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...

Test was done motion at 24fps and 1/48th shutter.

Jim

all these little tidbits are too much! When you spoke of touching up the raw files are you saying the results will be up tonight? or is the team taking the night off ( for once haha) and posting the results later?
 
Jim,

Can we use HDR variable mode as normal shooting tool at any situation instead of normal dynamic range ? Any artifacts ?
 
That assumes the 1st step is blown out. If there is tone in step #1 and separation between the last stop and the background, you have more than 17 stops (probably just under 18)... on this DSC chart.

Jim

I gladly stand corrected. Thanks Jim.
 
Jim,

Can we use HDR variable mode as normal shooting tool at any situation instead of normal dynamic range ? Any artifacts ?

That's a good question.

all these little tidbits are too much! When you spoke of touching up the raw files are you saying the results will be up tonight? or is the team taking the night off ( for once haha) and posting the results later?

At this point RED needs to take responsibility for everyone's sleep deprivation!

:w00t:
 
Throwing out a guess as to how this works...

There is 1/48 of a sec to take frames, and the Epic can shoot 120 fps. There is a read-reset time between frames, so let's pretend it's 1/50 sec we've got total. The way to get HDR in a motion picture camera is to take 5 frames in quick succession, but with *variable shutter speeds*.

The first frame is 1/100 sec; second frame is 1/200 sec; third is 1/400; fourth 1/800 and fifth 1/1600. That's a total duration of 31/1600 which is just under 1/50 second. Since each exposure takes in one-half the light, you're highlights are less and less likely to clip. Combine the frames using HDR software and you've got a huge amount of dynamic range within a traditional 1/50 second shutter.
 
Congrats Jim- that is VERY exciting news indeed. I understand about wanting to show the image in its best light after Gramme has worked his magic. But is there any chance you can give a little more specifics / details as to what you and your team was seeing? Even a range / ball park of what was being captured?
 
That's a good question.



At this point RED needs to take responsibility for everyone's sleep deprivation!

:w00t:

I don't think the sleep deprivation will be going anywhere until Epics and Scarlets are in our hands and we are shooting. Even then we'll be up all hours of the night pushing these new cameras to the limit!

Never has a camera company make me lose sleep...until the Scarlet was announced. Bravo Jim and company!
 
I live just south of Boston and its almost 3 AM here,
but it's hard to leave this "moment in cinema history"
and go to bed.
 
Throwing out a guess as to how this works...

There is 1/48 of a sec to take frames, and the Epic can shoot 120 fps. There is a read-reset time between frames, so let's pretend it's 1/50 sec we've got total. The way to get HDR in a motion picture camera is to take 5 frames in quick succession, but with *variable shutter speeds*.

The first frame is 1/100 sec; second frame is 1/200 sec; third is 1/400; fourth 1/800 and fifth 1/1600. That's a total duration of 31/1600 which is just under 1/50 second. Since each exposure takes in one-half the light, you're highlights are less and less likely to clip. Combine the frames using HDR software and you've got a huge amount of dynamic range within a traditional 1/50 second shutter.

Sounds very good.....but I think this would introduce a sort of strobing in the blurs.
 
Sounds very good.....but I think this would introduce a sort of strobing in the blurs.

Yeah... Every theory on temporal-based HDR suffers from this.

I'm not sure there's a clever way to interpolate between different blurs, it would be a hard thing to do.

My bet is on the Bayer sensor... But I'm just a noob.

I'm curious!
 
okay Jim, you had me at hello :-). This is a big night, light up a good one! I'm dying to see what went down!
 
Throwing out a guess as to how this works...

There is 1/48 of a sec to take frames, and the Epic can shoot 120 fps. There is a read-reset time between frames, so let's pretend it's 1/50 sec we've got total. The way to get HDR in a motion picture camera is to take 5 frames in quick succession, but with *variable shutter speeds*.

The first frame is 1/100 sec; second frame is 1/200 sec; third is 1/400; fourth 1/800 and fifth 1/1600. That's a total duration of 31/1600 which is just under 1/50 second. Since each exposure takes in one-half the light, you're highlights are less and less likely to clip. Combine the frames using HDR software and you've got a huge amount of dynamic range within a traditional 1/50 second shutter.

Imagine the motion blur hell that would produce.. Im still conviced that the HDR mode takes advantage of different lines for different exposures, but with using ISO rather than shutterspeed to make the difference.

OR the sensor has two sets of read circuits, again allocating different ISO's to each
 
Throwing out a guess as to how this works...

There is 1/48 of a sec to take frames, and the Epic can shoot 120 fps. There is a read-reset time between frames, so let's pretend it's 1/50 sec we've got total. The way to get HDR in a motion picture camera is to take 5 frames in quick succession, but with *variable shutter speeds*.

The first frame is 1/100 sec; second frame is 1/200 sec; third is 1/400; fourth 1/800 and fifth 1/1600. That's a total duration of 31/1600 which is just under 1/50 second. Since each exposure takes in one-half the light, you're highlights are less and less likely to clip. Combine the frames using HDR software and you've got a huge amount of dynamic range within a traditional 1/50 second shutter.

This was my initial guess as well. Perhaps different gain for each exposure as well. The only problem I see is there could be problems with motion. Fast moving object can move quite a bit in 31/1600s of a second. Though it would work well for many situations. In a way, it is intelligent bracketing.

In reality, I think/hope RED have an innovative approach we aren't aware of. Seeing as RED shoots RAW - the camera can intelligently recognize which parts are bright, which are dark, and adjust FLUT/ISO accordingly, in camera, for maximum DR. We could actually gain a lot of DR in post-production this way, by separating the dark and bright parts.

Importantly, I really want to know what the trade-offs are.
 
yet there will still be people who complain and bitch about RED.. because they are simply afraid...what a shame:(

Forever my friend kept saying Scarlet was vaporware...but why would RED, a quality company, fabricate a product? Seems a bit ridiculous. He is all for the AF-100, says the GH1 is all he needs with the firmware hacks.

But I keep insisting there is more to Scarlet that he is willing to see. Maybe he needs an increase in dynamic range to understanding what to expect of a video camera in today's and tomorow's world.
 
Back
Top