truth about Pocket 6k dynamic range

[thebrothersthre3]

I can send a google drive link for RAW files if anyone wants. But I think this demonstrates the dynamic range differences.

Ursa 4.6k over exposed

Pocket 6k over exposed

Ursa 4.6k under exposed (zoomed in to show noise pattern better)

 

Pocket 6k under exposed (zoomed in to show noise pattern better)

[androidlad]

Great, this aligns nicely with what C5D did with their over/under tests.

But, this is testing the latitude, not dynamic range.

[IronFilm]

On 6/27/2020 at 11:37 PM, Andrew Reid said:

Rather than go full frame as has been rumoured (let alone 8K), I think Blackmagic should bring back the old BMCC 2.5K sensor for a $999 model and do an updated Pocket 6K with same S35 sensor, but mirrorless L-mount or E-mount.

I'd be very strongly tempted to buy a Blackmagic Pocket 2.5K MFT for $599 (I think $999 is too close in pricing to the Pocket 4K to be a success)  or Pocket 6K MFT  for $1999 with an identical feature set/body as the current Pocket 4K/6K body design. 

[KnightsFan]

@thebrothersthre3 Raw files would be great, along with some information about your test. What ISO? How much over/under are these? What did you use as a base line for "correct" exposure? How are you measuring stops (aperture or shutter speed)? How did you process the images?

Did you process the images differently? Because on the under exposure images that the P6K clips more outside the window than the Ursa.

Not trying to say you're wrong here, just trying to understand what I'm looking at.

[thebrothersthre3]

1 hour ago, KnightsFan said:

@thebrothersthre3 Raw files would be great, along with some information about your test. What ISO? How much over/under are these? What did you use as a base line for "correct" exposure? How are you measuring stops (aperture or shutter speed)? How did you process the images?

Did you process the images differently? Because on the under exposure images that the P6K clips more outside the window than the Ursa.

Not trying to say you're wrong here, just trying to understand what I'm looking at.

The info outside the window was because I pulled down the highlights on the Ursa CDNG footage. With the Pocket 6k BRAW there is no option for highlights so I left it alone. 

Its a very unscientific test. Both cameras were set at 800 iso, 180 degree, F2.8. Then I blasted a 300w LED directly at my face 7 feet away. The idea was just to dramatically over expose and see which could bring back more detail. Same deal with the under exposed image. 

Post adjustments were just using the exposure slider. I believe I brought down the over exposed image 5 stops and brought the under exposed image up 4 stops. I didn't use the highlight recovery box. 

I am uploading the RAW files to Google Drive right now. I'd like to do a more scientific test and also use my XT3 in photo mode to see how they pair against those RAW files. But this test shows me what I was looking to see, which holds more highlight information and has more usable shadows.

 

5 hours ago, androidlad said:

Great, this aligns nicely with what C5D did with their over/under tests.

But, this is testing the latitude, not dynamic range.

Don't dynamic range and latitude directly correlate to each other tho? A sensor with more dynamic range will have more latitude and vice versa?

[KnightsFan]

Thanks for the info. I look forward to the DNG's.So you only used one camera setting total, and the only difference between the over- and under-exposure tests is that one has a 300W light and the other does not? Is that correct?

 

[thebrothersthre3]

3 minutes ago, KnightsFan said:

Thanks for the info. I look forward to the DNG's.So you only used one camera setting total, and the only difference between the over- and under-exposure tests is that one has a 300W light and the other does not? Is that correct?

 

We turned the light off and closed down the aperture to 5.6. 
The one lens was a sigma and the other a Rokinon so there could be a slight difference there but it would be super minimal. Got about 2 hours left on the upload to google drive. 

I was going to do a more structured test but I forgot to bring my light meter. The one issue is the underexposed images are probably too far gone. The Pocket 6k looks better still but its not really very usable haha. 

[majoraxis]

6 hours ago, IronFilm said:

I'd be very strongly tempted to buy a Blackmagic Pocket 2.5K MFT for $599 (I think $999 is too close in pricing to the Pocket 4K to be a success)  or Pocket 6K MFT  for $1999 with an identical feature set/body as the current Pocket 4K/6K body design. 

I would love both a 2.5k and a 6K pocket with micro 4/3 mount. I would also like to see a Micro Cinema Camera 6k.  Probably have to have a slightly bigger body,  but keep the m4/3 mount.  You get more than 4k resolution when cropping the 6k sensor to micro 4/3 size if your lens only covers micro 4/3 so there is no harm in having the 6k sensor with a micro 4/3 mount.

Any form factor of the 6k sensor and micro 4/3 mount and I will purchase.

[androidlad]

1 hour ago, thebrothersthre3 said:

Don't dynamic range and latitude directly correlate to each other tho? A sensor with more dynamic range will have more latitude and vice versa?

Related, but not really directly correlated.

