Welcome to our community

Be a part of something great, join today!

Register 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 :)

Getting full Dynamic Range from Dragon

That doesn't have much to do with compression as much as exposure. It's still sensor data being written to the R3d.

Battistella
 
In my experience Dragon has slightly less DR than Alexa, which has about 14 stops.
 
Dunkan Wolf said:
In my experience Dragon has slightly less DR than Alexa, which has about 14 stops.
Click to expand...

In this test http://indiecinemaacademy.com/red-e...l&utm_source=facebook.com&utm_campaign=buffer
made by Ryan E. Walters we see in the chart that the Dragon has more DR than the Alexa.

30005_Sekonic-ChartDragonAlexaMX.jpg
 
My 2p on this issue... I see nothing wrong with Red's DR claims. I'm one of likely a pretty small number of people to have tested it on a xyla-21

People underexpose Dragon ALL the time. Because it has such broad dynamic range, and because most scenes contain less relevant contrast than is available, people naturally expose typical scenes much lower than is optimum. Now you get decent results, but you're already digging the data up before you start. You cannot do it twice.

Gio scope fixes this, but then I've seen people first hand expose using it, then switch out of the overlay and stop down because they don't believe the image on their LCD...
 
I can't wait for GIO scope, is it still beta?
 
Why is 11 on the GIO Scope mid gray? I would hope that 8 would be.
 
Alexandros Tsoupras said:
In this test http://indiecinemaacademy.com/red-e...l&utm_source=facebook.com&utm_campaign=buffer
made by Ryan E. Walters we see in the chart that the Dragon has more DR than the Alexa.


30005_Sekonic-ChartDragonAlexaMX.jpg
Click to expand...
Alex you have misread the chart that you posted. Counting from 0 (where the chart says EV Scale) to the right this chart shows that Dragon clips (the flat purple line) at about +7.3, while the Alexa (yellow line) keeps moving to the right. Same thing on the underexposure side of the chart, where Dragon basically flat lines a little earlier than Alexa. What's interesting in this chart is where each camera places middle gray. Dragon puts mid gray point brightness at 1.3 stops darker than Alexa. As you move down stops each Dragon gray chip is exposed about 1.3 stops closer to the noise floor than Alexa. This chart seems to offer an explanation as to why there is noise associated with skin tone olpf when exposed at 800 ISO, as the gray chips under mid gray are mapped close to the noise floor. Exposing Dragon at 320 ISO would move all the gray patches up 1.3 stops away from the noise floor thus improving noise. Of course, by doing that you add 1.3 stop to the dark end of the scale and take away 1.3 stops of highlight latitude. In reference to the OP question, I would say that that dynamic range to do what he wants is there, but one has to expose correctly in order to do it. Trying to pull dark images out of a Dragon that has been exposed at 800 ISO is not the best way to do it. Shooting at 320 ISO is going to give one a lot better chance at bringing the darks up with less noise. Notice in the still camera article they shoot at 100 ISO, which probably gives their images lots of exposure in the darks thus allowing them to make the dark parts of their image brighter. Hope that helps.
 
Martin Stevens said:
Why is 11 on the GIO Scope mid gray? I would hope that 8 would be.
Click to expand...

I may just be imagining this, but doesn't it vary by ISO rating? I flipped it on briefly during a shoot last week just to take a look and I could swear 10 was showing as middle grey. But didn't have time to investigate.
 
Martin Stevens said:
Why is 11 on the GIO Scope mid gray? I would hope that 8 would be.
Click to expand...

Scott C said:
I may just be imagining this, but doesn't it vary by ISO rating? I flipped it on briefly during a shoot last week just to take a look and I could swear 10 was showing as middle grey. But didn't have time to investigate.
Click to expand...

No. In Gio Scope what you are looking at is the graphical representation of the individual stops available to the raw image. It will not change with ISO, Look, or any other adjustment.

The reference point of Middle Gray at 11 is exactly that. Just a reference point. Interestingly enough if you toggle between say Video Check and Gio Scope you can see the effect of the Color Space and Gamma Curve pretty clearly and what Gio Scope considers Middle Gray is actually rather broad. This is a new tool and is a bit different than how you would normally employ a common exposure method. For instance, you can light and expose to Middle Gray on set that would produce a value of 8 on Gio Scope being Middle Gray, however, in relationship of the Color Space or Gamma curve that might appear under exposed.

Where I find Gio Scope particularly useful is for Middle Gray, examining highlight roll-off and potential clipping issues, as well as defining areas of the noise floor I would like to put markers on in case I can expose differently or add some light into the equation.
 
Okay.... But, if RAW GIO view has the base ISO fixed at 800, then putting middle gray at 11 means LESS highlight latitude than if it were at 8.
 
Phil Holland said:
For instance, you can light and expose to Middle Gray on set that would produce a value of 8 on Gio Scope being Middle Gray,
Click to expand...

