Foot candles! Old school! I was taught that method by Bruce Surtees, an old school DP, trained by his father, Robert Surtees, an old-old school DP. Super effective! Thank you for your efforts here, Sir. Chappeaux.
But isn't changing a digital camera's 'native or base ISO' a way to generate artifacts anyway? Essentially because you are not operating at 100% of the sensor? (less dynamic range). I believe the Ursa Mini 4.6 native/base ISO is 800. (So each time you go up a stop in ISO, noise is doubled). So, in low light... shouldn't we shoot LOG and use it along with shutter and aperture before turning the ISO? Sorry, if this seems dumb but I'm s newbie doing a lot of investigating... so I don't really know for sure.
Shutter is typically set to 1/48 sec or 180 degrees, and aperture is a creative decision. We control exposure with ND filters and lighting- ideally. Most cameras employ noise reduction, when not shooting RAW, and I suspect some cameras still do a little noise reduction when shooting RAW. This test is pretty old in terms of camera technology, and I don't remember a lot of details about it to be honest. So from a practical/functional standpoint we use ISO variation within certain limits when exposing- usually camera dependent. Namely, what is the highest ISO you are willing to shoot with- it varies with each camera. For all practical purposes, on some shoots you have to go beyond your comfort zone if you want to get the shot, knowing someone has to "fix" it in post. Shooting LOG is very popular. HOWEVER, it all depends on your post workflow/budget. Some clients specifically ask NOT to shoot log. ALL of my clients the past 8 years have specifically requested NOT to shoot RAW- basically ever since RED sort of faded from mid tier projects. The post overhead is too great/expensive. This is a slightly long winded way of saying I don't remember how this camera processed the signal, other than anything above 800 ISO started to introduce fixed pattern noise, which greatly detracted from the image, which greatly deterred clients from using this camera. (or was that the original UrsaMini 4K- can't remember) Mind you BRAW and Prores RAW did not exist, at least not as a retail solution, at this time. Also, noise is not to be avoided at all costs if it is aesthetically pleasing. After all, we still add a little film grain on some projects, to make the image "feel" more organic- basically to get away from a hyper video look that is extremely common. So, it all depends on the project, the client, and your preferences. There is no "right" or "wrong" way to generate an image. For me, the "right" way is the way the client wants it and is willing to pay for it. But I do this for a living so I come at it from that perspective, not an academic or theoretical perspective. It's certainly worth looking into further from an academic/curiosity perspective.
@@cinemak1759 - Thank you for the quick and informative reply. You have confirmed some of what I’ve been looking into. It was very helpful! One of the reasons I was led to your page is because I’m considering buying into Black Magic. Of course now with the 12K camera, the game has changed dramatically since this video. (But I’m still considering the “older” camera.) And like you said, it’s all about application. For myself, it’s going to be more of a controlled and creative environment as opposed to ‘on the fly’ or an industrial set up. In short, I’m going for a cinematic look. Soooo.... controlling noise in my video should be “easier”... considering I will be using post production techniques. (along with many other applications ranging from lenses to lighting, etc.) No doubt… The road is long. Thanks again for the help. 🙏
Karl, great test, simple and straight to the point! have you compared the UM46 against a DSMC2 body? I have an WPN-MG but it was out on a rental and I have just wrapped a feature where I had UM46s and I was pleasantly surprised, aside the size/weight and especially the slightly more videoish motion cadence. What do you think? We can speak freely here ha! ;)
I feel most if not all of the 4K and higher rez cameras can start to feel like video real fast if you're not careful- Sony in particular, but definitely Canon, Panasonic (Varicam too) Red. Lighting and lenses make the difference- in particular lighting. I've been using ancient Nikkors primes and older Tokina ATX Pro zooms (the 28-70 designed by Angenieux) and that helps, but sometimes life would be a lot easier with a modern, cinema zoom. Lens tests are in order- and on order!
I totally agree but believe me, there's something else other than lighting, motion cadence is really different across cameras, Sony is the worst offender but Panasonic is close, BMD usually is not bad but the UM46 is "temporally sharper" (for the lack of a better description he he) VS the BMPCC/BMMCC/BMPC/UM4K
The UM46 is actually the first BM camera I've used, outside of a quick inspection. Can't voice an opinion on the other BM cameras for lack of familiarity.
Are you seeing any aliasing and moiré when shooting certain fabrics on clothing? I've only have 4 shoots under my belt with the UM4.6 and I'm seeing moiré on the talents wardrobe. It is certainly not as bad as the 5D MII but it is noticeable. I had a Red Scarlet MX running as a 2nd camera interview shoot so I had something to compare with. Granted the Red has a OLPF and the UM4.6 does not. Seems like you have a keen eye for these things. Do you have an issue with your UM4.6?
Yes, the camera does show moire. We ask subjects/talent to bring clothing options, and ask they don't wear clothing with tight patterns, sweaters etc. Same constraints video always had. So same old same old as far as moire is concerned.
The crews I work with all know footcandles. It's old school- film days. But with so many new cameras and so many new lights, looking for a constant to reference. Footcandles seems like a good place to start for me.
Foot candles! Old school! I was taught that method by Bruce Surtees, an old school DP, trained by his father, Robert Surtees, an old-old school DP. Super effective! Thank you for your efforts here, Sir. Chappeaux.
Yes! Finally, someone mentions the crosshatch patterns.
Thank You so much! the 1st low light camera test of the BM URSA 4.6k that gives me the information I was looking for :)
More than welcome. Hoping to do more camera and lens tests in the future.
Legend has it that he tests his car's horsepower with actual horses. (Just kidding! Great job on the video!)
LOL
This is a really great video. I loved the light meter bit at the end' that could be a separate video.
