Update: The proportions of the three different resolutions in Q5 are on average as follows: 4K: 1 GB 8K: 1,4 GB (140%) 12K: 1,5 GB (150%) The color differences are less pronounced in ACES, but still present. The name of the experiment should probably be more „slit“ or „grating“, I’m not quite sure how to translate it correctly.

@mustujabk

10 ай бұрын

What is the best setting for this camera? If you don't remind me asking, Or just make a video :-)

@deadendboredom

10 ай бұрын

@@mustujabk In general, the best settings depend on what you intend to do and your level of expectation. Personally, the best settings for me are 4K with Q5, providing an organic image while avoiding compression artifacts and keeping the file size as small as possible. (ISO 800, Shutter 180*)

Thank you for this thorough breakdown! After two years with this camera, I've found that 8k @ 18:1 has generally been my go-to. Now I'm over here doing 12k Q5 tests. While I rarely use it for 12k, this camera packs a huge amount of value. Form factor, being able to multi frame for horizontal and vertical deliverables, high resolution HFR, braw, pairing seamlessly with pocket 4k/6k footage, price, I'm really surprised the 12k hasn't been more popular. I will say that noisy shadows have turned some of my colleagues off to using this on projects, but everyone chasing "organically filmic" simultaneously expecting "clinically noiseless" is unrealistic. Really blown away with all camera systems on the market right now. Still feel like the blackmagic ecosystem is the best fit for my needs.

@raredreamfootage

Жыл бұрын

The noise looks very filmic though and if you bring the shadows down in post, it looks really good.

I'd love to see the same tests done on the 12K OLPF camera!

@John97857

5 ай бұрын

I was about to ask if a similar test just for noise can be done for the OLPF. I assume the file sizes would be the same.

@deadendboredom

5 ай бұрын

Yes, the file sizes won't change. If I ever come across an OLPF version, I'll definitely test it, but don't get your hopes up too high, that's unlikely to happen.

I'd also like the same tests dome on the 12k 2024 camera. With the higher quality camera the color should be more stable and less noise should be evident in all shots.

I appreciate your hard work and thoroughness. Thank you.

This is amazing thank you for your efforts

Great review as usual, really good video

thank you for doing the work for us. Really appreciate the effort

Thank you for your time and effort on this project. Exactly what was needed. BM should post this themselves.

This is an awesome video and I appreciate the honesty especially as I am looking into getting a 12k. God bless you and have a great day.

Thank you for this amazing content! I recently bought two ursas mini pro 12k, and your video is helping me a lot! Say hello to Brasil in next video! 😂

Beautiful Test!

amazing video. thank you!

Thank you sir. This is the first of things I warned about years ago, but you have done the work. The arrogant nuts were always pro brain falling out, and against this sort of examination, which shows clearly better image as you go up in the image settings. But what are you looking at. Let me take a stab at this: braw is not a raw recording in the sensor format. It is in between. The type of codec and image preservation and enhancement repair, is what I suggested they do, which is based off of my earlier recommendation to preserve image quality under mpeg 2 and h264 nearly decades ago, where quality increased. Firstly, I pushed for 2:1-2.5:1 to get optimum quality for post processing the image into highest quality. This was denied. I don't know what third party tools aim at doing this, but what you see as image colour artifacting, are clues to the original colour at the pixel pixel location. It looks bad because the sender pattern helps with low light pick up and extended dynamic range functions, as the clear and three primaries can be used a bit like nd filters. To me, the sensor colour pattern is too gappy and low on the colour tiles, and bound to lead to colour detail issues, particular at different resolutions. I had advocated for the random sensor pattern, which actually works out to be a diamond like pattern. Sparse patterns are OK for tv cameras, as the imare is heavily compressed and tanked by low component colour schemes, where you have a group of 4 pixels with compromised red and blue, and things get a bit muddled. I am ok with the Kodak rgbW semi Bayer pattern, but that's about as sparse as I want. This is the modern equivalent of the JVC HD1/HD10, that had a hybrid image sensor, some similar looking coloured noise, and not the greatest in low light. A 1.28k camera, this is a 12k, coincidentally. Anybody from those fdts, should have learned about strange sensor patterns. The only strange ones I support (the random one is said to have higher quality) is rygcbw primary complimentary colour pattern, which gets you a bit more of the visual colour space. I had to stop the video to write, but imagine that the colour changes in keying resolutions is related to the sensor pattern. Now, at 5:1 there is still problems and less quality than the Q modes. Unlike you saw, I can see a clear enhancement in image at each Q level down to the bottom. It's a natural enhancement. If you look at the paint on the wall behind the fireplace just near the corner, you see a more visually pleasing increase in quality of the flaking paint the higher the Q level quality. It's more sharp and contrasty, and audiences respond more to this. I have to see how it went on hair and faces, as that's one where people will notice more. If you look at the shaded eye area in the dark, it's not near lossless,band is clearly better noise pattern at higher quality. The JVC HD1/HD10 was a fun camera despite its limitations in the day, and this is a much more capable fun looking image. I've picked out the bug in the original Braw, detail retention issue with some fine things like light coloured fishing nets and patterns, and skin tones on this originally. But, I think the resultant sombre images this produced, were rather nice, and despite the low reported dynamic range compared to the Red 8k (I forget which one) I've seen it do better in real time dynamic range than the Red, as the top CMOS performance often is similar between different sensors, just the dynamic range is measured differently, with BM's figures a lot more conservative. However I suggested techniques on how to better recover range from the noise floor, recover image detail from surrounding frames, and using surrounding pixels to better estimate missing primaries, at the pixel and calculate missing dynamic range and stretch up the shadows and down the regenerated highlights, giving extra stops in the middle, more tonality, and natural looking image. A whole suite of things, including colour correction and restoration on monochromatic lights. So, this thing is sticking it to Red a bit, despite it's sensor being pretty good. It looks like a less colourful version of the Alexa or 4.6k when used without pushing the colour. So, what are we seeing in Q, repair. Hopefully calculated averaging, and use of ques from the surrounding pixels and frames, and not simple averaging progression. The problem you get in this industry is people who use formular rather than pattern. Formula will try to simplistically calculate something, where pattern dealing is more procedural decision making, and you have to work at it. A simple formula will try to guess what the value will be (like a blind folded person trying to throw a dart at a dart board), where with decision based procedure, it takes off the blind fold, tries to work out what's happening and what to do, then adapt or select the formula. A lot of people that lile to debate(/argue) in forums and and be negative on people with intelligence, are like this sort of person having a go at the previous sort, and they often gang up, the blind leading the blind, so to speak. Hopefully, we are seeing a faithful image reconstruction and levels adjustment as I've advocated over the last 20 years. But, you would have to test that with a good monochrome camera still comparison (not over with a low fill sensor) to see if the structure and relative levels through the areas, line up. Then use red green and blue matching spectral filters on the monochrome camera to see how well the relative levels of each primary follows the 12k across the image. If they are a lot different it maybe obviously they are some simple averaging, if they are a lot closer, they might be intelligent procedural pattern processing. Thank you. You have done a good job.

