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 :)

Matching an 8k Monstro with a 6k Epic Dragon?

Marcus Friedlander

Marcus Friedlander

Well-known member
Joined
Jun 30, 2014
Messages
1,124
Reaction score
2
Points
38
Location
Los Angeles
Hello everyone,

For an upcoming project, the director and I have decided that an 8k Monstro is the right tool for the job because of the physical sensor size, and the R3D workflow. Also, we are shooting a lot of night exteriors, so the insanse sensitivity is a huge plus because it allows me to drop my M40 and tow generator, and just run with an M18 and smaller genie as my biggest HMI.

However, as we've gotten deeper into prep, it's become painfully obvious that we really need to be shooting two cameras for most of the show.

We have the budget to bring in the necassary crew to run a second camera, but that already starts eating into our contingency. Also, raising the budget overall is tricky because of the wording of certain contracts that are already in place. So, while it is technically possible, it's definitely an absolute last resort.

So instead of renting a second Monstro, I'm looking into the possibility of using my Epic Dragon as a B camera.

Both cameras will be using the same OLPF'S, lenses, and filters. And the final look we are going for will be contrasty, and also have film grain laid over it, so I'm not too worried about the Dragon not being able to match the color aspects of the look of the Monstro (especially with IPP2 and/or ACES). But I am worried about the sensitivity and sensor size differences.

For sensitivity, this is how I'm thinking about rating the two cameras. (Our Master will be a 4k DCP. And we will be grading in resolve studio, so I'll have access to NR, but I really don't want to rely on it for every Epic Dragon shot unless I absolutely have to.)

Day EXT -
Monstro - T5.6 @ 800 w/STH
Dragon - T4 @ 400 w/STH

Day INT -
M T4 - 1280 w/STH
D T2.8 - 640 w/STH

Night EXT -
M T2.8 - 2000 w/STD
D T2 - 1000 w/STD

Night INT -
M T2.8 - 1600 w/STD
D T2 - 800 w/STD



As for sensor size differences, I plan to use the Monstro sensor for the wides and mediums and use the Epic Dragon for the close ups and inserts. I don't ever plan to shoot matching frame cross-coverage, because I think that'll be a nightmare in the edit and color grading suites. So the B camera will generally be used to grab reactions and cutaway moments, always complimenting the coverage the A camera is capturing, but never intending to exactly match.

Does anyone have any thoughts or advice on the matter? Would love to get the community's opinion!

Thanks!
 
Last edited:
in certain situations, like a dialogue two camera scene or such, ("matching frame cross coverage") you can window down the Monster to 6K and you will have essentially the same camera as the 6K dragon. They would then look pretty similar. Of course the Monstro will look great in more grand shots in 8K.
 
you can make the match exactly optically, perspective, dof wise. color balance /hue may not (actually I stand corrected hue should match as it is from the same wafer fab, processing notwithstanding). Sensitivity seems pretty close according to the tests.
Here is the sensor matching article link:
http://www.yedlin.net/170504.html
 
Mark Andersen said:
in certain situations, like a dialogue two camera scene or such, ("matching frame cross coverage") you can window down the Monster to 6K and you will have essentially the same camera as the 6K dragon. They would then look pretty similar. Of course the Monstro will look great in more grand shots in 8K.
Click to expand...

Yeah I know I can window down to match sizes, but I think it's pointless to pay the premium for the Monstro and lens package, but then not use the full sensor. If it came down to that I would just rent a 6k Weapon Dragon.

(Edit: Sorry!!! I just reread this and realized that this sounds way harsher than what I was intending!! My apologies Mark for coming off like an asshole!!! Totally not my intention!!! And thank you for taking the time to reply!)
 
Last edited:
I might end up being in a similar situation as yours.

I've been toying with the idea of using Monstro for wides (fior a more rectilinear perspective) and mid/tight coverage with S35 (Helium in my case). In your case, the 8K on a wide would also benefit from greater detail from the added resolution (tighter shots tend to have a shallower depth of field and therefore less detail in background).

Regarding sensitivity, I question the added 2 stops you're aiming for (I mean, test this but seems a bit much). I'd focus more on redcode compression. Mark from OffHollywood has a great blog post regarding redcode vs resolution and how they affect detail. I think it was something like 8K 10:1 = 6K 5:1.

And finally, I would recommend the LLO for the Epic Dragon vs the STD.
 
Helium is a different sensor family so you may have issues. But I'm sure somebody will come up with a "monstro" LUT for the Helium - Phil?!!;-)
 
I am confidant that with your strategy ...

As for sensor size differences, I plan to use the Monstro sensor for the wides and mediums and use the Epic Dragon for the close ups and inserts.
Click to expand...

...will make it a work.

I have on a couple of shoots shot my wide shots wide open and my MCU and CU work 1-2 stops down... net effect is similar.
 
Simran Dewan said:
Regarding sensitivity, I question the added 2 stops you're aiming for (I mean, test this but seems a bit much).
Click to expand...

