What Nikon gets right

[MrSMW]

19 minutes ago, ND64 said:

there is not a noticeable difference between Z9/Z8 and R5ii

Other than that one is green and one is magenta, but apart from that... 😉

The only thing about the above result is it's a pity it's not the same model because different skin tones in real life lead to different skin tones in the result.

In the above ref. pics, Jonny's skin tones look better than the other fella's green, but this might have been a little different with the same model.

[Andrew Reid]

58 minutes ago, ND64 said:

So CinedD published their new lab test and in imatest there is not a noticeable difference between Z9/Z8 and R5ii

5 stop under with NR shows different color cast

But AF striping noise is visible in Canon image, even after NR

 

What I always wonder about these tests is if you have skintones in the shot that are 5 stops under exposed, what the heck are you trying to achieve with the overall shot and lighting?

[MrSMW]

1 hour ago, Andrew Reid said:

What I always wonder about these tests is if you have skintones in the shot that are 5 stops under exposed, what the heck are you trying to achieve with the overall shot and lighting?

I guess showing what the worse case scenario is if your set up and capture is badly considered?!

I'm more interested in seeing real world footage than extreme studio set ups some seem to focus on, but each to their own I suppose, but walls and charts and even cats, have little relevance to anything I shoot.

Also, when a camera or lens review is 100% at night. Unless that is all you ever do...

It's also what makes out of season for practice sake, very difficult for me as the material, landscapes and lighting I have to work with May-Sep (in season) I cannot replicate out of season. So I don't bother anymore and just get up to speed very quickly in the Spring!

Anyway, Nikons...

Now that you have a Z8 Mr Reid, are you considering a Zf sidekick in the near future?

[ND64]

Its more like an engineering horse race and some people love to bet. Of course exposure mistakes happen, but nobody should make a 5 stops mistake. 

[Django]

Actually the R5 mk2 cined lab test reveals some quite interesting results. You can choose the binned 4K for an impressive 9.7ms RS or the 4Kfine option for detailed oversampled 4K in Clog2 with class-leading 13.3 (SNR2) 14.2 (SNR1) stops of DR! That is totally outstanding and something no one has talked about up until now. This places the R5 mk2 in this shooting mode above all other mirrorless cameras and makes it sit in between the Venice 2 & Alexa Mini LF. I was already blown away by R5 mk2 Clog2 files I've played with and now I understand better why. I think R5 mk2 has been overlooked as the specs sheet isn't that impressive but actual test results are and for DR alone I think its revealing itself to be a killer camera. 

[Andrew Reid]

34 minutes ago, Django said:

Actually the R5 mk2 cined lab test reveals some quite interesting results. You can choose the binned 4K for an impressive 9.7ms RS or the 4Kfine option for detailed oversampled 4K in Clog2 with class-leading 13.3 (SNR2) 14.2 (SNR1) stops of DR! That is totally outstanding and something no one has talked about up until now. This places the R5 mk2 in this shooting mode above all other mirrorless cameras and makes it sit in between the Venice 2 & Alexa Mini LF. I was already blown away by R5 mk2 Clog2 files I've played with and now I understand better why. I think R5 mk2 has been overlooked as the specs sheet isn't that impressive but actual test results are and for DR alone I think its revealing itself to be a killer camera. 

So meanwhile let's look at what that means in the real world away from pixel peeping tests and number of stops.

14.2 stops so you can get an unusably bad image 5 stops under exposed?

Or does it have a genuinely big creative impact / image quality benefit.

After all it is "class leading above all other mirrorless cameras" so it must be pretty special and not just 1% better dynamic range than a 3 year old Nikon Z9, right?

And just because it has 1% more "test setup" dynamic range does not make it a better mirrorless camera than the Nikon Z8, because it's a lot more expensive than one for a start and with a far worse range of native lenses, and a less flexible system mount.

I do like the sRAW codec for 4K/60p, smaller file sizes, huge EVF, and the body design is a decent evolution from the original.

But the DR and C-LOG2... Meh.

Many more important things to consider overall like the system mount, codecs, handling, prices used and new, and so on.

 

[Andrew Reid]

4 hours ago, MrSMW said:

I guess showing what the worse case scenario is if your set up and capture is badly considered?!

Here's the thing about this...

Yes it is showing the fuck-up scenario.

And people look at it as some sort of ultimate performance metric!

4 hours ago, MrSMW said:

I'm more interested in seeing real world footage than extreme studio set ups some seem to focus on, but each to their own I suppose, but walls and charts and even cats, have little relevance to anything I shoot.

The real world footage all looks the same as the flagship cameras are all capable of image quality that far exceeds the capabilities of our displays and the YouTube delivery network.

