Photosite density and sensor sizes

[Christoffer Glans]

I've been watching the Steve Yedlin tests on showing how resolution of cameras rarely matters for perceivable resolution and while that could be discussed back and forth, the thing that caught my eye the most was how clean, sharp and smooth the Alexa 65 looked when doing a true comparison with the Dragon 6K sensor. I've been holding off getting Helium since I'm not sure it's an economical investment for me in terms of what I get from the sensor vs the upgrade price, but one more thing has been the pixel density and smaller photosites of having an 8K sensor at the same size as the Dragon 6K. I've been much more interested in the 8K VV sensor since it's bigger, with the same photosite sizes.

Now, it made me wondering if it's economical and possible to make a VV Red sensor at 6K? Making the photosites larger as well as the sensor size set at Vista Vision. If the Dragon 8K VV is so expensive and hard putting together, would a larger photosite version at 6K be easier to make?

The reason is that at 6K resolution, Alexa 65 still kills any other camera because of how large the photosites really are. Wouldn't a Red VV sensor at 6K deliver a 4K image that is closer to that clarity and sensitivity?

I know Red has always focused on cramming more resolution into the sensors, but both the Dragon 6K and the Helium 8K are still cameras aimed for 4K distribution, why not do a VV 6K sensor? I mean, to really create an 8K distribution-able sensor with all that scale down magic, you have to go way up to 11 or 12Ks, right?

It's a serious question because it might also enable the visuals of the Vista Vision sensor size to a user base that might not afford the 8K VV and who has more interest in VV than the resolution bump to 8K.

 

[Anthony Berenato Jr]

I guess one of the biggest challenges is investment. Is it better to invest in the technology that makes an 8K VV sensor easier to manufacture and cheaper to produce, or spend the money in R&D / machines / production of a 6K VV sensor that would only serve as an interim and produce for a limited time (as the costs of 8K VV come down).

Though, I think it's a cool idea - as I love the way medium format makes everything look gorgeous... I'm just a bit worried that it wouldn't be practical in terms of time and cost. That's a real challenge, and sometimes it's better to invest in the future than redirect for the short term. I think the technology will get there.

As for the test, did they show any footage comparison between 8K Dragon VV and the 65? Helium has definitely been a treat here in comparison to Dragon, and I would love to see even that comparison as well. There's a certaint smoothness I can't seem to explain across my footage as well (vs 6K Dragon). Maybe that next tick / tock will bring us all there.

 

[Christoffer Glans]

I guess one of the biggest challenges is investment. Is it better to invest in the technology that makes an 8K VV sensor easier to manufacture and cheaper to produce, or spend the money in R&D / machines / production of a 6K VV sensor that would only serve as an interim and produce for a limited time (as the costs of 8K VV come down).

I have not found it to be reducing in cost. I hope I'm wrong, but so far it seems that the production cost is so high that it's mainly going to the Panavision DXL department and the cost is also high because it takes a big chunk of the "sensor cake". However, but correct me if I'm wrong, if a sensor was made with less photosites, it would be cheaper as well as not putting a huge strain on the boards inside the cameras, meaning that both the sensor and the board upgrade cost would be less, right?

Though, I think it's a cool idea - as I love the way medium format makes everything look gorgeous... I'm just a bit worried that it wouldn't be practical in terms of time and cost. That's a real challenge, and sometimes it's better to invest in the future than redirect for the short term. I think the technology will get there.

But as I said, the Helium 8K is made for 4K distribution and it hasn't any Vista Vision counterpart. The rest of the industry is moving towards larger sensors and Vista Vision sized sensors will probably be the "next big thing" for most cameras. If Red now only focus on Helium, that would mean they need to release an 11-12K version of the Helium (my math may be off) to enable a Vista Vision version of that sensor. Since most productions today still struggle to get 4K distribution, we are still focusing on getting good looking 4K over getting 8K distribution, also countering in the requirements for 8K VFX and so on. So in the end, a 6K Vista Vision sensor would probably be the most popular sensor made (if the cost is less than upgrading to 8K Dragon VV). I don't think that a VV Helium 12K would be anything but popular for resolution niche entusiasts.

As for the test, did they show any footage comparison between 8K Dragon VV and the 65? Helium has definitely been a treat here in comparison to Dragon, and I would love to see even that comparison as well. There's a certaint smoothness I can't seem to explain across my footage as well (vs 6K Dragon). Maybe that next tick / tock will bring us all there.

Nope, but my point is that a 6K VV sensor might cost less to produce. If the 8K VV would lower in price and be much more available then I would surely go for that instead. The comparison you can see is the shot made with the 6K Dragon and when you see the Alexa 65 of the same subject it's such an extreme difference in clarity that no other camera is even close. If that's not because of the large photosites and it's sensor size, then someone needs to fill in what does that. I'm just saying that I feel that a new 6K Red sensor in Vista Vision would create a much better image quality than I think the Helium can produce. But would love for Red to chime in on this.