I haven't had a chance to test yet, but theoretically, this is my calculation of the sensitivity difference between a 6k Epic Dragon and an 8k Monstro

Going from Dragon pixels to Monstro pixels = 2/3 of a stop
Going from DSMC1 to DSMC2 = 1/2 stop
Additional downscaling advantage of 8k over 6k = 1/2 stop

So that together equals 1 2/3 stops. Not quite the full 2 stops I know, but I have a feeling, based on the insanse noise floor tests people have been posting, that it's closer to 2 stops, rather than 1 2/3.

Simran Dewan said:
I would recommend the LLO for the Epic Dragon vs the STD.
Click to expand...

I was thinking about going LLO but the dragon may have issues with color matching as is, so I want to give it the most help possible. And the exposure difference between the Standard and LLO isn't worth it in my opinion.


And as far as red code compression goes, that's exactly what I was planning on shooting at. (Edit: 8k @ 10:1 and 6k @ 5:1
 
Phil Holland said:
If you do use the 6K FF format, note there is a resolution/slight format size difference between Weapon 6K S35 Dragon and Monstro 8K VV:
http://phfx.com/tools/formatCompare/formatCompare.cgi?formatOne_typeTech=RED%7CMonstro+8K+VV&formatOne_format=6K+FF&formatTwo_typeTech=RED%7CDragon+6K+S35&formatTwo_format=6K+FF&focalLengths=

Your friendly neighborhood Monstro man.
Click to expand...

Thanks for the heads up Phil! Appreciate it! I was planning on shooting a 8k 2:1 raster (with a 7k WS extraction) and a 6k 2:1 raster (with a 5k WS extraction. So I think I should be ok there.

But thanks again for your input!
 
Michael Lindsay said:
I am confidant that with your strategy ...



...will make it a work.

I have on a couple of shoots shot my wide shots wide open and my MCU and CU work 1-2 stops down... net effect is similar.
Click to expand...

Awesome! Glad to hear I'm on the right track! Thanks for your input Michael!
 
I’m going to give you a differing perspective. You’re free to borrow from it the parts you like and discard what you don’t.

By artificially compensating your ISO and stop values you’re creating an extra headache for yourself on set. Massively.

I’m someone who would recommend to never rate a red camera below 800asa in broad daylight (sure there are exceptions) - but rarely if ever. You’re going to clip your highlights faster and reduce your captured dynamic range if you’re light metering and ignoring the red exposure tools.

I would simply rate both cameras at 800 in broad daylight, rate them LOWER in night exteriors like 400-500, night interiors it’s up to how high key or low key your scene but anywhere from 320-800.

Sure you CAN deviate from these settings but I don’t recommend making it standard practice.

I would also keep your Standard OLPF on both cameras at all times, not swapping the olpf during the shoot unless you find you’re in a lighting emergency of some sort.

Focus on making a beautiful image with the aid of technology , rather than focusing on a technical image that might possibly support a story.

The last one is a bit overreaching as advice - I don’t know if you have shot a lot of feature length films , but you may find the level of complications outside of compensating exposure across two cameras demands so much of your attention that by reducing the technical complexity you have opened up your ability to support the story, the director, the actors and the sheer number of surprises that will be coming down the pipeline daily. Have a great shoot !
 
Hey Dan! Thank you very much for your detailed response!

Dan Kanes said:
I’m going to give you a differing perspective. You’re free to borrow from it the parts you like and discard what you don’t.

By artificially compensating your ISO and stop values you’re creating an extra headache for yourself on set. Massively.
Click to expand...

Since the Epic Dragon and VV Monstro have vastly differing sensitivities, isn't this a good way to create a balance between the two cameras? If I rated them the same, the dragon would be roughly a stop and a half nosier at every exposure, (and, if that isn't the case, then what was everyone complaining about when the Dragon sensor first came out haha.)

Dan Kanes said:
I’m someone who would recommend to never rate a red camera below 800asa in broad daylight (sure there are exceptions) - but rarely if ever. You’re going to clip your highlights faster and reduce your captured dynamic range if you’re light metering and ignoring the red exposure tools. I would simply rate both cameras at 800 in broad daylight, rate them LOWER in night exteriors like 400-500, night interiors it’s up to how high key or low key your scene but anywhere from 320-800.

Sure you CAN deviate from these settings but I don’t recommend making it standard practice.
Click to expand...

When I first started shooting with Red's I remember reading that this was the prevailing logic. However, after shooting with reds for 5+ years now, I've never really seen this earlier clipping problem ever play out in my negatives. When I've got a ton of light, like a day EXT, dropping from 800 to 400 has provided me with a cleaner negative, and, so long as I watch my goalposts, stoplights, and histogram, (which I never ignore!) I've never noticed a clipping problem with this method. Especially once the dragon sensor came out with its awesome rolloff. (Unless I point the camera at the sun lol)

After thinking about it for a bit, maybe I'm not pushing far enough for the early clipping effects to show? I'm only dropping down 1 stop, but if you dropped from 800 to 50, I could see it more obviously messing things up. But a 1 stop change in a 13-14 stop camera isn't the end of the world.

