Brett Harrison
Well-known member
It will be fun to look back on this thread in 4 months. To my eye the few things that have popped up have such a different texture to MX, even though they're arriving via the same color science.
Welcome to our community
Be a part of something great, join today!
Dragon/Epic comparison
- Thread starter Gunleik Groven
- Start date
Elsie N
Well-known member
- Joined
- Oct 31, 2009
- Messages
- 6,658
- Reaction score
- 0
- Points
- 0
I feel this thread is a +1 Dragon teaching tool based on Gunleik's methods. Very informative... and I think they will apply when the new color science is released, in re: choosing shooting methods.
Thanks Gunleik
Martin Stevens
REDuser Sponsor
Thanks.....thanks....thanks.......Gunleik.
Bobby Farokhzad
Well-known member
- Joined
- Feb 27, 2012
- Messages
- 576
- Reaction score
- 0
- Points
- 16
Just downloaded the R3Ds and played with them for a while.... the Dragon definitely holds better, no question there. Thank you for doing this Gunleik, much appreciated.
- Thread starter
- #165
Gunleik Groven
REDuser Sponsor
There is one thing I kinda get to say too often, it seems 
ISO does not affect or "create" or "diminish" noise
The light-levels hitting the sensor, determine resulting noise.
Thus it is fully possible to have a beautiful and noiseless image shot and posted at 12800 ISO.
And likewise, it is fully possible to create very noisy images, despite shooting at 400 ISO.
But this is more of a general camera thingy, and not really relevant for a comparison of these two.
Applies to both, though....
AND
A low exposure can be very clean, as long as you do not insist on pushing it silly.
Both of the last exposures I posted can be clean on both cameras.
But then they do not match anymore...
ISO does not affect or "create" or "diminish" noise
The light-levels hitting the sensor, determine resulting noise.
Thus it is fully possible to have a beautiful and noiseless image shot and posted at 12800 ISO.
And likewise, it is fully possible to create very noisy images, despite shooting at 400 ISO.
But this is more of a general camera thingy, and not really relevant for a comparison of these two.
Applies to both, though....
AND
A low exposure can be very clean, as long as you do not insist on pushing it silly.
Both of the last exposures I posted can be clean on both cameras.
But then they do not match anymore...
Hans von Sonntag
Well-known member
In other packages than RedCineX you'll find legacy CameraRGB. IMHO, a much more filmic colour interpretation than RedColor3. Because there are choices to transcode colour differently from RAW to RGB I hardly can say whether this or that is better. Fact is, a lot of people are attracted by the colour interpretation of Arri's Alexa and other get very nice colours out of MX. Nonetheless, it makes a lot of sense to work on new ways to bring colour from RAW to RGB.
RC3 with Drangon seems to be much more neutral in the hights than the MX in Gunleik's test which have a clear magenta cast. However, my own Epic does not exhibit such a behaviour. Interesting.
----
Hans
Mark Wilkinson
Well-known member
There is one thing I kinda get to say too often, it seems
ISO does not affect or "create" or "diminish" noise
The light-levels hitting the sensor, determine resulting noise.
Thus it is fully possible to have a beautiful and noiseless image shot and posted at 12800 ISO.
I'd like to see an example of an image "correctly" exposed at a metered 12800 with no noise. Anyone have an example they can post?
- Moderator
- #168
- Joined
- Apr 19, 2007
- Messages
- 13,392
- Reaction score
- 796
- Points
- 113
- Location
- Los Angeles
- Website
- www.phfx.com
Gunleik is merely producing a side by side moderate stress test and revealing the "stock" look straight out of camera. The magenta cast in overexposed highlights is something you should be experiencing normally with Mysterium-X, indeed interesting if yours does not produce this. You can minimize this effect utilizing the DRX feature in Redcine-X Pro to balance it correctly relative to the color of your light source and white balance choice.
