Welcome to our community

Be a part of something great, join today!

  • 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 :)

Why is the Helium sensor easier to manufacture than the Dragon sensor?

If you're Intel you can, because you have the volume + margin to be able to afford it. Red would probably have to raise its prices quite a bit to pull that off, especially with Red's volume. I got the impression from one of Jarred's posts that the Raven is barely paying for itself, which is why Red isn't able to bring on another production line just for Raven.

Fair enough. Makes sense. I absolutely get the impression Raven is a loss leader that happens to get some people into the ecosystem.
 
It would not surprise me if the Dragon design has various tolerances - e.g. Maximum time to read a horizontal line from the sensor.

With sensor data having to travel 33% further (VV vs 6K) - I would speculate that there is therefore a secondary issue: Ie that VV sensors need to be more perfect than 6K sensors to meet the same operational tolerances.

Chip Design : All chips have imperfect placement of their layers (masks). Variations in the lining up of masks causes variances in Inductance (L) and Capacitance (C) in the transistors, and circuit connections. It is these deviations from ideal LC that cause signal ringing, signal delay, increase in current drive and parasitic cross talk.

AJ
 
With sensor data having to travel 33% further (VV vs 6K) - I would speculate that there is therefore a secondary issue: Ie that VV sensors need to be more perfect than 6K sensors to meet the same operational tolerances.

The bigger sensor is bigger. The probability that something will fail is basically constant for a given die area. Ergo, a bigger sensor has a higher probability of defects, which makes yields lower.
 
I'd like to know if VV sized helium sensors are a possibility and would be easier to manufacture than Dragon.

Possibly, but they'll definitely run into the same size-related issues, since it would still be a huge chip.
 
Would be cool if the helium sensors are stackable like the amira sensor. Imagine if red built a tripple helium, just like the Alexa XL that has three amira sensors stacked next to each other vertically. For red that would give a 12x8k sensor.
 
My laptop after trying to play back those files

202350-laptop-fire_original.jpg
 
Would be cool if the helium sensors are stackable like the amira sensor. Imagine if red built a tripple helium, just like the Alexa XL that has three amira sensors stacked next to each other vertically. For red that would give a 12x8k sensor.



Arri doesn't stack sensors. :crazy: Stacking sensors is extremely impractical for cine camera use.

Canon once made a 201x201mm CMOS sensor out of a single wafer.

Canon+UHSCMOS.JPG
 
That is now owned by DJI as they bought Hassleblad. Not a bad purchase for their part I think

Yeah, probably a very good purchase for DJI. Probably also for Hasselblad, given the financial woes Hasselblad's been facing lately.
 
So far, I'd say that Canon is the least likely candidate. Technology wise, Fujifilm, Sony, Fairchild, and Dalsa all seem like more likely candidates, and I'm sure there are a few other companies that make sensors that do custom work as well.

Fujifilm is an add duck in imaging; Fujinon lenses are amazing (Hasselblad 645 lenses, for example), and has developed some very interesting sensor designs over the years including one that went into a Nikon dSLR clone, and now the lovely sensors for its mirrorless system camera line... yet still sourced a Sony sensor for its medium format camera.

Ummm didn't Canon release an 8K sensor during NAB....
 
Ummm didn't Canon release an 8K sensor during NAB....

Release? Nope. Just a demo. It was clearly a proof of concept demo; it was that rig with 4 Odysseys connected to it... hardly a practical setup.
 
Release? Nope. Just a demo. It was clearly a proof of concept demo; it was that rig with 4 Odysseys connected to it... hardly a practical setup.

You're completely missing my point. Canon demonstrated a functioning prototype sensor that has vanished (there hasn't been a peep about it since NAB in Spring 2016). Then out of nowhere RED releases a boatload of new Helium sensors months after NAB. In the meantime they're still having trouble with their Dragon sensors.