This makes no real sense to me because it contradicts middle gray being mapped to 11 and not 8 when in Gio view.
 
Mark Wuerthner said:
Alex you have misread the chart that you posted. Counting from 0 (where the chart says EV Scale) to the right this chart shows that Dragon clips (the flat purple line) at about +7.3, while the Alexa (yellow line) keeps moving to the right. Same thing on the underexposure side of the chart, where Dragon basically flat lines a little earlier than Alexa. What's interesting in this chart is where each camera places middle gray. Dragon puts mid gray point brightness at 1.3 stops darker than Alexa. As you move down stops each Dragon gray chip is exposed about 1.3 stops closer to the noise floor than Alexa. This chart seems to offer an explanation as to why there is noise associated with skin tone olpf when exposed at 800 ISO, as the gray chips under mid gray are mapped close to the noise floor. Exposing Dragon at 320 ISO would move all the gray patches up 1.3 stops away from the noise floor thus improving noise. Of course, by doing that you add 1.3 stop to the dark end of the scale and take away 1.3 stops of highlight latitude. In reference to the OP question, I would say that that dynamic range to do what he wants is there, but one has to expose correctly in order to do it. Trying to pull dark images out of a Dragon that has been exposed at 800 ISO is not the best way to do it. Shooting at 320 ISO is going to give one a lot better chance at bringing the darks up with less noise. Notice in the still camera article they shoot at 100 ISO, which probably gives their images lots of exposure in the darks thus allowing them to make the dark parts of their image brighter. Hope that helps.
Click to expand...

Hello Mark. Thanks for your answer. Like David said the alexa has more flat log/lut from the Epic MX/Dragon. That dose not mean that it has more or the same DR.
I have work with the alexa and i found out that the Epic MX has more DR in the shadows and less noise if it is properly exposed. In the DXOmark test http://www.dxomark.com/Reviews/RED-...rst-place-in-both-categories-14.8Ev-26.5-Bits we see that the DR of the Dragon is 14,8 stops. I think that the DR of the alexa is 14.
 
http://reduser.net/forum/showthread.php?117540-Dragon-Dynamic-Range-Test-21-stop-chart

As Stacey and I found, you can kind of arbitrarily pick any value below 19 and be "correct" from a strictly technical standpoint. If you say Dragon has "13 stops" then you really mean "It has 13 stops that are useful to me." Which is a perfectly reasonable statement. But you can't belittle someone who says they see 14 or 15--because it is there. Just exponentially less useful.

 
Gavin Greenwalt said:
http://reduser.net/forum/showthread.php?117540-Dragon-Dynamic-Range-Test-21-stop-chart

As Stacey and I found, you can kind of arbitrarily pick any value below 19 and be "correct" from a strictly technical standpoint. If you say Dragon has "13 stops" then you really mean "It has 13 stops that are useful to me." Which is a perfectly reasonable statement. But you can't belittle someone who says they see 14 or 15--because it is there. Just exponentially less useful.

Click to expand...

I would love to see this same image without the 6 stops of frame averaging.
 
Gavin Greenwalt said:
http://reduser.net/forum/showthread.php?117540-Dragon-Dynamic-Range-Test-21-stop-chart

As Stacey and I found, you can kind of arbitrarily pick any value below 19 and be "correct" from a strictly technical standpoint. If you say Dragon has "13 stops" then you really mean "It has 13 stops that are useful to me." Which is a perfectly reasonable statement. But you can't belittle someone who says they see 14 or 15--because it is there. Just exponentially less useful.

Click to expand...

I'm pretty sure you can push any full frame raw DSLR file to 16+ stops in post... it's just that they won't advertise with it as they know it's useless.
 
I'm still out on location shooting, so I can't send out new test images.

At this point with several people using Xyla-21s and individual patch reads landing on approximately 19 stops of measured dynamic range and have posted the samples here, I'm not entirely sure why this is still a discussion. However, if you want to call it 14, 15, or the more accurate 16+ that's fine.

If you want the "balanced" dynamic range distribution with equal stops above and below Middle Gray shoot between ISO 800-1280. That variance is mostly due to what people would consider is the last usable stop. For me ISO 1000 (or more those real picky around 1100) is that spot.
 
Dunkan Wolf said:
I'm pretty sure you can push any full frame raw DSLR file to 16+ stops in post... it's just that they won't advertise with it as they know it's useless.
Click to expand...

Seeing how you clearly believe Dragon has less than 16 stops of DR, I'd like to know...

how many stops do you think it has?

what test (to which you could perhaps provide a link) made you settle on that number?

why you believe that test over others that show Dragon having more DR?
 
Martin Stevens said:
Okay.... But, if RAW GIO view has the base ISO fixed at 800, then putting middle gray at 11 means LESS highlight latitude than if it were at 8.
Click to expand...

I was wondering the same thing about highlight DR and the 11th stop reference for middle grey.
 
You must log in or register to reply here.
Back
Top