Dragon however doesn't not need DRX. Which is why it is greyed out in Redcine-X Pro. Additionally Dragon itself has a very different highlight roll-off to the eventual clipping point that appears to "burn" rather than "slice". At least that's how I describe the effect. It does appear more organic this way and indeed provides less of a digital signature.
Hans von Sonntag
Well-known member
This is the reason why I question the RAW workflow in a way that I don't think that it is the best idea to have many different ways to transcode the RAW to RGB. I for my part set all my stuff, regardless how exposed, to 400 ASA 5000k RedLogFilm and CameraRGB or RC3. In many test I found out that I won't dig out more information or suppress noise when changing anything of these setting as long as I transcode to Log RGB.
In reality many users use funky settings in-camera for monitoring for whatever reasons. Then later the R3Ds get transcoded by a post-facility with these funky settings and the damage is done. The reason for the bad-mouthing Red often experiences is due to an unnecessary complex workflow. If there was an easy to use tool that let you select only between Log or Rec709 we had less problems. The same applies for camera settings. There is no need for selectable ISO settings. I shoot all my stuff with ISO 400. In environments with high contrast I just expose one stop darker. In environments with low light levels I stay with ISO 400 and give the sensor as much light as possible.
IMHO, the endless possibilities don't do any good and lead to funny tests such as developing the same R3D at ISO 400 and ISO 12000. Such a test does not show the sensor's abilities. For this, one has to change the amount of light that hits the sensor.
Arri's big advantage is an easy to understand workflow that does not lead to misunderstandings, misconceptions, pseudo know-how and badly exposed camera files. If I were Red the next RedCineX and SDK would be radically cleared out. But compressed RAW as a recording format is still a good idea and sets Red apart from the competition.
Hans
Hans von Sonntag
Well-known member
Phil that makes sense. DRX is a makeshift that naturally should be unnecessary when the DR is large enough to let the software create a nice, to the eye pleasing roll-off without artificial. secondary desaturation, etc... I never use DRX. When I find the roll-off of my Epic's pictures to "colourful" I desaturate the clipping parts with the grading application, in my case mostly Smoke.
Ok. Dragon seems to bring digital film more to or even beyond film, resolution-wise obviously in any event.
Hans
- Moderator
- #171
- Joined
- Apr 19, 2007
- Messages
- 13,392
- Reaction score
- 796
- Points
- 113
- Location
- Los Angeles
- Website
- www.phfx.com
I agree and believe that DRX's implementation with Mysterium-X is to allow access to the most possible usable dynamic range that MX has to offer. I go between grading it out and using DRX. Just depends on what I'm doing.
It would appear that with Dragon's rather "healthy" amount of total accessible dynamic range it's less of an issue as Red's calibration pushes that potential colorful effect outside the realms of what we see. Or perhaps they are manipulating it on the top end via some back end color science. Anything is possible really.
As Jim and Jarred said long ago when the original wedge chart showed up we all saw "a lot" of measured pure sensor dynamic range (hence people being confused about potential 20+ stops of DR), but once the color science and calibration is applied we'll get a large cut of that cake and the trimmings fall out of bounds or get used differently.
I've been down this street on a couple other cameras coming to market in the past and I really want to say kudos to Red for even showing us that wedge. That's stuff that usually stays behind the curtain. Way behind. However, it showed exactly how much potential there was going to be with Dragon and that in itself is what matters. I'd be curious how the difference between the RG3/RC3 and RDG/RDC color science reveals more or less or changes the overall color in comparison when that time comes. We might get even more out of Dragon than what we're experiencing right at this very second.
- Thread starter
- #172
Gunleik Groven
REDuser Sponsor
Hans.
You are in many ways right.
Hahahaha
Which means; I largely agree...
RLF ISO 640 WB in the 4400 - 6000 region is what we landed on after testes many times.
Or linear/320/5600 for some uses.
To some 16-bit linear format, or even out of the rocket into some grading app, if a transcoe is not on the table.