Dynamic range is a measure of a camera system - how far it can see into the shadows and how far it can see into the highlights. Dynamic range can be measured objectively, but even then there's a subjective component as each and every viewer will have their own noise tolerance threshold. This governs how much of the shadow part of the dynamic range they find actually usable.

Latitude is related to dynamic range, but it is also scene dependent. Latitude is the degree to which you can over or under expose a scene and be able to bring it back to a usable exposure value after recording. It is dependent upon dynamic range, which is going to set the overall boundary of by how much you can over and under expose, but it's limited by the scene too, and how bright and dark the scene itself goes.

Say a scene has a range of brightness of 5 stops (a typical Macbeth chart for instance), and let's use a camera that has a 12 stop dynamic range. If we place the scene in the middle of that camera's recordable range, we have 7 stops to play we can we could over or under expose by 3.5 stops and still recover the scene.

But if the scene was a real world scene of actor against a sunlit window and the range of brightness of 15 stops, you don't have any latitude at all - no matter how you expose that scene you're going to loose shadow or highlight information.

So yes, latitude and dynamic range are related, but different. Latitude can't really be used to infer how much total dynamic range a camera has.

[thebrothersthre3]

https://drive.google.com/drive/folders/1PxAM3CzdQrwPsCqm2DWTIZiKxoC2Cy5J?usp=sharing

Here are the RAW files. 

 

[thebrothersthre3]

28 minutes ago, androidlad said:

Related, but not really directly correlated.

I am slightly confused by latitude vs dynamic range. I'll get it eventually. 

[sanveer]

Dynamic range is how much information you can See in the shadows and highlights. All of that may not be usable in post, and depending upon signal to rise ration, it could vary greatly. 

Exposure latitude is how much of that, you can further push in post, without degrading the image enough, to make it unusable. It has got to do the sensor, and more importantly, the codec bit depth. Like a 14-bit codec would always be a (little) better than a 12-bit one, a 12-bit one would be better than a 10-bit one etc. I am guessing the arrangement if middle grey would also govern the latitude, and whether the highlights are better protected or shadow information. 

Sensor size, to some extent would also govern how much of latitude is available in the final image.

Please correct me if I an wrong. 

[thebrothersthre3]

10 minutes ago, sanveer said:

Dynamic range is how much information you can See in the shadows and highlights. All of that may not be usable in post, and depending upon signal to rise ration, it could vary greatly. 

Exposure latitude is how much of that, you can further push in post, without degrading the image enough, to make it unusable. It has got to do the sensor, and more importantly, the codec bit depth. Like a 14-bit codec would always be a (little) better than a 12-bit one, a 12-bit one would be better than a 10-bit one etc. I am guessing the arrangement if middle grey would also govern the latitude, and whether the highlights are better protected or shadow information. 

Sensor size, to some extent would also govern how much of latitude is available in the final image.

Please correct me if I an wrong. 

So like with the Pocket 6k it appears to be clipped but change the gamma and colorspace to rec2020 and all that information is there. Where as with the Ursa its still clipped white. 

[AlexTrinder96]

1 hour ago, androidlad said:

Related, but not really directly correlated.

Dynamic range is a measure of a camera system - how far it can see into the shadows and how far it can see into the highlights. Dynamic range can be measured objectively, but even then there's a subjective component as each and every viewer will have their own noise tolerance threshold. This governs how much of the shadow part of the dynamic range they find actually usable.

Latitude is related to dynamic range, but it is also scene dependent. Latitude is the degree to which you can over or under expose a scene and be able to bring it back to a usable exposure value after recording. It is dependent upon dynamic range, which is going to set the overall boundary of by how much you can over and under expose, but it's limited by the scene too, and how bright and dark the scene itself goes.

Say a scene has a range of brightness of 5 stops (a typical Macbeth chart for instance), and let's use a camera that has a 12 stop dynamic range. If we place the scene in the middle of that camera's recordable range, we have 7 stops to play we can we could over or under expose by 3.5 stops and still recover the scene.

But if the scene was a real world scene of actor against a sunlit window and the range of brightness of 15 stops, you don't have any latitude at all - no matter how you expose that scene you're going to loose shadow or highlight information.

So yes, latitude and dynamic range are related, but different. Latitude can't really be used to infer how much total dynamic range a camera has.

You could've at least quoted or linked where you copied this information from... 😆

https://cinematography.com/index.php?/topic/51973-dynamic-range-vs-latitude/

[KnightsFan]

@thebrothersthre3 Thanks for the files. I've played around with them a little bit. It's so close, I can't definitively say one looks to have more DR than the other. I do think that, although the FPN is subjectively displeasing in motion, the actual amount of noise looks slightly less on the Ursa... though it's obviously hard to say. One really interesting thing is just how different the colors are. The Ursa is very magenta. I can definitively say that from these images, I like the P6K version better.

