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 :)

RED Weapon - Dynamic Range and Latitude Examined

As always, you've gone to great lengths to share a lot of technical findings and insight that helps us to understand what Red has brought to the market, and in the process, you help to elevate this community. Thanks, Phil.
 
We all love Phil Holland.... and this new post of his, is just one small reason why.

Bravo, Phil.
 
  • Thread starter
  • Moderator
  • #24
Blair S. Paulsen said:
Wow Phil. That's a ton of effort and great choices on the way you portrayed the results in the chart design.
Click to expand...

Thanks Blair. It's certainly nice to look at it all organized. There's a whole bunch of other stuff I tested out that results in horrific images, but this tells the story effectively. It's an interesting thing, but here it is in reality. 16.5+ Stops. It's an amoebic description, but pretty accurate to how the camera and system works.


A. Clint Litton said:
As always, you've gone to great lengths to share a lot of technical findings and insight that helps us to understand what Red has brought to the market, and in the process, you help to elevate this community. Thanks, Phil.
Click to expand...

Thanks Clint. REDuser is an important community and important to me as well. And I think we're just at the tip of Weapons being used on productions. This and my two previous tests answered a lot of my questions. I think it will reveal much to those who are asking the same things.

RED's breaking a bit of new ground right now. They essentially (and arguably) have the highest quality motion picture camera available (which is nice as current rental rates reflect that) while at the same time introducing an entry level model with Raven. I think what we learn about Weapon 6K will all translate into Weapon 8K. Sounds like there's something else on the horizon in between with a more Scarlet minded beast as well. I just had two of the best years of my career shooting on Epic Dragon all over the place. Some of those shots were really jaw dropping in terms of color and resolution. Now I know exactly what I can get out of Weapon 6K today and it's giving me a fair idea of what to expect in the near-ish future from Weapon 8K.
 
If the camera has 7 stops of highlight latitude above mid gray @250 ISO, There's rarely a need to "protect the highlights" by shooting at the recommended 800 ISO?
 
  • Thread starter
  • Moderator
  • #29
David Mullen ASC said:
Any place to rent a DCS Xyla-21 chart in Los Angeles?
Click to expand...

Couldn't find one, nearly bought one, likely co-designing a new one.


jussi rovanpera said:
If the camera has 7 stops of highlight latitude above mid gray @250 ISO, There's rarely a need to "protect the highlights" by shooting at the recommended 800 ISO?
Click to expand...

Depends on how and what you shoot, how and if you light, and how you choose to expose. It's pretty easy to clip even at ISO 800 with a not-so-bright source in frame, though most don't mind that. Overexposure isn't necessarily a bad thing, it's just a bad thing when it occurs in a distracting manor that draws too much attention or when it feels like the exposure decision has been made in a reckless sort of fashion.
 
I know on epic dragon the Gio-Scope puts 320 iso's middle grey at stop 11 and clips past stop 16. If 320 iso now has 7+ stops of room before clipping are we seeing another 1 or 2 stops of highlight protection on weapon? What's your estimate, Phil?
 
Great test!!! Thanks Phill!!!

One more that's always the same question:

What about wide color gamut with Weapon?

How far we can push or pull RGB that for example caucasian skin tones remain the same or accurate???

It's about a possibility of wide colour gamut that could during a grade process get rid of to much magenta or yellow tinged skin that can occur so often with all cine digital cameras that use CMOS sensors.
 
Very good work, Phil.
Thanks.
 
Thanks for these tests. Very useful info here.

I'm not trying to be a troll, just have an honest discussion. Though I'll probably be banned for playing devil's advocate here.

As I've been saying since dragon first came out, there are still no tests anywhere which show the dragon to have more DR than the ALEV III in the Alexa. It's neither here nor there to me. Just an objective observation. And certainly not an criticism of the effort that you put into, or my/the community's appreciation for these tests.

And people will always argue about size and resolution and color, etc...

But, if you were to do these exact same tests with an Alexa, or likely f65 as well, by all other available data, speaking of DR alone, you'd very likely get a wider range.

So, my point from the beginning has been, if we want to continue saying that Dragon is 16.5+ stops, then we should, in fairness and accuracy, do the exact same test objectively with the other major players and plot them on that same scale.

Or, conversely, use the same methodology that the other manufacturers used to reach the 14+ stops for Alexa and f65, and plot Dragon on that scale.

Because as it stands, saying that Dragon, in either Epic or Weapon, has 2.5 stops more DR than Alexa or f65 is simply not true. And is very misleading.

And again, this isn't a discussion about any of the other merits or features of any of these cameras. Just DR. And the inconsistency of the methodology used to rate it. And how it gets labeled by manufacturers.

In the end it doesn't really matter either way. Most of us know how to test a camera to figure out how it can, or if it can, suit our needs. I just have a problem, I suppose from an ethical standpoint, of seeing cameras that clearly have more DR being labeled as less capable for marketing purposes. It just doesn't sit right with me. Just my personal feelings.

And also, again, this is not a criticism at all of your very thorough tests or to lessen the value of what you've provided here. So, above all, thank you for the work that you have consistently put into helping this community.
 
Paris Remillard said:
Or, conversely, use the same methodology that the other manufacturers used to reach the 14+ stops for Alexa and f65, and plot Dragon on that scale.
Click to expand...

The issue with this is that no one publishes their methodology for measuring dynamic range and there is no agreed upon industry standard. Geoff Boyle does a nice job of using the same methodology for all of the cameras he tests.
 
  • Thread starter
  • Moderator
  • #40
Sanjin Jukic said:
What about wide color gamut with Weapon?

How far we can push or pull (flexibility) RGB in R3D (RAW) that for example caucasian skin tones remain the same or accurate???
Click to expand...

I guess I would need to test that to be precise. It will also likely depend on what your rated ISO was as well if you think about it.


Stacey Spears said:
The issue with this is that no one publishes their methodology for measuring dynamic range and there is no agreed upon industry standard. Geoff Boyle does a nice job of using the same methodology for all of the cameras he tests.
Click to expand...

This is true. But I can say two things. Most manufacturers and companies count Total Captured Stops. This is reflected in their numbers. I can say that for over 12 years the "patch method" I use is what a few 4 letter companies use as well as two celluloid companies have used. It's the most time consuming process, we usually dedicated a full day for each "thing" to test, but it makes measuring tonal variation and density accurately fairly bullet proof. The Xyla-21 is indeed nice and allows for contamination elimination via the shutters to occur and that's truly needed when testing the chart. You can essentially transform it into a single patch readout which is nice too. I would like to have two changes to a future chart to assist with these types of measurements. Hopefully DSC can accommodate that.
 
You must log in or register to reply here.
Back
Top