BTW in my original comment I'm specifically referencing Canon's ability to manufacture 8K sensors since a few people downplayed the likelihood that they're the one's making the Helium.

Also this was the prototype, I have no clue what you're talking about it being impractical...
canon_cinema_eos_8k_camera_2.jpg
 
Last edited:
You're completely missing my point. Canon demonstrated a functioning prototype sensor that has vanished (there hasn't been a peep about it since NAB in Spring 2016). Then out of nowhere RED releases a boatload of new Helium sensors months after NAB. In the meantime they're still having trouble with their Dragon sensors.

No, "release" and "demo aren't at all the same thing.

BTW in my original comment I'm specifically referencing Canon's ability to manufacture 8K sensors since a few people downplayed the likelihood that they're the one's making the Helium.

Well, that's also completely unclear. There's no indication that Canon IS able to manufacture 8K sensors in anything vaguely resembling volume.

Also this was the prototype, I have no clue what you're talking about it being impractical...

I suppose as long as you don't have any interest in recording anything from it, sure. That was what the 4 Odysseys were for.

I'm not going to pretend that it's not POSSIBLE hat Canon is the supplier that Red is using, but I don't think that it's likely. For one thing, Canon hasn't ever produced a sensor for a competitor that I've ever found an indication of, let alone a custom job for another company.

Plus unless that 8K was in volume production last March, it wasn't ready to go in time for Red's launch. On top of that, I have yet to see any evidence of a Canon sensor that could match the Helium and Dragon dynamic range.

My guess is that Red worked with a semico that specializes in scientific imaging solutions like Dalsa or Fairchild Imaging and worked up a custom sensor design in collaboration with them. But that's a guess, and odds are no one who doesn't work for Red will ever know for sure.
 
No, "release" and "demo aren't at all the same thing.

I'm aware of that, however, when a company as large as Canon creates a working demo they are pretty capable of launching it as a product if the cost and demand is practical.


Well, that's also completely unclear. There's no indication that Canon IS able to manufacture 8K sensors in anything vaguely resembling volume.

Technically theres no indication that any of the sensor brands you mentioned can manufacture an 8K sensor. On the other hand Canon demonstrated a working prototype while all the other brands have shown NOTHING.


I suppose as long as you don't have any interest in recording anything from it, sure. That was what the 4 Odysseys were for.

I'm not going to pretend that it's not POSSIBLE hat Canon is the supplier that Red is using, but I don't think that it's likely. For one thing, Canon hasn't ever produced a sensor for a competitor that I've ever found an indication of, let alone a custom job for another company.

Plus unless that 8K was in volume production last March, it wasn't ready to go in time for Red's launch. On top of that, I have yet to see any evidence of a Canon sensor that could match the Helium and Dragon dynamic range.

My guess is that Red worked with a semico that specializes in scientific imaging solutions like Dalsa or Fairchild Imaging and worked up a custom sensor design in collaboration with them. But that's a guess, and odds are no one who doesn't work for Red will ever know for sure.

I never mentioned that Canon manufactured Dragon sensors. Also I highly doubt Canon or anyone else would be running around telling people that they've manufactured sensors for a competing brand. Everyone is likely under a strict NDA.

*BTW Canon is entering the 3rd party sensor business within the next 2 yrs. https://petapixel.com/2016/09/01/canon-sell-cmos-sensors-companies-first-time/
 
Last edited:
Technically theres no indication that any of the sensor brands you mentioned can manufacture an 8K sensor. On the other hand Canon demonstrated a working prototype while all the other brands have shown NOTHING.

You tease an interesting point. Why would a competitor like Canon sell RED their 8K sensor? Unless of course (thinking beyond Canon), RED's sensor partner is no longer...in the camera business...
 
I thought 5D was an 8k+ sensor / more pixels than helium, is it not?
Sure it´s not recording the amount of frames but it still kind of proves that canon has a source for quite capable sensors, no?
 
Back
Top