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 :)

Epic has changed...

So does the behavior of larger sensors make everyone want to make bigger ones or is there a mathematical or otherwise hypothetical sweet spot of sorts, considering all aspects of usability for a 4K master.
(I know capturing larger is always better, I mean other reasons)
 
I was assuming an equal field of view (equivalent, not equal, focal lengths)

Maybe Sumfun and I are pedantic. Here's how I would put it:

  • Larger sensor standing in the same spot with FOV corrected by longer lens: thinner DOF, same FOV.
  • Same lens, larger sensor moved to get the same field of view: thinner DOF.
  • Same lens, larger sensor standing in the same spot: same DOF, wider FOV.
  • Same sensor, more telephoto standing in the same spot: thinner DOF, narrower FOV.
  • Same sensor, more telephoto moved to get the same field of view: same DOF, same FOV, higher magnification of background blur (often confused for thinner DOF, see Myth busted: telephoto lenses have thinner DoF).

Really, DOF can all be summed up as aperture (not focal ratio) and focus distance[*]. To get the same FOV on a larger sensor, a longer focal length is needed. If the focal ratio is kept the same, then the aperture increases (75mm f/2 is wider than 50mm f/2). Since the aperture increases, DOF is thinner. So the mental shortcut is that sensor size contributes to DOF.

[*] Plus, the apparent depth of field will be affected by the sensor and post production. But it all starts at the lens and goes downhill from there.
 
Maybe Sumfun and I are pedantic. Here's how I would put it:

  • Larger sensor standing in the same spot with FOV corrected by longer lens: thinner DOF, same FOV.
  • Same lens, larger sensor moved to get the same field of view: thinner DOF.
  • Same lens, larger sensor standing in the same spot: same DOF, wider FOV.
  • Same sensor, more telephoto standing in the same spot: thinner DOF, narrower FOV.
  • Same sensor, more telephoto moved to get the same field of view: same DOF, same FOV, higher magnification of background blur (often confused for thinner DOF, see Myth busted: telephoto lenses have thinner DoF).

Really, DOF can all be summed up as aperture (not focal ratio) and focus distance[*]. To get the same FOV on a larger sensor, a longer focal length is needed. If the focal ratio is kept the same, then the aperture increases (75mm f/2 is wider than 50mm f/2). Since the aperture increases, DOF is thinner. So the mental shortcut is that sensor size contributes to DOF.

[*] Plus, the apparent depth of field will be affected by the sensor and post production. But it all starts at the lens and goes downhill from there.

I'll buy that...

Jim
 
So does the behavior of larger sensors make everyone want to make bigger ones or is there a mathematical or otherwise hypothetical sweet spot of sorts, considering all aspects of usability for a 4K master.
(I know capturing larger is always better, I mean other reasons)

I don't know about hypotheticals, but I can say that for me the primary concerns are lenses and sensor technology. There is a high concentration of lens quality and quantity at Super35 and Still35 sizes. First lenses:

Wider aperture lenses cost more to manufacture to the same quality level and larger sensors require wider apertures for the same field of view. For example, 50mm f/2 on the RED ONE has an aperture of 25mm. It requires approximately 80mm f/2 on still 35, which is an aperture of 40mm, 15mm wider.

One way around that problem is to use slower lenses on the larger format, as is typical with Medium Format compared to Still 35. That way the aperture and depth of field stays the same, but you have to use slower shutter speeds to get the same exposure. Another way around that problem is to not manufacture them to the same quality level.

In DSLR cameras, one of the difficulties is moving the mirror up and down fast enough, which gets exponentially harder with larger mirrors. That's part of the reason why the crop cameras can be faster than the full frame.

Sensor technology is divided unequally among sensor sizes. I'm not sure if it's market forces or engineering difficulties, but it's where we are. The very smallest digicam sensors, with over 25 times less area than Still35 sensors, still have much higher technology.

For at least the last five years they have led the industry in quantum efficiency per area, read noise at base ISO per area, read noise at high ISO per area (not always), resolution per area, and sheer marval. Of course, it doesn't really matter how good they are "per area", because their area is so tiny that even if they are 10X better than a DSLR, the DSLR is still 25X bigger, so it wins in the end. Anyway, technology is trickling up from digicams into bigger sensors, such as gapless microlenses.

Of course, there are some areas where bigger sensors have higher tech, too. All that extra space might as well be used for noise reduction circuits, which is why big sensors can equal (or sometimes beat) digicams in high ISO noise on a per-area basis.

So anyway, my point was that some sensor sizes have better technology on an area basis, so a smaller sensor with higher quantum efficiency is more desirable than a larger sensor with lower QE, unless the larger sensor also has a faster lens available.

For example, the Canon 50D has gapless microlenses and higher QE on a per area basis than the 5D mark 2. So if you crop the 5D mark 2 down to the same size as 50D, the 50D will have much higher signal-to-noise.
 
Ok, so let me see if I picked up the basic and most important aspects from your post. (Thank you BTW!)

1. Larger sensors mean exponentially more expensive lenses... especially since 35mm lenses may seem cheaper in comparison due to their availability, when you go bigger you get hit from both the manufacture cost and the low supply sides at once.
2. High ISO noise performance is possibly the only area where most companies seem to already be able to beat the smaller sensors.

