"H.265
ISO800 N-Log - noise reduction destroys all the high resolution details.
RAW
- 8 stops of exposure latitude is the limit! Putting this into context, the Panasonic LUMIX S1 II managed 10 stops of exposure latitude.
-The Sony A9 III (Lab Test here) with the global shutter sensor exhibited 9 stops of exposure latitude, just using the 10-bit internal compressed codec!
- average but OK dynamic range figures, and an average 8 stops of exposure latitude put it in the middle ground of full-frame consumer cameras."
I'd also add CineD had an interesting reply in the comments;
Question/Post: "The video is great, thanks! I have one question: on paper, the dynamic range looks bad, but in real footage it looks amazing — even compared to cameras with 2 or 3 more stops. How is that possible?
CineD comment: "This is exactly what we are saying. Average camera (specs), gorgeous images . At the end of the day, we are not filming charts but actual footage, and this camera is doing a good job of capturing images. "
neoshazam wrote:
I'd also add CineD had an interesting reply in the comments;
Question/Post: "The video is great, thanks! I have one question: on paper, the dynamic range looks bad, but in real footage it looks amazing — even compared to cameras with 2 or 3 more stops. How is that possible?
CineD comment: "This is exactly what we are saying. Average camera (specs), gorgeous images . At the end of the day, we are not filming charts but actual footage, and this camera is doing a good job of capturing images. "
Indeed - seems to be general consensus that the raw video looks very good.
It’s the H.265 that we want fixed ;-)
This is likely a dealbreaker for me - but hopefully firmware fixes that pronto.
Meanwhile, hopefully the other camera companies are going to put these big juicy screens on their bodies starting ASAP.
(though that might be complicated because the competitors have/need fans on them, which means they already have extra battery consumption…)
Some initial thoughts after spending a little more time with the ZR, though still not quite enough time:
- The display is incredible. Every other camera screen is absolutely painful to look at after this. I hope this kicks off a trend of larger, better displays on new cameras.
- However, the above makes the UX very touchscreen-centric. Maybe with some practice it'll become easier, but taking me a minute to really get into it. Muscle memory has me reaching for buttons that simply don't exist on this thing. Have had a couple of frustrating moments.
- Sony lenses work surprisingly well with it via the Megadap adapter. Autofocus is obviously not as dependable as Sony lenses on a Sony body, but it's worked well enough to not have to worry about it. Obviously, no distortion or other lens aberration corrections but it's livable.
- IBIS is pretty decent. I'd argue better than the one on my FX3, which isn't all that amazing to begin with.
- I struggled a little bit with exposing R3D at first, but after some RTFM and watching the Cine D video on it, I think I'm starting to get the hang of it. Definitely benefits from exposing to protect highlights, as opposed to ETR. Using zebras has been helpful, especially set to highlights at "245" (specifically for R3D). There's more helpful info here.
- R3D footage is of course noisy by default, but cleans up well in Resolve. And as long as you expose properly, it actually looks... REALLY F***ING GOOD.
I'm sure I'll have more thoughts as I get more time with it, but so far I'm excited to keep putting it through its paces.
n8rv wrote:
- The display is incredible. Every other camera screen is absolutely painful to look at after this. I hope this kicks off a trend of larger, better displays on new cameras.
- However, the above makes the UX very touchscreen-centric. Maybe with some practice it'll become easier, but taking me a minute to really get into it. Muscle memory has me reaching for buttons that simply don't exist on this thing. Have had a couple of frustrating moments.
I guess this is the compromise they made, to make the display large and the camera small, there is not much space for physical buttons.
I have a DJI Mini 5 Pro drone which has a controller that has physical controls for flight control but for exposure and generally camera settings, they are located in the rim of the controller display which is a touchscreen. I often accidentally turn the camera into Auto mode where it merges photos and does a funny kind of multi-shot hdr merge which I dislike a lot. I can turn this functionality off by touching the corner of the screen to turn the camera into Pro mode which gives manual controls and takes single photos instead of merged multiexposure fancy thingies that don't look natural to my eye. However, the area of that control on the touch screen is quite small and is not easy to use those touch screen controls to get what I want. Exposure compensation is also a touch screen operation. However, once set correctly the picture quality of the drone is actually quite good and while I can't get the results to look like they were from my Nikons, they can print well and add valuable perspectives to documenting a location, for example, from otherwise inaccessible or unsafe camera positions, and I love that.
