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 :)

Off the chart...

How about each pixel can be read before it clips?
 
How about each pixel can be read before it clips.

Yup, that's kind of what I meant. The camera AI detects which areas are overexposed, and brings the exposure (i.e. FLUT/Gain/whichever way) down for those pixels. It does the opposite with dark pixels. Could work. I still think RED have an approach we know nothing of.
 
Well, there will always be personal bias. We will see when the Scarlet is here. Would he really choose an inferior product?

No, I think with some hands on use of my Scarlet he will see the bright light haha. I do not mean to discourage those who have bias and are comfortable with their own cameras now, for me my D90 is a great camera, but everyday I am learning and understanding that I can and should expect more of the products I'll be depending on professionally in the future.
 
Yup, that's kind of what I meant. The camera AI detects which areas are overexposed, and brings the exposure (i.e. FLUT/Gain/whichever way) down for those pixels. It does the opposite with dark pixels. Could work. I still think RED have an approach we know nothing of.

Now that's a real FLEXIBLE LUT!
 
How about each pixel can be read before it clips.

That's similar to another idea I had, but requires hardware I don't know exists in the MX sensor: self-resetting sensors that increment a binary counter when they clip. This would (in theory) turn a linear photon counter into a logarithmic one. Even with just 2 bits for each sensor, you could quadruple the amount of photons the sensor could "sense" without clipping.

(I have no idea if this is even remotely feasible in actual hardware.)
 
No, I think with some hands on use of my Scarlet he will see the bright light haha. I do not mean to discourage those who have bias and are comfortable with their own cameras now, for me my D90 is a great camera, but everyday I am learning and understanding that I can and should expect more of the products I'll be depending on professionally in the future.

Mind you, it is perfectly understandable as to why a person may choose a DSLR or AF100 over a Scarlet. They shouldn't just dismiss it without considering it, though.

Now that's a real FLEXIBLE LUT!

There are many ways to gain DR from RAW files in post by separating the dark parts from the bright. We had a member here posting about it as well, a while back. Perhaps the RED DSMCs have become intelligent enough to do it in camera?
 
So the obligatory wet towel question:

Why wouldn't we leave our Epics in HDR mode?

And you refer to this as "epic" HDR mode... will this only work with Epic?

Yes, it is the trade-offs I am most interested in.

And no, the first time Jim mentioned "HDR Mode", it was for both Epic and Scarlet. Of course, everything is subject to change, but let's hope it makes it.
 
As far as I remember HDR was only meant for Scarlet 2/3... Maybe I am wrong... It was said a long time ago...
 
HDR video mode might reduce your maximum recording frame rate. But if most of the content you're going to be shooting will be 24 fps, who cares?
 
Why wouldn't we leave our Epics in HDR mode?

1. Data size

and one or both:

2. (a) Temporal Artefacts

or/and

2. (b) Decreased Frame Rate

You see, you can trade DR for the above drawbacks.
 
How about each pixel can be read before it clips?

There are a bunch of ways it could work, even without very special sensors - all because EPIC and Scarlet are such high-bandwidth cameras internally. The MX sensors have really low read-reset times, even if you just do two reads in 1/24th of a second and HDR blend them with Graeme magic... the results would be astounding. I'm Psyched.

Jim, are you ready to let us in on the part of the secret?
 
Yup, that's kind of what I meant. The camera AI detects which areas are overexposed, and brings the exposure (i.e. FLUT/Gain/whichever way) down for those pixels. It does the opposite with dark pixels. Could work. I still think RED have an approach we know nothing of.

Ugh.. you might be right, but I'd hate to see that be true, as much as I hate dynamic contrast in modern TV's
 
Send film to the retirement home, and let the revolution begin!
Guys, I'm sorry to say that I'm not a Red One owner, but reading posts like this makes me beleive that the Epic isn't such a bad place to start either!
I am ready to enter the world of RED now :-)
 
There are a bunch of ways it could work, even without very special sensors - all because EPIC and Scarlet are such high-bandwidth cameras internally. The MX sensors have really low read-reset times, even if you just do two reads in 1/24th of a second and HDR blend them with Graeme magic... the results would be astounding.

If you do two exposures instead of one, you will gain at most 1 stop only and introduce motion artefacts: Hardly HDR.
 
1. Data size

and one or both:

2. (a) Temporal Artefacts

or/and

2. (b) Decreased Frame Rate

You see, you can trade DR for the above drawbacks.


If it somehow did sub-frame sampling and you just lost framerate or shutter speed options or base sensitivity--that would be amazing. I hope that's the tradeoff vs some form of progressive exposure system which results in temporal artifacting. Temporal artifact would make it far more limiting in when it could be used.
 
Back
Top