Due to all the above, I've never really gelled with this concept the way everyone else has, but hopefully you can open my eyes a bit!

The core thing that always confused me is that the mantra around here is that ISO is a digital gain, and not an analogue gain, BUT, somehow that digital gain that happens behind the sensor, can can cause the sensor itself to clip? That seems like a contradiction, no?

If you aren't clipping at 800, then you shouldn't be clipping at 200 or 3200 either because it's just metadata right? The actual pixel wells are still capable of holding the same amount of photons no matter what is going on behind the sensor in the processing stage?

So as long as you don't incorrectly change your lighting or exposure to cause clipping, then you'll be fine. (Or at least I have been so far hahah.)

But if I'm doing this all wrong because I'm misunderstanding a key concept I would love to be corrected!!!! Hahaha

Dan Kanes said:
I would also keep your Standard OLPF on both cameras at all times, not swapping the olpf during the shoot unless you find you’re in a lighting emergency of some sort.
Click to expand...

Im going to respectfully disagree with you on this one. I believe swappable OLPFs are one of the biggest advantages of the RED ecosystem. Obviously for its exposure characteristics, but also for its optical characteristics as well. While a good colorist can seemingly make anything look like anything these days hahaha, I still find a lot of benefit in using different OLPF'S to help create a base difference between varying sections of a film (which for this project is exactly what I'm looking for.)

Dan Kanes said:
Focus on making a beautiful image with the aid of technology , rather than focusing on a technical image that might possibly support a story.
Click to expand...

Ironically, even though I'm talking about a super high end camera, this direction I'm researching is actually not about making a more beautiful image; it's entirely about better telling the great story that's already in the script. On top of that, this isn't a film that's supposed to be beautiful, it's a film that's supposed to be effective. And more importantly, it's a film that will live or die by the strength of its performances, and not necessarily by the beauty of the images.

So, I'm focusing on what I can do on my end to best facilitate and capture great performances, while still creating the imagery that will be right for the film. (And if we do our jobs right, give our editor the footage he needs to cut together a great movie :) )

Dan Kanes said:
The last one is a bit overreaching as advice - I don’t know if you have shot a lot of feature length films , but you may find the level of complications outside of compensating exposure across two cameras demands so much of your attention that by reducing the technical complexity you have opened up your ability to support the story, the director, the actors and the sheer number of surprises that will be coming down the pipeline daily. Have a great shoot !
Click to expand...

Thank you for this advice as well. It's definitely something I took into serious consideration before pitching this idea to the rest of production.

I can't go into detail unfortunately, but, luckily, this project has several key elements that'll make the complexities you mentioned above, significantly more manageable.

As far as experience goes, I'm definitely not Janusz Kaminski hahahaha. But, this will be my 6th film as a DP, and my company's 8th that we'll have produced, so I feel confident in my abilities to handle the extra complexities. I generally find that things that make me nervous, or even scared, are the exact things I need to be doing with my life at that moment in time hahaha. I had the same feeling when we started our production company, and when we started pre-pro on our first feature, etc. etc. hahaha. This is just another one on the list :)



Thanks again for your lengthy response Dan! Really appreciate it!
 
The idea is to rate the camera at a different ISO, not simply change the setting. If you were rating 400 and clipped, changing the rating to 800 will mean adding a .3 ND or stopping down, thereby reducing how much light is actually hitting the sensor.

The tonality of a daylight scene typically is above the noise floor even at 800 so this won't result in too much noise, likewise the tonality of a night interior is probably mostly below middle grey so you want to rate lower ISO so that you expose above the noise floor.
 
I would definitly use the LLO on the Dragon (and use a Dsmc2 body if possible) as the noise characteristic is much nicer when shot at 1600iso as with the STD. As you want some noise/grain in the final picture you'll have something nice already in camera. So you can match better the high ISO performance from the Monstro when you shoot yours night exteriors.

Pat

PS : My base Iso for Weapon MG Dragon on LLO is 1600. I use 800 when I want something terribly clean.
 
Anton Shavlik said:
The idea is to rate the camera at a different ISO, not simply change the setting. If you were rating 400 and clipped, changing the rating to 800 will mean adding a .3 ND or stopping down, thereby reducing how much light is actually hitting the sensor.
Click to expand...

And this is where I get confused. Why do I have to change a digital gain ISO on camera to add an analogue ND, or stop down, in order to save something from clipping? Why can't you just stop down while staying at 400???

I get that my monitor will look darker, but the alternative is to artificially brighten up the image with a digital gain in camera, when it can be more accurately done in a color grading suite, and both operations will increases the noise..... So it seems like a wash to me???

As long as you haven't clipped anything on the day, there is proper light on the subject you are trying to see, and your overall lighting ratios are what you want, then this is really not an issue I've been able to replicate in the real world. In other words, if I'm even remotely competent as a DP, this should almost never be an issue even worth thinking about hahahaha.
 
You must log in or register to reply here.
Back
Top