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 :)

Iso 8000...

Status
Not open for further replies.


Here are a few different interpretations of the MacBeth chart [based on their calibration tiff]. Don't take these as anything more than me tinkering in Photoshop - there's no accuracy here.
 
This was a shot that is 5 stops under-exposed in camera. Then it was pushed 5 stops in post.

Graeme
 
Not a bad Friday the 13th for the Red Team!

maybe 6 to 9 months from now there will be an excitement about a few thousand extra lumens on a new projector ....

what a crazy ride!! truly history in the making!

I was just thinking about all the fun stuff I can soon shoot ... Central Park in winter at night looks pretty amazing ...

magic hour in September ....

the skyline and bridges at night from Brooklyn ....

almost makes you forget about making money ...
 
Regarding these high ASA conditions... Since the retrieval of the near noise floor information when shooting REDCODE RAW is a post process for RED CINE... how is viewing facilitated at the EVF/LCD/monitor outs when shooting?

I figure some sort of "gain-up" is done for the live outputs, else you'd be hard-pressed to see your image at f22. I think that has been touched on once or twice before, but...

... my question...Is the "gain-up" of the viewing signal of good enough quality to make all esthetic and technical judgements?

Perhaps more importantly, if you choose to record REDCODE RGB this "gain-up" would need to be baked into the file... is it equal in quality to the results you'd get from post manipulation of REDCODE RAW?

My gut intuition is that it wouldn't be, but seems a fair question to ask as the answer would reveal more about the on-camera processing of debayer, gain, and scaling and would be of importance to those hooking up decks for HD-SDI out recording and 720p high speed work. I'm sure this stuff is designed locked now, and is answerable.
 
There's an electronic gain in the RGB pathway, so, if you shoot 5 stops under, you just put the EV compensation +5.0EV.

Graeme
 
Graeme Nattress said:
Yes, it was 4k REDCODE RAW, to a compact flash card. I wouldn't have believed it if I hadn't seen it myself.

Graeme
Click to expand...

I wouldn't have believed it either if I hadn't metered it at T4 and set it at T22 myself.

d
 
offhollywood said:
I was just thinking about all the fun stuff I can soon shoot ... Central Park in winter at night looks pretty amazing ...

magic hour in September ....

the skyline and bridges at night from Brooklyn ....

almost makes you forget about making money ...
Click to expand...

Have fun and make money at the same time - shoot those shots as 4k REDCODE RAW stock footage! It would be way cool to be the first to have RED stock of the scenery you described.

I regularly disappear into the beaches, mountains, coastal wilderness, and festivals near my home on the Central Coast, with cameras, and have fun getting great images - then I lay them up as stock footage and stills. They license really well, which always makes me want to have more fun out there - and make money at the same time!

Just one more good use for RED One...
 
Stuart English's recent gain remarks also shed some light (pun intended!) about gain on RED One:

"Yes, there is a Gain adjustment measured in 3dB steps available. That affects the RGB recordings, video outputs and EVF feed. When recording RAW the value is stored as metadata, so you can use this value for gain in post, or ignore it - your choice.

Stuart English
7/1/07
RED User
http://www.reduser.net/forum/showthread.php?t=3039&page=2


"In the same way that you can push a filmstock, you can adjust gain when "developing" RAW into RGB.

The visual result is similar - a brighter image but an increased noise floor."

Stuart English
(later in same thread)


"If you record REDCODE RGB then any Gain selected is applied prior to compression i.e its already affecting the RGB image the camera records. If you record REDCODE RAW, then Gain is just stored as metadata, and it is then used (or not, as its your choice) in the conversion of 12 bit RAW to 10 bit or 8 bit RGB / 4:2:2 in post production."

Stuart English
(later in same thread)
 
I'll definitely being advising everyone who uses our Red's to expose for highlights and let the rest fall into the black.

If in grading you can go that deep into the noise floor - well maybe not quite that far if the rest is well exposed and noise free - then what on earth does that do to the dynamic range rating?
 
this is astounding to me....absolutely astounding. This pic may be the biggest testament to the abilities of this camera...even more so than the beautiful pics of the milk girls and crossing the line.
 
Well if they lit it to F:4.0, that means that [at F:4.0] the 18% grey square should spot to "perfect" exposure [I.E. it should show a line at 50% on a waveform monitor]. I don't know the exact stop differences, but the darker squares would spot to what I assume to be several stops darker. I'm sure somebody here knows the exact stop difference.

Since this shot was underexposed by 5 stops [to achieve the 8,000 ASA rating], the 18% gray square would be 5 stops under "perfect" exposure. And the black squares? The ones that still clearly maintain detail on this chart? They'd be that many stops darker.

So from the "center" of the exposure, you'd have 5 stops [18% gray square] plus __ stops [black square] plus however many stops are visible beyond the black square. That'd just be the dynamic range below 50% of the exposure... who knows what's on top of the exposure.

This is by no means an exact science and I've probably screwed the pooch mightily in this post... but no matter how you measure it, it rocks the freakin' house.
 
Brook Willard said:
This is by no means an exact science and I've probably screwed the pooch mightily in this post... but no matter how you measure it, it rocks the freakin' house.
Click to expand...

lol Brook, you've got my brain bubbling as well, I see your logic behind the post. :biggrin:
 
Thanks Gibby and Greame for the info on the gain aspects of the signal path... I did not refine my question very well... I suppose the heart of what I am asking is if you will be compromised by the HD-SDI outs or the RGB codec if you are doing significant gain (ASA 4000-8000) vrs recording REDCODE RAW and doing it in post.

I understand the fact that you will have far greater room to manipulate with RAW in REDCINE, but if excluding fruther post manipulation from the question, would your live outputs via hardware have as robust de-bayering, downscaling, and gain as, say, the same default settings applied in REDCINE?

I would think this would be very important point to know for those wanting to record out to HDCAM SR decks, or the like as well those shooting high speed with 720p RGB...

If no metrics are known on this, I would imagine that it would be another good puzzle for the testing team...
 
Well I think the black square on a MB chart is roughly two stops under the 18% square... though I'm not 100% sure. Anybody with more knowledge of these things want to chime in?
 
It's always better to shoot RAW and work with it in post.

Graeme
 
Status
Not open for further replies.
Back
Top