G
Guest
Guest
Last setup i used pre lockdown was master wheels, R2, and Fx9 with a prime. Inverted setup on a dolly. We left the ring on, but didn’t need to.
Very responsive setup.
Canon R5 Announced with 8k
- Thread starter Ken1212
- Start date
Last setup i used pre lockdown was master wheels, R2, and Fx9 with a prime. Inverted setup on a dolly. We left the ring on, but didn’t need to.
Very responsive setup.
ahalpert
Major Contributor
If you're interested, we started a new thread to focus just on operating aesthetics-
http://www.dvxuser.com/V6/showthread.php?367218-Operating-Aesthetics
John Brawley
Well-known member
larger payloads mainly, higher degree of stabilization and better camera integration (with an Arri camera of course).
I spoke to one of the the main car / pursuit camera operators on Ford Vs Ferrari at the bSC show and he showed me as lot of their rigs and the SRH3 was outperforming way more expensive and larger heads.
JB
G
Guest
Guest
Although, the setup where the guy wanted a 180 degree turn around was Alexa Mini LF anamorphic lens and R2. A bit heavy.
G
Guest
Guest
Thanks
ahalpert
Major Contributor
Thinking about this comment -
If you shoot 8K 10bit and finish in 4K, won't you produce a 4K 12bit image to work with? And with higher color sampling frequency as well?
G
Guest
Guest
You may be thinking of chroma super sampling.
I think bit depth is in the gradation and steps and shades. You aren’t really added steps when you down res. I guess you could blur or dither in a 12bit space and then down res, and sort of fake it. Haha.
However, if you have a 10bit 8K 4:2:2 image, you could make a 10bit 4K 4:4:4 image. Or something like that.
Ryan Farnes
Well-known member
What is everyone's best guess on dynamic range of the R5?
I've been really drawn to the BMPCC6k. The S1H didn't impress me as much with highlights. I haven't spent as much time reviewing the A7RIV or other Sony mirrorless options. I was hoping to see something announced about the A7SIII today but alas, no. Sony Alpha Rumors predicts a June announcement.
I've owned a few Canon cameras, but was getting ready to jump to Sony or Blackmagic. I like the Blackmagic images, but the concept of high quality stills and motion in the same body is compelling to me so I was thinking Sony was probably going to be the route.
Then the R5 specs were confirmed. The R5 seems like it will be an absolute beast. Dream camera. The 30fps full sensor RAW motion. Plus all the other goodies...
There may be some other small details, but the latitude is the *big* thing that I'm wondering about.
I've been really drawn to the BMPCC6k. The S1H didn't impress me as much with highlights. I haven't spent as much time reviewing the A7RIV or other Sony mirrorless options. I was hoping to see something announced about the A7SIII today but alas, no. Sony Alpha Rumors predicts a June announcement.
I've owned a few Canon cameras, but was getting ready to jump to Sony or Blackmagic. I like the Blackmagic images, but the concept of high quality stills and motion in the same body is compelling to me so I was thinking Sony was probably going to be the route.
Then the R5 specs were confirmed. The R5 seems like it will be an absolute beast. Dream camera. The 30fps full sensor RAW motion. Plus all the other goodies...
There may be some other small details, but the latitude is the *big* thing that I'm wondering about.
G
Guest
Guest
Did you try the recent prores raw update on the S1H?
The EVF on the little panny is pretty great. Best evf on a mirrorless i’ve tried so far. But, yeah, the pocket is $2K. If i were an ef user, that would be hard to pass up. However, the S1H has that Sony sensor, so it is likely to have better dynamic range, especially in raw.
John Brawley
Well-known member
No !
All kinds of no.
This came up in another thread. There's this ongoing myth about sensor resolution and "colour" resolution.
It leads to statements like this. And I don't get it.
I think people confuse the ratio of RGB photosites on a sensor with the way video encoding gets described like with 444 / 422 / 420 etc and assume they're the directly related.
They aren't.