That is fine, and corresponds well with the Lightiron workflow case studies I have read, and fits with most testing I have done myself.
But that approach is kinda preventive and frankly a bit confusing for many users.
And I have come to an approach that there is no point in being religious about "best", if that approach is preventive and gives a "less than good" result for a lot of users.
Many insist on a pregrade, and I recently had a discussion with a good friend and otherwise good imagetreater, who was concerned with the results he got by grading in RCX.
I tried to suggest another way, and kinda showed him that what he was complaining about was the consequence of "RAW grading".
His response was;
Well, that is really counterintuitive. It does look like a color app, right?
And I do that at times myself. Quick grades in RCX.
I know the tech tradeoffs, and sometimes there is no point in not doing that.
"Whatever gets you where you imagine you want to get".
But I have had some real issues with the wb tools for the MX sensor in particular.
What I see on the Dragon, at least at first glance, is that they work "like you would think they should".
But I admittedly have not been digging, and I guess that technically the "flat" approach will have an edge.
but as the "easier" way starts to "work well", there is no reason to not advocate it over a more tech- and knowledge-hungry approach.
I really dig it.
Simplicity in post is only good!
Understanding exposure is also only good.
Dragon is a lot more allowing than any other digital sensor I have seen.
but that does not take away the edge for those who can expose for the image they want.
But as I discussed in an earlier thread.
there is really no "RAW" workflow.
What we grade is allways an RGB image, and thus normal RGB/Signalchain/color-rules and knowledge apply.
there is no magic to exposure or color.
Even with RAW...
But Dragon makes it more likely that more people will maybe get closer to what they envision, because it has so much more playroom, and will make the good ones shine even more.
that is kinda a win-win-win situation in my eyes.
Still, if people expect that it will set light for them or color there images or make their highlights and contrsta all by itself, they will be in for a hard time...
OTOH
That hard time is really called "learning"
Learning is good.
Do you not think?
I used to be very hardcore in my workflowadvices, as "this will get the best result", but to many that wa only confusing, as they did not have the knowledge to get "more" by getting more control over the image.
And then it is counterproductive, would you not say?
LOL
As to CamRGB.
THAT (if my memory serves me, maybe Graeme will chime in), was the unmatrixed RGB for the original M sensor.
If it works the same way on the MX... I really do not know...
Or if it applies an M matrix to the MX.
Whatever it is in an MX or Dragon context, most people will not have any reason or knowledge to push that button.
And it is at least a bit undocumented to what it does to the "post M" sensors.
But if you like the look;
All the power to you!
THAT is kinda my point
Technically, it is possible to measure what way that could be considered "the right".
But a lot of people without that tech knowledge can get really far with just very intuitive tools.
And familiarize themselves with images and workflows, until they - like my friend, start to see things they do not like, and will have to learn ways to deal with them.
What I see with Dragon is that the threshold for "need to know" has been risen significantly.
Which is good.
less geeky questions for me.
More fun work for all of us.
bless RED for going there!
G
(pragmatic)
- Thread starter
- #173
Gunleik Groven
REDuser Sponsor
Hahahaha
Know what, I agree. Actually you reverse-state my point by including "metering" in the argument!
Take the filmstock equivalent...
I guess you would hardly rate a 200 ASA stock at 12800
At least you would know that IF you chose to do that, you would create a very noisy image...
To achieve a somewhat useable image you would swap to a more lightsensitive stock. Right?
Similarily I know that I do not swap the sensor or add any analogue gain to it to increase sensitivity by sliding the ISO slider.
So I would do what most everyone who have showed off "nice" 12800 images here have done.
Overexpose like hell, and "recover highlights" in post.
Now to me that is not "magic" but just a very impractical way of aquiring images.
That said.
I am capable of doing it...
A lightlevel of X will allways create the same Signal-/noiselevel on the same sensor calibrated the same way.
The difference ISO does, is where in the visual specter (where in the curve) that signal/noise is visible.
