Which would be better for low light performance?

[Luis Avila]

Of these two, which would be better to shoot in low light.

A Scarlet S35 with Red Primes at T1.8
OR
Scarlet FF35 (larger sensor) with Compact Primes at T2.1.

Would they be about the same in terms of low light capabilities or am I completely off in my logic. Which would you prefer?

 

[Chuck Z]

The Scarlet FF35 will be developed with the Mysterium Monstro in mind. Therefore, one might assume that it will have better low light capabilities.

 

[Luis Avila]

The Scarlet FF35 will be developed with the Mysterium Monstro in mind. Therefore, one might assume that it will have better low light capabilities.

Well the S35 with Red lenses would have a lower Tstop than the CP2's, wouldn't that slightly make up for a smaller sensor?

 

[Bruce Allen]

FF35 has:
- greater surface area (overall AND per sensor-site) for collecting photons
- next-generation sensor and AD circuitry (so better at collecting those photons)
- higher res (better chance for clever NR that trades off some resolution for noise)

S35 has:
- slightly faster lens

I'd say the FF35 with slower lens will win.

Bruce Allen
www.boacinema.com

 

[Jacek Zakowicz]

not so simple

not so simple

Well the S35 with Red lenses would have a lower Tstop than the CP2's, wouldn't that slightly make up for a smaller sensor?

All things being equal it would be true. But this is not the case.
There are many factors involved
Sensor design, sensor native s/n ratio, DSP, DSP hardware design, firmware, compression(on the recording side)
You can safely say that once next generation of components arrives all bets and comparisons are off- typically the newer generation is vastly better than the older one...

 

[Michael Olsen]

Indeed. RED ONE featured the first generation "Mysterium" chip.

RED ONE can now be upgraded to the second generation "Mysterium-X" chip. This same chip will live inside the 2/3" and S35 Epic and Scarlets.

The FF35 Epic and Scarlet will use the third generation chip "Mysterium Monstro".

Assuming the advances made between M-X and Monstro are anything like the advances made between Mysterium and M-X, the FF35 would be superior.

This is, of course, speculation more than a year (I would think) into the future. Epic isn't even shipping with the M-X chips, yet. And Monstro is still a prototype and has a long development journey before it.

 

[Alexander Ibrahim]

I think we can safely say that no one outside RED can even make an educated guess.

So long as its just for fun, we can waste time.

If you want advice to use in your business planning, then my advice is to work out a plan to make money NOW using existing technology.

Don't worry about future technology until the future when the future technology in question is real.

So, for entertainment purposes only...

I can say that t/1.8 and t/2.1 aren't very far apart.

As everyone else has noted, it is expected that Monstro sensors will deliver much more than the MX sensors.

Right now I think the Mysterium MX sensors can let you shoot as fast as ISO2000 with good results. I expect the Mysterium Monstro sensors will allow us to shoot with similar results at ISO 4000. That is actually a more modest gain in performance than we saw between Mysterium and Mysterium MX.

I don't think I'd set Mysterium as high as 800, and certainly not 1000 without the help of post noise reduction software. Typically I won't push it past ISO 640, and I prefer to set it at ISO320.

Still that expected performance gain far outstrips the aperture difference between these two lenses.

So... based on two pieces of speculation, my wild ass conjecture regarding a future guess is that the FF35 "Monstro" camera with a t/2.1 lens will provide a better low light performance level than "current" s35 MX camera with a t/1.8 lens.

 

[Subhadip Sen]

Unless something goes wrong with development, the FF35 will certainly have greater low light. Not because it is the larger sensor - remember it will have to capture more pixels too. But because of the Monstro sensor. Much like the M-X reduced the noise floor and increased useable sensitivity greatly. One could hope that Monstro is as much, if not more, of a leap forward. I remember Jim mentioning that Monstro has a target DR of 15 stops.