Bit depth is directly related to dynamic range. With a higher bit depth you can store a great number of values OR a great discretion of values.....let's call it all the fractional values within that tone curve
Bit depth is the very first thing I want to know about a camera because to me everything else goes from there.
The only rough rules are if they are LOG or LIN encoded, then....
12 bit lin = 10 bit log
12 bit log = 16 bit lin
And I say roughly....
I know that Arri process off their sensor at greater than 16 bit lin, but the pipeline internally dithers down to 16 bit lin and then stored as a 12 bit LOG file in it's RAW form. A 12bit log file is more or less the same as the 16 bit file.
BMD do something very similar. Greater than 16b processing before the encoding phase, then stored as 12bit LOG in the file (Raw or prores)
You can't turn a 10 bit file into a 12 bit file (either log or lin) because you have more resolution. It just doesn't work that way.
You might get an improvement in noise, and therefore an apparent improvement in DR (because it's based on the noise floor)
But you don't all of a sudden get better dynamic range or better tonal / colour discretion.
Supersampling means you can improve things a little, but I don't think it's anything very meaningful and certainly not what the mathematical difference of 10 or 12 bit would suggest.
High bit depth images allow you to store more dynamic range OR allows you much finer discretion on tones and colours if you're grading. Now you can't just "create" values out of nothing by supersampling and using higher resolutions. It will improve a little, but not radically. A difficult key is still going to be a difficult key.
It's not like you can do an average key in 8K 10 bit and then it's behaving like a 12 bit acquired file in 4K. I've never seen that in evidence in pictures. Less bit depth means less wriggle room in the grade.
If you can nail the exposure and WB and it's squeezing all the DR of the sensor into the 10 bits you have available to record then GREAT ! But if you're trying to push that grade hard, a 10 bit file will fall apart faster, it won't take as much grading abuse. Usually seen in the obvious ways like banding, but in my view more critically, in trying to balance colour. Imagine a face that's lit by two different colour temperature lights.
In a lower bit depth grading environment NO MATTER WHAT RESOLUTION you have, the lower bit depth file with blotch and block faster in any subtle hue or secondary keys you're trying to do to improve the skin tone. The blocks might get smaller and not as noticable on the lower resolution output, but the same limitations are there.
Therefore, a 4K shooting 12 bit log camera like the pocket 4K or even the OG pocket with it's dual gain sensor and "only" 1920 sensor but still recording 12 bit LOG is better than an 8K camera like this.
Obvious resolution differences aside....
That's just because I've come to consider the bit depth to be one of the most important specs of a camera. Of course if there's a shot that can only be done using this camera then you use it. But there are plenty of small camera options with larger than 10 bit recording that I'd choose before this.
JB
Last edited:
Ryan Farnes
Well-known member
Hehe, yeah. Funny how those values are often inflated.
I'm not shooting these, just reviewing available samples online. Perhaps the S1H is underrepresented with competent shooters/graders, but skies rarely retain detail in the stuff I've seen. If there is detail, it has an inorganic blend into the digital mush of pure burned white...from what I've seen. Could also be the camera itself just has more of an electronic vibe to the image that I don't like.
I see a lot of Blackmagic 6K footage that has delightful detail retention in skies in full sun. It simply looks more like actual film stock with that natural rolloff with highlights to me. The 6K does do some weird stuff when pointing straight at the sun though, so it's not without its faults, but the image just feels more like an Arri or Red type of image. The S1H feels like a super high end Japanese video camera. I dunno, but my initial impressions of the latitude left me wanting more. I own a GH2 still, and think the world of Panasonic for making amazing DSLR form factor video cameras.
The R5 simply checks too many boxes, especially as I've gotten into shooting more and more still photography. If the latitude is there, I can't imagine any other camera wooing me, but if it's mediocre, despite all the other capabilities, I may pick up a 6K and a used Canon body like a 5Div and call it good.
G
Guest
Guest
Yeah, I think BM does a good job of getting the film maker sensibilities straight from camera. A bit strong on the "skin tone", but I like the overall tone curve and how that represents 3 dimensional space. The skin tones have some grit and texture. The colour is the current ideal too. Bias towards orang/teal. However the greens are almost yellow. Which, actually helps simplify the palette. The highlights at the extreme break up though. Get that yellow fringing. But better than all the compressed mirrorless stuff in general. I think the 1DC was great in how the channels clipped gracefully. It clipped nicer than the C500.