That's basically it right? well till Red came along and now god knows what we are gonna get :D
 
What I don't get is why Epics old S35 sensor camera was praised? Was it a true S35mm size and not the 16:9 RedOne design, or a 4:3 standard? I could only guess that RedOne owners would feel that theirs are almost equal because of the ability to upgrade their sensor to whatever the Epics is, and even use RedOne as a deposit for a Epic. To me Epic should be another whole new beast instead of RedOne Draft 1.5!

Epic should be at least another sensor size if anything?
 
What I don't get is why Epics old S35 sensor camera was praised? Was it a true S35mm size and not the 16:9 RedOne design, or a 4:3 standard? I could only guess that RedOne owners would feel that theirs are almost equal because of the ability to upgrade their sensor to whatever the Epics is, and even use RedOne as a deposit for a Epic. To me Epic should be another whole new beast instead of RedOne Draft 1.5!

Epic should be at least another sensor size if anything?

What do you mean Epic's "old sensor"? Maybe you need to read what the differences are and why before posting this? Just a thought.

Jim
 
What do you mean Epic's "old sensor"? Maybe you need to read what the differences are and why before posting this? Just a thought.

Jim

I did not mean to sound that bad, I apologize. I was trying to talk about the sensor size. The RedOne can get a new Sensor Mysterium X(Monstro, ect.) anyway. The new sensor sounds amazing I just dont think it needs another camera unless it is a differ size.
 
I see what you and tons of other people are saying Ayoji about making each camera hold a different sensor size... it feels "right". But I am sure (and I bet you'll agree) that if Jannard makes a S35 sensor for Epic it will likely be only cause he pushed funded and demanded from his team to cram something that will blow your mind into the S35 surface area so that we also got access to the S35 glass that is much more affordable and varied. And because they reached the conclusion that there is S35 glass out there that is good enough for the insane amounts of resolution they are cramming in the surface area. Or some other reasoning that escapes my limited knowledge. And I am sure that Jim knows you didn't mean to make it sound bad and that it was instead some of that poor writing that happens when you get so exited about something you can't see until the oven's bell dings. It is the are we there yet phenomenon, Ayoji, Jim. And we are stuck in here cause it will be worth it. :w00t: :shifty: :biggrin:
 
Hello Jim!
please let us know your take on tone mapping for expanded HDR.Why not take that direction hardware wise built within EPIC.Instead of playing the game of bigger sensor.
thank you.

Alain
 
Hello Jim!
please let us know your take on tone mapping for expanded HDR.Why not take that direction hardware wise built within EPIC.Instead of playing the game of bigger sensor.
thank you.

Alain

Are you kidding me?

Jim
 
Yes a vista vision sized sensor would be fantastic. although expensive, red are definitely in a position to do a large sensor and make lenses for it and give the option of using the windowed center with regular lenses just like we would do with 2k.

maybe an 8k sensor. thats one way of future proofing it. christopher nolan loved shooting on imax cameras but wasn't too happy about the weight and has also said he'd shoot the next batman completely in that larger format. if epic is something along those lines with a big sensor on it maybe it could be used on the next batman. although the lenses would be very expensive studio productions could easily swallow that cost.
 
Bigger Sensor Baby!!!

VistaVision(Full Frame 35mm) is doable for Red, maybe even IMAX!
They can beat the Phantom(and future others) who would make 65/70mm size cameras!:gun:
 
Yes a vista vision sized sensor would be fantastic. although expensive, red are definitely in a position to do a large sensor and make lenses for it and give the option of using the windowed center with regular lenses just like we would do with 2k.

although the lenses would be very expensive studio productions could easily swallow that cost.

If it's a Vistavision sensor, the glass would be cheaper than S35 glass because of economies of scale. There is a huge amount of fantastic 35mm stills glass that is CHEAP compared to cine glass. Any parts that can overlap with consumer products will be cheaper than specialized low volume parts. I was just thinking about this the other day when I realized that I could buy a pretty nice bicycle for the cost of 2 Red top mounts.
 
S35mm (possibly 4:3) is the most sensible spec for Epic if its aim is to replace existing chemical-film cameras in motion pictures and TV, because all the existing glass and supporting systems will remain usable, for the most part.

Probably the next step after Epic will be to full frame "Vista Vision", and this will be driven by cameras like the 5Dm2, 1DsM4, DSMC, etc. It will take some time for people to adjust to the idea of using still glass or for people to modify or produce new "full frame 35mm" glass.

I could be wrong. Jim could try to leapfrog directly to that step with a "full-frame" Epic, but that's something only he and a few people around him know right now.
 
If it's a Vistavision sensor, the glass would be cheaper than S35 glass because of economies of scale. There is a huge amount of fantastic 35mm stills glass that is CHEAP compared to cine glass.

Two comments:

To my knowledge there are no consumer grade lenses that would cover a Vistavision sized sensor. The lenses would have to cover a much bigger field than the s35mm or even full-sized 35mm lenses.

And there are already plenty of people using high-end stills lenses on their Red One, thanks to the Nikon mount and the IMS (and apparently soon some other mounts).
 
To my knowledge there are no consumer grade lenses that would cover a Vistavision sized sensor. The lenses would have to cover a much bigger field than the s35mm or even full-sized 35mm lenses.

According to Panavision and Wikipedia, Vistavision is the same width as still 35, but varying heights based on desired aspect ratio (between 1:1.63 and 1:2.00). Any still 35 lens would cover it. (Super35 is film run vertical. Still 35 and Vistavision is the same film run horizontal, just more perfs).
 
Back
Top