Just to go back against the other measurement, C5D puts the P6K at 11.8 in 6k at ISO 400, and the Ursa 4.6K at 12.5 when the 4.6k is downscaled to UHD in ISO 800. That's gotta be really close if you also downscale the P6K to UHD. Also worth noting that they measured at 400 instead of 800. I don't know which way that would change things--you'd expect worse DR in a non-native ISO, but it could also have different noise reduction. So overall I'd caution against saying this disproves the other measurement, but it is evidence against the Ursa having more DR than the P6K at any given setting.

 

The other thing I will say about latitude is that while latitude does not equal DR, its is true that if Camera A has both more over- and under-exposure latitude than Camera B, then Camera A also has more dynamic range (discounting subjective opinions on one having a "nicer" noise pattern).

[thebrothersthre3]

36 minutes ago, KnightsFan said:

@thebrothersthre3 Thanks for the files. I've played around with them a little bit. It's so close, I can't definitively say one looks to have more DR than the other. I do think that, although the FPN is subjectively displeasing in motion, the actual amount of noise looks slightly less on the Ursa... though it's obviously hard to say. One really interesting thing is just how different the colors are. The Ursa is very magenta. I can definitively say that from these images, I like the P6K version better.

Just to go back against the other measurement, C5D puts the P6K at 11.8 in 6k at ISO 400, and the Ursa 4.6K at 12.5 when the 4.6k is downscaled to UHD in ISO 800. That's gotta be really close if you also downscale the P6K to UHD. Also worth noting that they measured at 400 instead of 800. I don't know which way that would change things--you'd expect worse DR in a non-native ISO, but it could also have different noise reduction. So overall I'd caution against saying this disproves the other measurement, but it is evidence against the Ursa having more DR than the P6K at any given setting.

 

The other thing I will say about latitude is that while latitude does not equal DR, its is true that if Camera A has both more over- and under-exposure latitude than Camera B, then Camera A also has more dynamic range (discounting subjective opinions on one having a "nicer" noise pattern).

I don’t think the noise pattern is subjective in this case. It’s unusable on the URSA. Do you not think the pocket is keeping the highlights better tho ? My face can pretty much be recovered with the P6k where it’s clipped to white on some parts with the URSA. I want to do another test on the URSA to see where the weird vertical lines start happening. I would agree the overall noise is slightly less. But those vertical lines just aren’t removable with NR. The pocket and ursa do better in shadows at 400 iso but I prefer highlight to shadow info usually. 

[KnightsFan]

17 hours ago, thebrothersthre3 said:

I don’t think the noise pattern is subjective in this case. It’s unusable on the URSA. Do you not think the pocket is keeping the highlights better tho ? My face can pretty much be recovered with the P6k where it’s clipped to white on some parts with the URSA. I want to do another test on the URSA to see where the weird vertical lines start happening. I would agree the overall noise is slightly less. But those vertical lines just aren’t removable with NR. The pocket and ursa do better in shadows at 400 iso but I prefer highlight to shadow info usually. 

The pocket is keeping the highlights better, yes. That doesn't necessarily equate to DR because we don't know the distribution, and I do think that the Ursa is better in the underexposure strictly in terms of amount of noise, not that you would use it because the fixed pattern is very unpleasant. But Ursa is also keeping the sharpness in the underexposure just slightly better--though that could be a slightly sharper lens. But of course that's splitting hairs as you wouldn't use either one at that point.

If you do any more tests, what I would do is get a gradient of light and ensure that they clip at the same point, and then check the shadows. That would avoid any variables with them distributing DR differently around middle grey, or any discrepancies between what the two cameras use as middle grey to begin with. Maybe I'll do a similar test with my cameras just for fun.

[TheRenaissanceMan]

The P6K handles noise reduction in the shadows much better than the UM46, which struggles with patterns and ugly color shift even after generous NR. 

[thebrothersthre3]

57 minutes ago, TheRenaissanceMan said:

The P6K handles noise reduction in the shadows much better than the UM46, which struggles with patterns and ugly color shift even after generous NR. 

Yeah that’s the issue with the ursa underexposed 

 

6 hours ago, KnightsFan said:

The pocket is keeping the highlights better, yes. That doesn't necessarily equate to DR because we don't know the distribution, and I do think that the Ursa is better in the underexposure strictly in terms of amount of noise, not that you would use it because the fixed pattern is very unpleasant. But Ursa is also keeping the sharpness in the underexposure just slightly better--though that could be a slightly sharper lens. But of course that's splitting hairs as you wouldn't use either one at that point.

If you do any more tests, what I would do is get a gradient of light and ensure that they clip at the same point, and then check the shadows. That would avoid any variables with them distributing DR differently around middle grey, or any discrepancies between what the two cameras use as middle grey to begin with. Maybe I'll do a similar test with my cameras just for fun.

The ursa had a sigma lens vs a rokinon but it’s hard to say. I want to do more exact testing with my Xt3 as well, before I sell the ursa. 

[Andrew Reid]

The big mystery of the day.

Is @androidlad actually Graeme Nattress?

😃