I know what you mean about the S1H footage. I don't really like the in camera look it has. Now that it has raw prores, I am going to give it a second look, as I really like what they did with it. If the HDMI were in a better spot, I'd likely buy one immediately. However, no telling if anything like that will ever happen, as i isn't likely people are going to think about ergonomics when they are distracted with raw and K's, and don't want to spend more than $3,500. I think the panasonic has the HDMI really close to where the meat of the left hand goes. -.-
The irony of the the two cameras is this. The Panasonic has an excellent EVF and a swivel screen, but requires an external ATMOS recorder... The BM Pocket does Braw internal, but requires an external EVF or Screen to use outdoors. So, either way, they are going to be a built out thing at some point. Kind of ruins the compactness. Oh well. 10 more years of reinventing the wheel?
ahalpert
Major Contributor
That is very interesting. I must have been misinformed - I had read in numerous places that you could improve color fidelity by at least having a blue and a red receptor inside that 2x2 patch that down samples to 1 pixel instead of a single green pixel.
As far as increasing bit depth - I had read that 4 different 10-bit values in a 2x2 patch could average out to a more accurate intermediate value in 12-bit than if they were a single pixel encoded in 10-bit. Now that I think about it, if all 4 pixels are the same value, they will all get rounded to the nearest value in 10-bit encoding and the outcome will be the same as if they were all 1 pixel encoded in 10-bit. So, there may be an improvement in tonal discretion in random places by a random degree?
In any case, you've basically convinced me that if/when I ever shoot a micro budget feature, a full-frame pocket 6K would be awesome. The pocket because the bit depth alone may account for the prettier colors I see coming out of it. 6K or 4K but ideally not higher so I can maximize sensitivity (and perhaps dynamic range?) And full-frame both to line up ideally with the stills focal length ranges I have, but also for maximum sensitivity. When I was an AC, i worked under many DPs on indy shoots on REDs and Arris who set a limit for their maximum usable ISO (usually 1600) and then broke their own rule and pushed it higher so they wouldn't have to change their lighting set-up (like when they lit a beach at night with a single M18 through a silk, or an interior with a heavily diffused Diva next to a large reflector). The takeaway for me was that if you have an inadequate lighting budget, you're better off having a higher usable ISO.
John Brawley
Well-known member
I'm not saying there wouldn't be a difference. But in terms of real world results it's sure as hell not anywhere close to what the actual difference between 12bit and 10 bit files is.
Only you can make those calls. You can also test for yourself. You shouldn't believe me or what you read on the internet. You should question what anyone says and find out what works for you in your shooting circumstances ! Maybe the physical way you work with a camera might suck. So who cares then what the bit dept or resolution is !
If you make a better film because the R5 allows you to get shots you can't get any other way then of course that's the best camera. You have to develop your own critical thinking approach to all this information and spec porn.
I can only go by my own testing and having gone though a few shows. Bit depth has been the thing that has helped me the most. Resolution not so much. There's never really right or wrong in all of this. Just try it and see for yourself !
JB
ahalpert
Major Contributor
Believe me - I trust you as far as i could throw you
. But if something makes sense to me, I listen.Do you have an opinion on what we know of the upcoming Komodo vs the Pocket 6K? Komodo is a touch more expensive than I'd like to buy for something that probably won't increase my paid work. But I could put a speed booster on it and get back the edges of my FF lenses and that extra stop. I'm a little concerned if the near-global shutter will reduce DR. All the global shutter cameras I've seen have had reduced DR. Thoughts on that + codecs/IQ?
