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 :)

All the Epic/Dragon Vs. Alexa/Sony threads are belong to us...

Is it wrong to say that this is a result of where you set mid-gray/ISO during capture (like it was on MX)? e.g. Had Dragon been set to ISO2000, its images would have had a more even over/under exposure, and hence more comparative to Alexa (which has a base of approx +7/-7 at ISO800)?
 
Phil Holland said:
See the above image and you make the call.
Click to expand...

I don't see anything in that image other than one is sharper and white balanced. You think there's more detail in blacks and whites in the top?

Mike P. said:
Is it wrong to say that this is a result of where you set mid-gray/ISO during capture (like it was on MX)? e.g. Had Dragon been set to ISO2000, its images would have had a more even over/under exposure, and hence more comparative to Alexa (which has a base of approx +7/-7 at ISO800)?
Click to expand...

This seems like a very attenuated attempt of a cop-out. At this point I'm not even sure I understand what anyone is trying to claim and what their reasonings are behind it. Though Phil has tried to explain it multiple times now, I do not believe I'm the only one still very much confused about this mythical reasoning behind 16.5 stops.
 
Phil - Regardless of what the base exposure of a particular camera looks like in terms of its exposure level, it is relevant to itself. So if We count the stops above and below that for each camera and then compare, thats all perfectly fair. I do not see 16stops in the R3Ds. The Alexa has more in total in this test. I'm clearly not the only one to think so. How many do you count in the CML test?
 
James, it's not a cop-out. I'm saying all the cameras were set to ISO800, and that might not work for displaying all of Dragon's "16" stops... (too much in the low end and/or makes it appear to have less stops in the high end).

What's confusing is that, at base exposure, you'd expect Dragon to have +8/-8, and it apparently does not. More over, having to process Dragon footage to show a latitude advantage over Alexa will automatically cram Dragon into the same boat that MX was in (the one where you need to use finishing that is above/beyond the norm/standard.) A lot of Alexa's strength is that it mimics film fairly well right out of the camera; ArriRAW/2.88k is merely used to adjust/tweak the image further. With RED/MX, you kind of need to process/leverage it's raw to get it closer to film... Don't get me wrong, a lot of people like that control, but I understand when directors/DoPs just want what looks best the fastest, and when most producers just want what gets their post to the finish line cheapest.
 
James_Mills said:
I don't see anything in that image other than one is sharper and white balanced. You think there's more detail in blacks and whites in the top?
Click to expand...

Both are white balanced and in theory they should be "identical", however they are not. Read the values and more importantly the separation of tones between those two images.
 
Phil Holland said:
Both are white balanced and in theory they should be "identical", however they are not. Read the values and more importantly the separation of tones between those two images.
Click to expand...

I'm getting it. Tonal value, clear separation between those values = color separation. Combine that with more resolution and you have some very sexy pictures.


Battistella
 
Mike P. said:
Is it wrong to say that this is a result of where you set mid-gray/ISO during capture (like it was on MX)? e.g. Had Dragon been set to ISO2000, its images would have had a more even over/under exposure, and hence more comparative to Alexa (which has a base of approx +7/-7 at ISO800)?
Click to expand...



It makes no difference. The native dynamic range of the DRAGON sensor is static (it has a hard end on the top and bottom). Choosing different "development ISOs" only serves to shift where 18% Grey is mapped within that range. You won't "find" any extra DR solely by changing development ISO.
 
Mike P. said:
James, it's not a cop-out. I'm saying all the cameras were set to ISO800, and that might not work for displaying all of Dragon's "16" stops... (too much in the low end and/or makes it appear to have less stops in the high end).
Click to expand...

David Battistella said:
The base ISO conversation has been kinda done to death here.

http://www.reduser.net/forum/showth...g-Native-ISO&p=1326373&viewfull=1#post1326373

In this thread where the consensus seems to be to call the base ISO a RANGE and not set a specific value as a baseline. Let's not go there in this thread but that thead is a good read if you have the time.

Battistella
Click to expand...

I'm aware of what is being stated here, but are we supposed to then go around shooting natively at 2,000 ASA and expect other cameras to compete at 2000 also in shootouts? And do you guys really think that if we did there would be a clear 2.3 stops of distinction between the Alexa and the Dragon? 2.3 stops of dynamic range is HUGE. That's worse than the difference between an Alexa and a 5D. I cannot begin to imagine changing my ISO to 2,000 and suddenly I'm shooting with a whole new camera with 3.3 more stops of dynamic range (since it would have to account for the stop that Alexa already has on it in these tests).

