a problem or opportunity in the market. Based on my experience, the Patent Office favors those claims that provide a solution to an existing shortcoming, problem, or opportunity or creates a new business or industry where previously there was none. The problem you identify as trying to solve with your idea; data throughput and storage, utilizes largely off-the-shelf solutions that may already have some degree of Patent protection or prior art.

The problem is to locate a source for the cmos.
4K cmos are in the works but will only be available to the elite with an initial $250k membership fee.

This 2K one could be a good start.
http://www.altasens.com/ProductFlyer...oductFlyer.pdf

What cmos would you like to see be used?
What would you want the camera to have?

I advise that you build this thing of yours and sell it, bypassing the Patent office altogether and get the protection you crave for the idea from the revenue you generate in the market. Technology Patents have an extremely truncated relevance, especially a design Patent, which is what yours looks like. Build that great idea, saturate the market with it before the competition has the opportunity to oppose you and capture the market share. Speed and stealth is where the competitive edge lies and the money is the only reward and protection you can count on. Entrepreneurs Act, Government Reacts. Put your faith in yourself and show people your idea, don't tell them. Does that sound like a product we're all familiar with?

Good Luck!

 

[Dylan Reeve]

A patent is only a good as the money you can expend on protecting it too. If someone with deeper pockets than you wants to rip off your idea, there's a good chance they'll get away with it.

Patents, sadly, have limited benefit to small businesses and individuals, and even less so in a globalised world.

 

[tonyp12]

The 16 x 4GB address issue will be avoided by only having each 4Gb module accessible at at time, no need to have the whole 64GB randomly accessible.
SSD or any other type memory solutions is not fast enough when it comes to RAW 4k images or extreme high speed (for slow mo)


We have a USA patent pending on the cmos module only.
Though we will build complete camera systems,
the cmos module will be available for sale by itself.

Plug it in to anything that have a pci-e slot, you are allowed to do anthing you want with it.


Initially we are not gone spend time and money having it patent in every country in the world.
Hopefully by the time we start actually making money some Chinese company have not already flooded the market.
At least they will have problem selling in the USA.

 

[GlennChan]

Based on my experience, the Patent Office favors those claims that provide a solution to an existing shortcoming, problem, or opportunity or creates a new business or industry where previously there was none.

I've seen some pretty silly patents... like method for exercising a cat.
http://www.freepatentsonline.com/5443036.html

 

[GlennChan]

Back to the original topic... I think there have been many similar projects in the past.

SI-2K
Andromeda-modded DVX100
Drake (didn't turn into commercial product AFAIK)
Kinetta (not in active development anymore?)
Nox (doesn't really count / not in same price range)

If you look at the SI-2K... originally they were going to go uncompressed, but then they realized that it made a lot more sense to go with very gentle compression (gentle in terms of image degradation... somewhat similar goals as to what Redcode does). Both the SI-2K and the Andromeda DVX100 are sensors attached to computers... so in that sense it looks very similar to what you're doing.

2- In general terms, I'm not sure if your solution would be a good balance of image quality, price, and practicality (e.g. size, weight, ease of putting the whole package together, camera features like timecode, lens mounts, etc. etc.).

There are many good camera systems on the market already that would be suitable for low-budget productions...
Red
SI2K
EX1
HV20
Any of the HDV cameras but recording the HD-SDI signal coming off it
etc. etc. etc.

2b- On the other hand, apparently there are some really high dynamic range sensors around (20 stops claimed). The image quality probably isn't that great but people will buy things like Holga cameras if they deliver an interesting image (then again, people wouldn't pay that much for a Holga).

http://www.dvinfo.net/conf/showthread.php?t=59314