There needs to be a new Zacuto challenge screened at a cinema, which really gets under the skin of the different cameras and how straight forward they are to expose, grade and whether getting totally optimal results means a lot of faff or a few easy tweaks, as that's what really matters isn't it?

I could have 1% less noise in skintones accidentally exposed 5 stops too dark but what good is that if C-LOG2 is a bitch to grade and takes hours?

And that 1% becomes less than 0.000001% in footage which is exposed correctly, less than 0.000000000000000001% in footage which has been lit well, and if the footage is going to end up on YouTube in 4K with web compression you can reduce that to a big fat NAUGHT.

4 hours ago, MrSMW said:

Anyway, Nikons...

Now that you have a Z8 Mr Reid, are you considering a Zf sidekick in the near future?

I did consider the Zf but decided there's more fun factor in an X-Pro3 or something a bit weird.

Still a bit put off by the fake-ness and mixed build quality of it. In some ways it's really nice and a bargain, but probably not needed alongside the Z8.

The Z8 does everything I need on a high-end camera and was a better option than EOS R5 II due to price and lens mount, and a better option than S1R II due to overall capabilities and a better sensor, plus again a more flexible mount.

[MrSMW]

1 minute ago, Andrew Reid said:

I did consider the Zf but decided there's more fun factor in an X-Pro3 or something a bit weird

Possibly and Fuji does have their funky SOOC Jpegs which IMO do beat the equivalent Zf (which is a bit of faffing to achieve) but for someone like me who only shoots raw, those Zf files are a definite step up over those cropped Fooj files.

3 minutes ago, Andrew Reid said:

Still a bit put off by the fake-ness and mixed build quality of it. In some ways it's really nice and a bargain, but probably not needed alongside the Z8.

Possibly not needed with a Z8 and a better companion to the Z8 for video or hybrid would probably be a Z6iii, but as a stills + hybrid, I do think the Zf would be better and more fun. But only if you needed/wanted a sidekick and you probably do not.

It's a slightly quirky one for sure and I bought it, used it on 3-4 jobs, sold it, re-bought another, have debated selling it (for another A7RV and I still might) but for now at least, it stays.

I can't see it not being around for at least the whole of this season actually because as I have mentioned multiple times, I AM more than happy with my current set up, but always considering alternative options and the most likely of those is a switch to wholly Nikon for 2026.

I want my in camera crop overlays a la LUMIX though in both stills and video plus full LUT support and until that happens, it is not happening for me (a switch to Nikon).

And even if they did bang those in with firmware next week, I think it's probably too close to my season kick off now to change so if there is going to be any change of system(s) it would not happen until this time next year.

But who knows what will happen between now and then...

[eatstoomuchjam]

1 hour ago, Django said:

You can choose the binned 4K for an impressive 9.7ms RS or the 4Kfine option for detailed oversampled 4K in Clog2 with class-leading 13.3 (SNR2) 14.2 (SNR1) stops of DR!

As for DR, is it really that exciting to be 0.5 stops (SNR2) and 0 stops (SNR1) better in 4K than the Canon C70, a super 35 camera that was released in 2020?  Or 0.4 stops better (SNR2) and 0.2 stops better (SNR1) than the Sony A7 IV, a camera that Sony users have been enjoying since 2021?  Or 0.2 stops better (SNR2) and 0 stops better (SNR1) than the Ronin 4D 8K which was released in late 2023?  Keep in mind that the DJI numbers, at least, are also raw which, presumably, doesn't have (much?) NR baked in, as any of the other cameras would when using non-raw codecs.

None of that is to say that the R5 II is a bad camera!  It's a really fantastic camera.  It's really just to say that it's not really meaningfully ahead of several other cameras.  That said, if someone offered me a free swap for my OG R5, I'd take it.  😉

[Django]

23 minutes ago, Andrew Reid said:

So meanwhile let's look at what that means in the real world away from pixel peeping tests and number of stops.

14.2 stops so you can get an unusably bad image 5 stops under exposed?

Or does it have a genuinely big creative impact / image quality benefit.

After all it is "class leading above all other mirrorless cameras" so it must be pretty special and not just 1% better dynamic range than a 3 year old Nikon Z9, right?

And just because it has 1% more "test setup" dynamic range does not make it a better mirrorless camera than the Nikon Z8, because it's a lot more expensive than one for a start and with a far worse range of native lenses, and a less flexible system mount.

I do like the sRAW codec for 4K/60p, smaller file sizes, huge EVF, and the body design is a decent evolution from the original.

But the DR and C-LOG2... Meh.

Many more important things to consider overall like the system mount, codecs, handling, prices used and new, and so on.

 

Oh I agree with you and not trying to argue R5 mk2 is the best camera in each category nor was I even comparing it to Nikons.