However, it would be nice if my drone could have dedicated physical controls for basic things such as shutter speed, ISO, exposure compensation, and aperture (the thing has a fixed aperture, which may be OK due to the small sensor). I imagine the ZR does have enough dedicated controls for most of those functions when properly configured.
Interesting comment on Cined:
“Regular ProRes HQ is also soft like H265 on the Nikon ZR. I believe that Nikon purposely dumbed down the compressed codecs to keep the camera from overheating. In my tests, the data rate of H265 is about 50% less compared to Z63 and ProResHQ is about 30% less on ZR compared to Z6III. Nikon is doing this on purpose. I wish they would’ve charged a couple hundred dollars more and added a fan. Because I think that is the issue, it’s a thermal throttling.”
ronno wrote:
Interesting comment on Cined:
“Regular ProRes HQ is also soft like H265 on the Nikon ZR. I believe that Nikon purposely dumbed down the compressed codecs to keep the camera from overheating. In my tests, the data rate of H265 is about 50% less compared to Z63 and ProResHQ is about 30% less on ZR compared to Z6III. Nikon is doing this on purpose. I wish they would’ve charged a couple hundred dollars more and added a fan. Because I think that is the issue, it’s a thermal throttling.”
Interesting theory, but I don't see how reduced processing to better manage heat has anything to do with datarates.
I like listeing to other people on cameras but always like to verify-
I snagged a copy and can confirm 2 things-
The built in microphone with 32 bit audio works shockingly well- Best in a camera I have ever used.
H265 is soft like microwaved butter-
The rest of the camera is dialed...R3D is awesome and having the ISO felxibility is a really nice touch in post- Probably stupid not to use it.
Its a bit annoying there is no Auto White Balance but thats not that big of a deal -
I am traveling to NYC for 3 weeks tomrrow and will have this with me. Have a bunch of projects I will get to test this out on. Hopefully a firware update comes and corrects H265 because at current, its 100% not useable.
I think basically if you do a simpler/worse job at compression, it will take less computing power and heat the camera less. Thus no overheating when the cameras is already doing a lot of work to make the H265 files, and has no fan.
I am obviously not an engineer, so I don’t know why the data rate is also a lot lower.
Interesting theory, though.
Gary Irwin wrote:
Interesting theory, but I don't see how reduced processing to better manage heat has anything to do with datarates.
Also, thanks for the info and let us know how it goes in NYC.
RustyRus wrote:
I like listeing to other people on cameras but always like to verify-
I snagged a copy and can confirm 2 things-
The built in microphone with 32 bit audio works shockingly well- Best in a camera I have ever used.
H265 is soft like microwaved butter-
The rest of the camera is dialed...R3D is awesome and having the ISO felxibility is a really nice touch in post- Probably stupid not to use it.
Its a bit annoying there is no Auto White Balance but thats not that big of a deal -
I am traveling to NYC for 3 weeks tomrrow and will have this with me. Have a bunch of projects I will get to test this out on. Hopefully a firware update comes and corrects H265 because at current, its 100% not useable.
ronno wrote:
Back up a second…No auto white balance??
Also, thanks for the info and let us know how it goes in NYC.
Sorry- Shooting R3D - There is no Auto White Balance. You can select one but it really is set in post. Same with ISO- You can set one to try and get your exposure correct but ISO section doesn't matter when shooting in R3d. You set that in post as well.
Did a little back-to-back handheld comparison between the ZR and my FX3, trying to keep everything as evenly matched as possible.
6K R3D on the ZR and 4K UHD at the highest bitrate H.264 S-Log 3 on the FX3.
No grading, just basic color space conversion, a bit of highlight recovery, and a bit of noise reduction in Resolve.
I think they both look comparable-ish. The ZR retains extra detail thanks to the 6K resolution downscaled onto a 4K timeline, plus the RAW codec. FX3 looks a bit mushy in comparison due to the 4K resolution (no downscaling) and baked-in NR from camera. Any other thoughts?
Since the memory card is a major source of heat, lower data rates mean they run cooler and there is less risk of overheating.
However, since R3D NE has the highest data rates and there are no reports of overheating with it, I am not sure how they would have overheating if they had higher data rates in h.265 since those are a fraction of the R3D NE data rates. So to me this seems an unlikely reason for the stronger NR or less detail. The Z6III is also not known for overheating.
Gary Irwin wrote:
Interesting theory, but I don't see how reduced processing to better manage heat has anything to do with datarates.
n8rv wrote:
Did a little back-to-back handheld comparison between the ZR and my FX3, trying to keep everything as evenly matched as possible.