From what I remember there was never a definitive answer from RED about what would be ASA throughout the year leading up to the Dragon's release. I seem to remember it being stated that there was no native ISO in terms of a target dynamic range, but perhaps I'm wrong. Adrian seems to agree though:

Adrian Jebef said:
It makes no difference. The native dynamic range of the DRAGON sensor is static (it has a hard end on the top and bottom). Choosing different "development ISOs" only serves to shift where 18% Grey is mapped within that range. You won't "find" any extra DR solely by changing development ISO.
Click to expand...

A whole lot of confusion going on. I'd like to hear others' professional opinions about Phil's pictures and whether or not a clear distinction in latitude can be made (although I'm not even positive that Phil's argument is the whole 2000 ISO route; I think he's stating something too scientific for me and perhaps a lot of us in here). So far the only thing concrete is this test that clearly shows the Dragon (with its old OLPF) to have less than 14 stops of dynamic range.

It would be great to have another objective outsider comment on this. It's unfortunate that Geoff cannot elaborate without the threat of expulsion again.
 
Phil's argument isn't the whole 2000 ISO route :)

I'm doing my best to avoid getting into camera response curves and looking at these a different way.

The best I can do from here is test the desired cameras and produce a controlled experiment to showcase and measure the total dynamic range. I'll have to leave it at that. There is a difference between the "differences seen here" and the "measured number" people are after. That's essentially what I'm trying to convey.

One piece of advice I can give is push and pull the CML UWE files around as much as possible and carefully study the results. This will reveal the latitude and possibly the dynamic range differences, however, it's not the right way to measure them precisely.

We are getting a bit off the track of the test though. The results are valid in comparing the systems with the technology available at the time. It's good to look a bit under the covers though.
 
I'll just leave this here... DSC labs make a multi-thousand dollar chart for a good reason.

xyla21dragonannotated.jpg


Side note: I didn't share this at the time, as the camera was pre-release and undergoing significant change and improvement. I'll have my new OLPF setup in hand very shortly, and will re-test of course.
 
Ok I feel better now. I've just had a play with the CML footage again. If you play around with the ISO like Gunleik said then I get 5 over and 8 under, which was the limit of the test. If you then drop the ISO you can see separation for another 3 stops. 5 over and 11 under seems a pretty mad balance so I guess that's why there's a new calibration.
 
John Marchant said:
I'll just leave this here... DSC labs make a multi-thousand dollar chart for a good reason.

xyla21dragonannotated.jpg


Side note: I didn't share this at the time, as the camera was pre-release and undergoing significant change and improvement. I'll have my new OLPF setup in hand very shortly, and will re-test of course.
Click to expand...

John,

I use RED approved DSC labs charts as well.

Can you share the lens stop and ISO setting of the camera when that was captured?

Battistella
 
Slate says T4 1/2 on a 50mm UP, 5KHD, RC 3:1

Posted RLF, ISO 250

I have even more convincing ones not at hand on my laptop - this shot was done without leveraging the whole sensor area (this is actually quite a crop from the 5k, so a little grainy). The later tests I did, I built a proper light-tight hood for the Xyla-21, rather than just using a darkened room, and shot a variety of glass and f-stops, filling the 6k width with the chart. I curtailed these further test though when last year it became apparent that more significant changes were coming, and I didn't want to blow a significant effort testing something which would shortly be redundant ;)
 
I guess what I'm saying is if we have 16.3 stops, I'd like for it to be apparent at all times of capture if I am giving the shot that much range. I shouldn't have to dig it out in post, it should be there in the RC3,RG3 shot. Does that make sense?

The Alexa is chosen because it's 14 stops are apparent at all times and do not require subsequent digging and post ISO adjustments. I would like to see the 16.3 stops at all times at every ISO as long as I expose it to leverage the 16.3 stops properly. I want to see the 2 stops improvement vs. Alexa in a side by side instantly without having to play around in post. I think that's what most of us want to see...



Edit: Don't get me wrong, I'm a fanboy who recommends Epic for a million reasons. I just want it to be more obvious the improvement vs. Alexa otherwise people will not listen to these 16.3 stops DR claim. I don't want to have to tell people they need to do more work in post to dig out those 16.3 stops...
 
You must log in or register to reply here.
Back
Top