I was just pleasantly surprised to find out that in 4K fine mode the DR in Clog2 puts it all the way up to the 6th position in their rankings between Venice 2 and ARRI. In comparison the original R5 is ranked 72nd with 10.8 stops of DR. Quite a significant leap that no one seemed to really notice.

I'm not that concerned with exposure latitude tests as indeed I never intend to under expose an image to those limits however usable DR in a well exposed scene is definitely always a plus, especially in backlit situations which can be crucial in certain scenarios.

Also from my own tests, Clog2 is just a lot less contrasty and saturated than Clog3. Skin tones are much more natural. So that and the added DR is to me a significant advantage of the R5 mk2 vs mk1. 

Now comparing it to the Z8 that's a tougher one, different pros & cons. I do really like the Z8/Z9 and was so close to grabbing a Z9 last summer as it ticks so many boxes. But as mentioned I had exposure trouble shooting N-log and grading workflow was tricky versus Canon C-log. Maybe with more time on the camera I'd get to satisfactory shooting/grading level but sadly that just wasn't at all the case for me in my short time with the Nikon.

I did love many things about Z8/Z9, namely the ergonomics, tilt display, speed, ProRes codecs, 4K60p oversampled, color science and neutral/flat profiles. What I didn't like was menu/banks system, AF that would often focus on eye lash instead of iris, no LUT support, unreliable log exposure assistance. These things could all be corrected in firmware and the pros do outweigh the cons. In the end its the overall lens ecosystem that'll drive you one way or the other. I'm more invested in Canon so its going to take a lot for me to switch to Nikon. 

Rumor has it R5 mk2 is getting open-gate. Obviously to be taken with a huge grain of salt, and Nikon could very well add that as well. That would be a significant draw for me.

[eatstoomuchjam]

2 hours ago, Django said:

I was just pleasantly surprised to find out that in 4K fine mode the DR in Clog2 puts it all the way up to the 6th position in their rankings between Venice 2 and ARRI. In comparison the original R5 is ranked 72nd with 10.8 stops of DR. Quite a significant leap that no one seemed to really notice.

I'd continue to be cautious with those numbers, though.  Given that the two cameras are only about 1/2 stop apart in 8K raw mode - which, if we're talking about an oversampled mode, is the same one (or close to the same one) that gets downscaled to 4K, higher numbers might be less indicative of a huge jump and more indicative of more aggressive noise reduction.  If not, great, but rather than taking test chart results at face value, we should recognize them as a useful tool for understanding certain aspects of performance.

(And if getting a lot of detail in the image isn't a goal, then the higher NR settings SOOC might be just fine anyway!)

[ND64]

2 hours ago, eatstoomuchjam said:

And if getting a lot of detail in the image isn't a goal, then the higher NR settings SOOC might be just fine anyway!)

I mean even iPhone could achieve 12 stops! with NR. 

[Marcio Kabke Pinheiro]

On 3/10/2025 at 3:36 AM, MrSMW said:

It should have been LUMIX and just LUMIX from the start.

It keeps circling back to the same broad topic; branding and marketing.

I think the opposite. Panasonic was a very known brand in the VHS times - their camcorders were some of the best, their VHS recorders sold millions of units. Could not be recognized as a stills camera company, but as a video brand, was on pair with Sony (at least here in South America).

Lumix appeared with their point & shoot cameras. Is like Sony changing their logo in front of the cameras to "Cybershot" instead of Sony.

Every non-camera person that ever saw me with a Lumix always ask "what brand is this, never heard", I say "It's made by Panasonic" and immediately people make a face of "oh, THIS i know".

[IronFilm]

On 3/11/2025 at 4:16 AM, MrSMW said:

There is an argument that the FF market is already saturated so maybe they should not try and compete in that arena, but then MF is 'sexy' over FF which most want over APSC which most want over 4/3.

I suspect in the long run then Fujifilm might be proven right in their decision to target differently both Medium format and APS-C. And to not try to attack so called "Full Frame" head on. 

Fujifilm has APS-C so that people who want highly portable kit (smaller lens!) or very cheap kit (such as the X-M5, or an upcoming X-T300 or X-A8) then they can get a Fujfilm for that. 

But for those who insist on "the best" then the ever decreasing prices of sensors means we'll see Medium Format drop down into reach of more and more people. 

Just like with what happened in the DSLR Era. Initially nobody could afford Full Frame DSLRs (they didn't even exist!), then only the very richest hobbyists or working pros could afford it, but with time, eventually anybody could afford a Full Frame DSLR. (Nikon D600 / D700 or Canon 6D / 5D mk1 goes for pennies these days!) 

I predict the same will eventually happen for Medium Format as well. And who will be positioned the best to take advantage of this? Fujfilm will be. 

[MrSMW]

Fuji avoiding FF and sitting both sides seems like a good call to me also.

I have often thought about going back to them for cropped video and GFX stills.