6K R3D on the ZR and 4K UHD at the highest bitrate H.264 S-Log 3 on the FX3.
No grading, just basic color space conversion, a bit of highlight recovery, and a bit of noise reduction in Resolve.
I think they both look comparable-ish. The ZR retains extra detail thanks to the 6K resolution downscaled onto a 4K timeline, plus the RAW codec. FX3 looks a bit mushy in comparison due to the 4K resolution (no downscaling) and baked-in NR from camera. Any other thoughts?
Are they sharpened? Looks like the Sony could use a bit more ;-)
Also, does the FX3 have 8k like the A1 does?
No sharpening applied to either.
But the ZR footage is definitely sharper and more detailed. It helps that it's 6K footage downscaled into a 4K timeline, as you're essentially packing in 1.5X the number of pixels. The fact that R3D Raw doesn't have any NR baked in from camera also helps.
The FX3 does not do 8K. It is 4K max and only records in H.264/H.265 internally, which also bakes in NR. All of these contribute to the FX3 footage being softer.
ilkka_nissila wrote:
Since the memory card is a major source of heat, lower data rates mean they run cooler and there is less risk of overheating.
However, since R3D NE has the highest data rates and there are no reports of overheating with it, I am not sure how they would have overheating if they had higher data rates in h.265 since those are a fraction of the R3D NE data rates. So to me this seems an unlikely reason for the stronger NR or less detail. The Z6III is also not known for overheating.
I don't know but I think there might be something to Ronno's theory.
My Z6III and Z8 will shut down in 30 minutes in the sun when continuously recording H.265 at high frame rates. They can record longer when shooting NRaw and ProRes because the processor isn't working as hard -- it's not denoising and compressing like it has to with H.265. Bitrates have nothing to do with it.
It's plausible that Nikon chose to forego the bulk, noise and cost of a fan knowing the camera would support long recording times when shooting R3D, ProRes and NRaw, but had to sacrifice some extra processing needed for H.265 to maintain good thermals, with the result being a substandard H.265 quality. The ZR is clearly targeting the cinema crowd which shoots raw. Anyone satisfied with H.265/264 already has the Z6III...which coincidently appears to have been reduced in price to match the ZR.
Ah I see.
Thanks again for going to the trouble and sharing! I would apply at least the basic sharpening to the Sony file before making a comparison. No one is going to use footage without any sharpening…
Cheers.
n8rv wrote:
No sharpening applied to either.
But the ZR footage is definitely sharper and more detailed. It helps that it's 6K footage downscaled into a 4K timeline, as you're essentially packing in 1.5X the number of pixels. The fact that R3D Raw doesn't have any NR baked in from camera also helps.
The FX3 does not do 8K. It is 4K max and only records in H.264/H.265 internally, which also bakes in NR. All of these contribute to the FX3 footage being softer.
ronno wrote:
Ah I see.
Thanks again for going to the trouble and sharing! I would apply at least the basic sharpening to the Sony file before making a comparison. No one is going to use footage without any sharpening…
Cheers.
Cheers.
Typically, yes, some sharpening might happen, but also depending on what the final product is meant to be.
In this case, I wanted to try the highest resolution and best codec each camera can offer, while keeping the processing on both to a minimum in order to directly compare footage at its most basic level. The difference in sharpness and detail "straight out of camera" is definitely something worth observing and considering.
Also, here’s another good “first impressions” video for the ZR, with interesting observations on dynamic range, etc.
This video scientifically compares various Nikon ZR codecs, testing their performance in different lighting conditions. The creator meticulously films test footage in diverse settings, including outdoor and studio environments, to showcase each codec's capabilities. Expect side-by-side comparisons and a detailed analysis of the results.
Yes, I saw this one. Unfortunately the dude shot often at narrow depth of field (manually focused) and while conducting the tests, he is moving in and out of the focus zone. So not that helpful IMO.
In any case, it seems pretty obvious that if you want to shoot RAW, this camera is a damn good deal in lots of ways.
sport wrote:
This video scientifically compares various Nikon ZR codecs, testing their performance in different lighting conditions. The creator meticulously films test footage in diverse settings, including outdoor and studio environments, to showcase each codec's capabilities. Expect side-by-side comparisons and a detailed analysis of the results.
. At the end of the day, we are not filming charts but actual footage, and this camera is doing a good job of capturing images. "