G
Guest
Guest
The GS on the Komodo is anticipated to reduce DR, but it can be turned off when you don't need GS. RS is fine for most things. It is a bit funny to think of shooting in a bus or car, and wanting the extra latitude, but also wanting straight lines in the background. haha. But, the camera is likely to have somewhere around 13.5 stops of "usable" DR, so it has some to spare. But again, the global shutter feature can be turned on or off as needed. Not recommending Red here though.Believe me - I trust you as far as i could throw you
Do you have an opinion on what we know of the upcoming Komodo vs the Pocket 6K? Komodo is a touch more expensive than I'd like to buy for something that probably won't increase my paid work. But I could put a speed booster on it and get back the edges of my FF lenses and that extra stop. I'm a little concerned if the near-global shutter will reduce DR. All the global shutter cameras I've seen have had reduced DR. Thoughts on that + codecs/IQ?
ahalpert
Major Contributor
I read more recently that RED thought that their global shutter mode was good enough that they may ONLY implement global shutter, or something to that effect. Obviously, the camera is not finalized yet, though.
13.5 stops of usable DR would be quite a lot if you rate the Alexa at 14 stops!
What are your quarrels with RED? The overemphasis on resolution? You prefer Arri color?
R3d codec seems nice to me
G
Guest
Guest
I don’t rate the Alexa, the Alexa rates me.
I own a Red, and will likely buy a Komodo when the times comes. Will be nice to have a smaller camera that has a similar look and same workflow.
Arris is great, but until the LF didn’t have enough resolution for me. Arri is great for movies and TV. Things were the viewing distance makes resolution irrelevant. But 2 megapixels is not a lot. Video gets away with less resolution for a variety of reasons, but editing and being able to move the camera are basically showing details. So, in the context of moving images craft, resolution is not as important. However, it is still lovely to have more resolution. Looks amazing. Arri slaughtered the MX sensor, but i could go either way with Red’s latest three sensors. Or Venice for that matter. They’re all pretty amazing. Looking forward to see if Arri improves on DR in their s35 4K sensor. But really difficult to beat the Red+Komodo pairing for owner op.
That would be a shame if they go fulltime global shutter, but may make sense from a business perspective. Sigh.
As far as DR numbers, we all have our own numbers, but i try to go with general consensus when i mention DR numbers. Just so everyone is on the same page. I rate Arri Alexa as ~15, and Red Gemini and Venice as 14+ With the edge given to Venice. Red doesn’t do in camera noise reduction, so, in comparisons, Venice and Arri come out ahead by more than strictly just DR, as they show less noise. To be honest, i like having the in camera processing, but i also like having the better battery life. So, kind of a wash.
Few cameras are good enough that i just like the image no matter where it is pointed. Canon 1D series cameras (stills mode) and the Mini LF are possibly the only cameras that no matter what i just kind of gasp at how good everything is. Every other camera including the Alexa SXT don’t really do that for me. Red has gotten closer, but takes a little work, and the Venice and BM just have a bit of a mojo signature that doesn’t always vibe with me. Panasonic Varicam has great colour, but something gets lost along the way to the final image in all the projects i’ve seen it on. I assume that is because it is only used when low light is needed. Idk, i haven’t really used it other than jumping onto a project real quick. I don’t know why i never really liked the Canon C series. The C500ii is the first one i like. Though i still feel the 1D cameras do it better. But i am going way down into broad sweeping statements and loose observational preferences.
Last edited:
John Brawley
Well-known member
I think it's what RED should have been doing from the start. They got a bit obsessed with the high end / high resolution market.
Like other RED packages, you know that it won't be 5K to buy, you'll need to throw a lot more at it.
If it's a new sensor, which it seems to be than I think it could be a great little camera. I'm personally not much of a fan of the modular boxy machismo that RED like to pursue, but it's ideal for the gimbal lovers
I think it will be very successful if it does what's promised. R3D is a great and well established codec so no post issues either.
BMD went through trying to make sensor that was GS and RS switchable but it's SO MUCH WORK to make both modes work. It's kind of like the sensor work starts again. It's actually the same amount of work as if you were building two unique cameras. So them saying "only GS" makes me think, I remember hearing that before when it was many many more months and months of work to do both
In their case they predicted that there would also be less DR on GS, but it wasn't clear cut either.
JB