Welcome to our community

Be a part of something great, join today!

Log in
  • 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 :)

Noisy Epic footage?

Patrick Tresch said:
Who told you so?
Click to expand...

Hey Pat,

I'm going on my understanding of blackshading from Deanan (when he was at RED) and when I first ran into HOT PIXELS with the original RED ONE.
At that time Blackshades were only required at the shutter angle, framerate and, TIMEBASE.

That's it.

Other that what was recently posted in this thread as quoted by Deanan, that is all I have to understand it from the source.

Most of the "problem noise" is blue channel noise. From what TED has explained to me in interviews for articels I have written, it's wise (in the M days) to "activate" the blue channel, particularly with Tungsten sources (by using blue gel or a blue filter) to help eliminate some of the blue channel noise caused by low light or tungsten sources.

In further tests I have found that this works well with the MX sensor as well in both the EPIC and the RED ONE MX.

I've never heard from any RED source that the metadata affects BLACK SHADING CALIBRATION other than the shutter, framerate and timebase.

Hope this helps.

David
 
"How to expose" is really a subject for another thread. Hearing of noise issues from the common user it would be great to have some feedback from the factory how features like Spiderman4 and The Hobbit dealt with it as it doesn't seem to be a problem isolated to a specific Epic.
 
Will Keir said:
Shooting with the red for a couple years now, this is exactly how NOT to get the shot. What I see on the monitor MUST be check with a histogram to avoid shooting over/under exposed.
Is your claim that with an epic this is all fixed? That the view finder shows what you will see in post? That, I would like to believe.
Click to expand...

All I need to see in my monitors is that there is no blown out skies, and there is info in my black areas... It's as simple as that.
Some of you guys are making this out to be way more than it is. Or more difficult than it needs to be.
My kids shoot my EPICs with ease, none of them come to me with badly exposed images either and they don't even know what a histogram is or let alone understand it. I just say spin the aperture ring around until the picture looks right... So far they have 100% success rates.

I've been shooting RED for 4 years now... (yes I was a very early adopter - build 14 days.. camera 158 was out first)

Back when a highlights were a huge problem... I was getting perfect exposures then, and still get perfect exposures now.
It's not rocket science Will... why do you think it has to be.

Also, What I see in post is what I want I shot on set... why would it be any different? ... All my data, goes to REDLOGFILM, all my onset meta data gets deleted, as I don't need it. I create new LUTS that suite my purpose, I don't use any onset meta data at all, ever.

Maybe you should get to know your cameras a bit better. :))) The word "EASY" is an understatement.
 
Hello David,

No, the metadata doesn't affect the black shading.

People think they expose correctly a scene by changing metadata to 3200k, when in fact they are just pushing a channel and adding noise.

On the present case shown by Mr. Haase, I don't know if he had another issue? Is the Epic really noisier than the One-MX? I should test... But my "impression" : it's not...

Black shading can have an effect on the noise level IMundertanding if the BS calibration wasn't done properly for that current setting (Build/FPS/shutter).
 
That's right.

If the calibration was done for 1/2 second exposures then that setting might not be great for high speed shots.

Thats sort of it basically.

I looked at ERICs R3D and I did not think it had excessive noise compared to a RED one MX. But if I zoom past 200% I start to see some.

David
 
Patrick Tresch said:
Do you say "less exagerated" due to the black shading? As compression from 8 to 3 has only little effect over noise?
Black shanding, exposure and control over color temp could be the key?
What do you think?

I remember the early days of ONE-M, with blue screen issues due to misbelief that color temp is only a metadata thing that can be handled in post. Poeple got a lot of noise in the blue channel when lit in T.

RAW is not only Metadata! ;-)

Pat
Click to expand...

As per my original post, I black shaded properly before the shot, exposed as I normally would and have been doing since day 1 of the MX sensor under lighting conditions I have shot a hundred times. The shot is noisier than I would expect. I understand everything about exposure, histograms, color temp, blue channel, etc etc but the shot is noisier than it should be.
 
Eric,

Thanks for starting this great thread. There is a lot of rich information here. If you have looked at tons of MX before and then this is your conclusion then I just have to trust your judgement. It could be the larger frame size. What if you do a 4K comparision to a 4K shot you did on R1 MX and do 4K and 4K side by side.

If you say it's noiser, I trust that based on what you are saying here. Strange, personally I find the EPIC stuff sharper with less noise, and, then again, they are different cameras, even though they share the same sensor.