People playing tricks with high ISO "overexposure" and redlogfilm "recovery".
Seriously.
If it gets them what they want, I think the approach is good.
Still I find it to be a slightly complicated way to work.
Detlev Eller
Well-known member
... ... love it!
... love it!
Detlev Eller
Well-known member
love this data!
- Thread starter
- #176
Gunleik Groven
REDuser Sponsor
If I am not much off, this thread is sorta "done".
General exposure and workflow stuff can be discussed elsewhere.
And I have actually requested the thread to be closed...
Just one last thing (hahaha)
The last two R3Ds I posted can be quite usefull if you play with them to match them through ISO and look at the results you get.
try:
400/12800
200/6400
100/3200
50/1600
For the lower, it gets increasingly "useful" imho as you actually lower it to what it physically is....
There is in my opinion a lesson to be had from that.
Expose low. It works.
If you want a dark image...
If not, add some contrast.
LOL
Thx
Has been fun.
I will post general Dragon stuff somewhere else.
G
General exposure and workflow stuff can be discussed elsewhere.
And I have actually requested the thread to be closed...
Just one last thing (hahaha)
The last two R3Ds I posted can be quite usefull if you play with them to match them through ISO and look at the results you get.
try:
400/12800
200/6400
100/3200
50/1600
For the lower, it gets increasingly "useful" imho as you actually lower it to what it physically is....
There is in my opinion a lesson to be had from that.
Expose low. It works.
If you want a dark image...
If not, add some contrast.
LOL
Thx
Has been fun.
I will post general Dragon stuff somewhere else.
G
Humberto Ramos
Well-known member
I was paying with some of the R3D and I think I'm in love with this camera :drool5:
Last edited:
Mark L. Pederson
REDuser Sponsor
Well ... I hear this all the time but I think it's a sad statement that just points to how lazy and resistant to learning some people are today.
When I went to film school(s) (I was kinda thrown out of two of them) we had to learn and understand how to expose motion picture film. We learned how to use a light meter. We tested various film stocks, under exposed, over exposed - we experimented and learned. When I first learned how to use RAW it just felt like digital film stock to me.
RAW isn't exactly the most complicated thing in the world. The guys in stills managed to learn it - leverage it - and use it without "misunderstandings, misconceptions, pseudo know-how and badly exposed camera files" every day all over the world.
Sometimes I feel like there is this "point and shoot" generation that has never even touched a light meter or learned the basic, fundamentals of cinematography / photography.
Making the tools "dumber" might get you a bigger market share - but ... we'd all have better images if more people paused to learn the fundamentals.
Again, ask any still photographer who makes a living shooting images if he or she finds RAW difficult.
Nick Morrison
Well-known member
- Joined
- Nov 9, 2011
- Messages
- 9,141
- Reaction score
- 50
- Points
- 48
- Location
- Brooklyn
- Website
- www.smallgiant.tv
To add to this, I work at huge cable network where we get Alexa files all the time...and they are not always "easy" to work with either. They rarely come in graded, they usually look flat and sickly, and most editors throw up their hands and ask "why does this look like shit??". Sometimes they come looking for me to answer this question, cuz they know I "know stuff about cameras".
To me, a lot of the blame falls on production companies, period. Not post houses. In this day and age, I strongly believe you need to own your image. To me that means always delivering GRADED shoot footage to clients, accompanied by the the raw elements so they can re-grade if they need to. We agree on a look on set, and plan to deliver that way. Otherwise the workflow in post can get messy. Regardless of what camera you shoot on.
My clients can have up to a dozen editors's work on a project at one time, so if my footage doesn't have a ONE LIGHT baked in, deliverables canstart to look radically inconsistent as each editor grades independently. This feels too messy for me, so I personally chose to control the data and pregrade my deliverables.
But's just me.
Mark L. Pederson
REDuser Sponsor
Put that on a T-Shirt.