Agreed, he gave a very thoughtful explanation on its use vs. film stock.
Thank you for the test.
Thanks very much. I think the footcandles measurement is a good baseline as well.
I love the foot candle references!
Can you compare it with Sony A7SII, please ?!
Great review man!
Really helpful video, thank you!
But isn't changing a digital camera's 'native or base ISO' a way to generate artifacts anyway? Essentially because you are not operating at 100% of the sensor? (less dynamic range). I believe the Ursa Mini 4.6 native/base ISO is 800. (So each time you go up a stop in ISO, noise is doubled). So, in low light... shouldn't we shoot LOG and use it along with shutter and aperture before turning the ISO? Sorry, if this seems dumb but I'm s newbie doing a lot of investigating... so I don't really know for sure.
Shutter is typically set to 1/48 sec or 180 degrees, and aperture is a creative decision. We control exposure with ND filters and lighting- ideally. Most cameras employ noise reduction, when not shooting RAW, and I suspect some cameras still do a little noise reduction when shooting RAW. This test is pretty old in terms of camera technology, and I don't remember a lot of details about it to be honest. So from a practical/functional standpoint we use ISO variation within certain limits when exposing- usually camera dependent. Namely, what is the highest ISO you are willing to shoot with- it varies with each camera. For all practical purposes, on some shoots you have to go beyond your comfort zone if you want to get the shot, knowing someone has to "fix" it in post.
Shooting LOG is very popular. HOWEVER, it all depends on your post workflow/budget. Some clients specifically ask NOT to shoot log. ALL of my clients the past 8 years have specifically requested NOT to shoot RAW- basically ever since RED sort of faded from mid tier projects. The post overhead is too great/expensive.
This is a slightly long winded way of saying I don't remember how this camera processed the signal, other than anything above 800 ISO started to introduce fixed pattern noise, which greatly detracted from the image, which greatly deterred clients from using this camera. (or was that the original UrsaMini 4K- can't remember) Mind you BRAW and Prores RAW did not exist, at least not as a retail solution, at this time.
Also, noise is not to be avoided at all costs if it is aesthetically pleasing. After all, we still add a little film grain on some projects, to make the image "feel" more organic- basically to get away from a hyper video look that is extremely common. So, it all depends on the project, the client, and your preferences. There is no "right" or "wrong" way to generate an image. For me, the "right" way is the way the client wants it and is willing to pay for it. But I do this for a living so I come at it from that perspective, not an academic or theoretical perspective.
It's certainly worth looking into further from an academic/curiosity perspective.
@@cinemak1759 - Thank you for the quick and informative reply. You have confirmed some of what I’ve been looking into. It was very helpful!
One of the reasons I was led to your page is because I’m considering buying into Black Magic.
Of course now with the 12K camera, the game has changed dramatically since this video. (But I’m still considering the “older” camera.)
And like you said, it’s all about application. For myself, it’s going to be more of a controlled and creative environment as opposed to ‘on the fly’ or an industrial set up. In short, I’m going for a cinematic look.
Soooo.... controlling noise in my video should be “easier”... considering I will be using post production techniques. (along with many other applications ranging from lenses to lighting, etc.)
No doubt… The road is long. Thanks again for the help. 🙏
Karl, great test, simple and straight to the point! have you compared the UM46 against a DSMC2 body?
I have an WPN-MG but it was out on a rental and I have just wrapped a feature where I had UM46s and I was pleasantly surprised, aside the size/weight and especially the slightly more videoish motion cadence. What do you think? We can speak freely here ha! ;)
I feel most if not all of the 4K and higher rez cameras can start to feel like video real fast if you're not careful- Sony in particular, but definitely Canon, Panasonic (Varicam too) Red. Lighting and lenses make the difference- in particular lighting.
I've been using ancient Nikkors primes and older Tokina ATX Pro zooms (the 28-70 designed by Angenieux) and that helps, but sometimes life would be a lot easier with a modern, cinema zoom.
Lens tests are in order- and on order!
I totally agree but believe me, there's something else other than lighting, motion cadence is really different across cameras, Sony is the worst offender but Panasonic is close, BMD usually is not bad but the UM46 is "temporally sharper" (for the lack of a better description he he) VS the BMPCC/BMMCC/BMPC/UM4K
The UM46 is actually the first BM camera I've used, outside of a quick inspection. Can't voice an opinion on the other BM cameras for lack of familiarity.
Did you mention your t-stop? I know it's irrelevant to your test, but I'm curious what it is for 1 footcandle to expose so high.
T-stop: 1.4 The exposure is low, raised in Adobe Premiere's Lumetri so my face is @70 ire
Are you seeing any aliasing and moiré when shooting certain fabrics on clothing? I've only have 4 shoots under my belt with the UM4.6 and I'm seeing moiré on the talents wardrobe.
It is certainly not as bad as the 5D MII but it is noticeable. I had a Red Scarlet MX running as a 2nd camera interview shoot so I had something to compare with. Granted the Red has a OLPF
and the UM4.6 does not. Seems like you have a keen eye for these things. Do you have an issue with your UM4.6?
Yes, the camera does show moire. We ask subjects/talent to bring clothing options, and ask they don't wear clothing with tight patterns, sweaters etc. Same constraints video always had. So same old same old as far as moire is concerned.
ty for sharing karl!
thanks for this video.
Hmm... I don't know anyone that talks in footcandles. I'd like to know more.
The crews I work with all know footcandles. It's old school- film days. But with so many new cameras and so many new lights, looking for a constant to reference. Footcandles seems like a good place to start for me.
Footcandles are probably the most tried and true objective way to run any comparative light test.
Great test, thank you so much.