David
 
David Battistella said:
Eric,

Thanks for starting this great thread. There is a lot of rich information here. If you have looked at tons of MX before and then this is your conclusion then I just have to trust your judgement. It could be the larger frame size. What if you do a 4K comparision to a 4K shot you did on R1 MX and do 4K and 4K side by side.

If you say it's noiser, I trust that based on what you are saying here. Strange, personally I find the EPIC stuff sharper with less noise, and, then again, they are different cameras, even though they share the same sensor.

David
Click to expand...

I am trying to get a hold of the side by side R1 MX shot I did. In my opinion the Epic is noisier on this shot. I am trying to find out if it is just this particular Epic, some black shading gone wrong, firmware, compression, or just a general difference. I'm hoping today to be able to shoot a test with this Epic and an MX and another Epic to compare.
 
Thanks Eric,

Remember the default calibration is done at about 23degrees.
I have done my calibration at 48degrees with good results.

It might be good to grab a shot right after startup and then about 15 minutes later on the EPIC just to see if it is an ambient temperature thing.
Maybe see if there is a difference between default and USR .
A 5K/4k then 4K/4K would be good to see too.

Thanks for doing all this.

David
 
Mark Toia said:
True.. but bad exposure is the biggest reason people get noise.
Click to expand...

LOL!!!!!!!!!

There are many ways to get a good exposure, but what Mark says here, is just soooooooo right.

Might just add that bad metadata-treatment is the 2nd biggest reason.

Get friendly with the camera for the situations you are shooting for.

Personally I think noise is a non-issue with the Epic.

Not that I CANNOT get noise, just that I know how to avoid it if that is desireable.

You need some contrast to make images... use the lights, natural or studio to create images.
Without contrast and in low light, you'll basically get noise.


I am tempted to trelease my side-by-side R1 MX/Epic test, shot with 35mm masters on both, by now...
They are really different beasts IMHO...

This thread is kinda funny...

:))
 
'Bit short on time, but If I released a set of a/b single frame r3d's and let's say 3 5 sec exposures from each cam, shot in parallell with the same glass, would it be of some interest?

I did it under strictly tungsten, to provoke any issues.

Shot 4kHD on both (or 16:9), not to give the Epic the extra edge of downsampling, and as that is the relevant format for TV here at least.

Cheers!

G
 
Gunleik Groven said:
'Bit short on time, but If I released a set of a/b single frame r3d's and let's say 3 5 sec exposures from each cam, shot in parallell with the same glass, would it be of some interest?

I did it under strictly tungsten, to provoke any issues.

Shot 4kHD on both (or 16:9), not to give the Epic the extra edge of downsampling, and as that is the relevant format for TV here at least.

Cheers!

G
Click to expand...

Just do it. You know you want too..... :)


David
 
OK.

Won't get around it today, but within a week...

That comparison gave us the answer to whether we wanted to shoot a 112 shootingday series with a couple of Epics or a R1 MX and and an Epic...

The series has been shooting on two Epics since Oct 15th.

BTW, noise was not the single parameter we were judging the images from...
 
It seems like the vast majority of 'The Social Network' was shot under low light tungsten sources. Does anyone know how Cronenweth activated the blue channel while maintaining the low lit, tungsten tones without too much noise?
 
Andrew Huse said:
It seems like the vast majority of 'The Social Network' was shot under low light tungsten sources. Does anyone know how Cronenweth activated the blue channel while maintaining the low lit, tungsten tones without too much noise?
Click to expand...

I believe he rated and shot with an asa rating of 500 instead of 800. Cronenworth knows how to expose. GWTDT utilized Kino's for most of the lighting of interiors.
 
Ok, I was able to shoot a few tests today. Here is the methodology. Please keep in mind my primary purpose was to test my Epic against another Epic to see if there are any differences in noise at 800 ISO.


1) Cameras powered up for 10 minutes
2) Both Epics on firmware 1.7.10
3) Both Epics set to 180 degree shutter, 23.98 fps, 3200K, 5KFF, 7:1 compression - R1MX set to 4K 16x9 RC36, 23.98, 3200K
4) Both Epics and R1MX blackshaded immediately before shoot using Factory option and selecting Current fps/exposure (on Epic)
5) Scene photographed under tungsten lighting with bright exposure on face of subject and dark background. Aperture set to 2.8/4 split on each take.

Here is a link to 3 R3D trims. I was specifically looking for noise in the right and lower right part of the frame. There are slates on each shot. You can draw your own conclusions first.

http://dl.dropbox.com/u/904379/epic_mx_noise_test.zip

Have fun.



 
You must log in or register to reply here.
Back
Top