Welcome to our community

Be a part of something great, join today!

Register Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Dream Super - Vistavision Chip Concept : "Hydrogen"

D

Dan Kanes

Well-known member
Joined
Nov 25, 2009
Messages
5,496
Reaction score
12
Points
38
I call it Hydrogen because it would be "LIT"

3 helium wafers stacked top to bottom and rotated 90 degrees. ~ 47.31 x29.90mm total silicon area , TBD how much would be utilized fully.

Pixel density is combined by a DSP multiplexer so that 3 or more pixels become 1 "super pixel" which doesn't need additional interpretation for the processing back end past the Front end DSP.

DSP multiplexer stitching reduces rolling shutter artifacts through its interconnect technique.

Pipeline then works normally after that so we have 8K (and smaller) recording from the open gate sensor.

Obviously this is just a daydream for the people like me who have a vistavision deposit in from NAB 2015.

I'm not "tripping" on my deposit, I kept it in because I believe the dividends will pay off way better than any other investment of that 10k.

Red always takes care of us.
 
Haha you can trade mark an element? Only in America. Mendeleyev is turning in his grave....
 
Jacek Zakowicz said:
Haha you can trade mark an element? Only in America. Mendeleyev is turning in his grave....
Click to expand...

Heh, I was surprised when they trademarked a color... after that, trademarking an element is not surprising at all. '-)
 
yes I know- it never seizes to amaze me what the government agencies make money on...
The sensor question is not if it can be done- but why? I can only imagine the complexity of triple processing sync, color matching, etc, etc, etc....
 
Andy Roberts said:
Hmmm. I guess I don't feel that jaded about the USPTO. It is an essential service to keep things orderly where otherwise chaos would abound.
Click to expand...
Don't want to jack the thread but patents are irrelevant until defended in the court-and the company with the bigger whip always wins. Trademarks are more relevant- but Hydrogen? really?
 
Sensor stacking doesnt happen in cinema or stills sensors. Lol.
Not sure where the whole idea came from. Wouldn't we have 4x5 large format CMOS by now if it really was an efficient way to increase sensor size?
 
Last edited:
Dan Kanes said:
It does happen.

This is precisely how the alexa65 sensor works.
Click to expand...

According to Arri it doesn't. It's a single cut from a cmos wafer.
Besides, putting all 3 sensors in perfect allignment would be a pain in the ass. we're talking about microns here. Not to mention keeping them in place after heavy use. Can you imagine them shifting out of allignment after a bump? Screen tearing...
 
Last edited:
Milano Questa said:
According to Arri it doesn't. It's a single cut from a cmos wafer.
Besides, putting all 3 sensors in perfect allignment would be a pain in the ass. we're talking about microns here. Not to mention keeping them in place after heavy use. Can you imagine them shifting out of allignment after a bump? Screen tearing...
Click to expand...

By stack I mean during the fabrication process, in terms of conceptual planning, not after the fabrication process in a physical Tetris manipulation of existing, non matching dies frankensteined together with super glue.

I mean, what do you think a vistavision dragon sensor is ? If you want to break it down to colloquialism, it's a larger cut of Dragon wafer.

They still had to do a custom masking layout, which is the same thing Arri did for A65, it's about taking existing IP and laying it onto a silicon substrate in a different way via an opto/photo lithographic process.

I assumed people reading this already had some rudimentary understanding of electronics manufacturing and could survive reading, virtualizing these concepts and ingesting them mentally without deciding to argue a semantic point.

Nothing I'm talking about doesn't exist or is science fiction, so I'm politely going to ask you please take a chill pill.

I'm guessing it might be a language barrier or that you just have an argumentative feeling you're looking to explore today.

Let me put it in different wording for you.

Imagine if Red took their Helium sensor IP and laid it out in a manner that would make it effectively 3 times larger, and decided to do a triplicated readout from different CMOS lanes, using an additional FPGA fabric to handle readout and combination only without moving into a "DSMC3" level of processing architecture and keeping effective recorded 8K size, with intelligently combined "super pixel" light gathering - RGB or RGBW "light gathering buckets" made up of 4 traditional photosites.

In any case - have fun thinking about it, it's not a "real thing" yet for you to even argue with me about. When it exists or doesn't exist, get back to us about how much you know about electronics.

Feel free to showcase your argument about how I'm wrong.
 
Dan Kanes said:
By stack I mean during the fabrication process, in terms of conceptual planning, not after the fabrication process in a physical Tetris manipulation of existing, non matching dies frankensteined together with super glue.

I mean, what do you think a vistavision dragon sensor is ? If you want to break it down to colloquialism, it's a larger cut of Dragon wafer.

They still had to do a custom masking layout, which is the same thing Arri did for A65, it's about taking existing IP and laying it onto a silicon substrate in a different way via an opto/photo lithographic process.

I assumed people reading this already had some rudimentary understanding of electronics manufacturing and could survive reading, virtualizing these concepts and ingesting them mentally without deciding to argue a semantic point.

Nothing I'm talking about doesn't exist or is science fiction, so I'm politely going to ask you please take a chill pill.

I'm guessing it might be a language barrier or that you just have an argumentative feeling you're looking to explore today.

Let me put it in different wording for you.

Imagine if Red took their Helium sensor IP and laid it out in a manner that would make it effectively 3 times larger, and decided to do a triplicated readout from different CMOS lanes, using an additional FPGA fabric to handle readout and combination only without moving into a "DSMC3" level of processing architecture and keeping effective recorded 8K size, with intelligently combined "super pixel" light gathering - RGB or RGBW "light gathering buckets" made up of 4 traditional photosites.

In any case - have fun thinking about it, it's not a "real thing" yet for you to even argue with me about. When it exists or doesn't exist, get back to us about how much you know about electronics.

Feel free to showcase your argument about how I'm wrong.
Click to expand...

From Paralinx wireless to now plotting on how to develop a super sensor, you're an impressive dude Dan. =)
 
Haha thanks Alexander.

By the way, this posting is just for fun.

I'm daydreaming about the future.

In no way would I ever try to take credit for the hard work of others actually doing the real work making great cameras.

Looking forward to seeing you all at Cinegear!
 
I wouldn't be surprised if Hydrogen was next.
At this point I would be happy with S16 Helium cut. Or three S16 RGB Heliums and no debayer.

In Vista Vision Helium tech with Dragon pixel pitch would be a great combo.
 
You must log in or register to reply here.
Back
Top