Tolles Video, dass du Deutsch als Muttersprache hast erkennt man in deinem Englisch sehr gut 😬✌🏻

Thank You!

Wonderful. That first frame looking better was a shocker. I have mixed feelings about the 8K and 4K because of those different colors. I just got a Pocket 6k Pro and also have an FX9, 6, 3. I will try a test in ACES to see which Red, Green, and Blue are the most similar to the 12k or the 8k. (unless you beat me to it)

@deadendboredom

Жыл бұрын

I didn’t show it in the video, but I tried to adjust the colors of the different resolutions with the help of a colorchecker and even after that the blue wasn’t completely the same. My FX6 looks more like 12K in terms of colors, in the video you can see a small clip at 13:38 and there the blue looks like pure blue - but since it didn’t look so blue to the naked eye, I’m really wondering if my light has a much wider spectrum than most cameras can record. That leads to the question to what extent RGBACL is better than RGBWW when the image arrives at the viewer After beeing recorded through the RGB sensor and then shown through an RGB monitor. You are welcome to pre-empt me with the test, as I currently do not have the technical possibilities to take a closer look. P.s. I would be interested to know if the grain and rolling shutter at 12K bothers you compared to 4K.

@elvisripley

Жыл бұрын

@@deadendboredom I have shot a lot which is why I appreciate you making these tests I was getting suspicious about some things that you tested. 1. sometimes clips don't record because of what looks like a card datarate failure when using the Q settings but by looking at the time remaining on the card and your find that the first few frames look much higher quality, it seems like it takes the 12k a little bit to settle into the needed data rate. Sometimes it is too much for the card in the first couple seconds. If that happens I stared closing down the aperture so the first few moments are black and then opening back up. Nice to see your confirmation. 2. those colors. I got annoyed on a commercial where all the 12k was 24 and there was also 8k/120. The talent had this blue shirt that of course was different. Now that I know I would just shoot 8k. I like the colors in 12k but don't dislike 8k. At Q5 I don't find much difference and sometimes 12k is smaller than 8k for my real world shots. But that difference in color mid shoot isn't good. 3. I am in post production on a doc I shot that was going to be all FX9 but then I got an Ursa and thought I should try it out. I shot some extra interviews and all b-roll with it and all 12k because it seemed like it looked the same as the FX9. The noise is never annoying to me and I like the ability to have extra detail if needed for reframing or messing around. In Resolve I used Smoother for my scaling setting and that gave a pleasant look. I have done a test with an Alexa LF, Canon C300 III, FX9, FX6, FS7, Raptor, Komodo, shooting a scene and so I know the FX9 and 6's ACES look match the Alexa colors. Actually they all pretty much matched except for the Raptor and Komodo which were both a little different but in different ways, FS7 didn't have great deep reds.

@deadendboredom

Жыл бұрын

Thank you for your answer, interesting observation and solution with the aperture at the beginning. I have to try this out, although I often have the problem in 4K Q5 and also other quality levels that are normaly not a problem. However, only with external SSDs. I have to admit that I have always worked color managed so far, I just loaded the scenes again with ACES and had to find the following: The color difference with the resolutions is no longer quite as strong but it is more visible when the blue is more powerfull. However, you can overall still see the differences. I also find what you said about the Sony cameras very interesting and can confirm to my astonishment that the colors in ACES are actually very similar to BlackMagic. It is also interesting that the Color Managed Work Flow sometimes produces significantly different results (Sony). Thank you for the impetus to give Aces another chance. This may lead a little too far, but I would be very interested in your ACES workflow if you pass material back and forth between different programs.

@elvisripley

Жыл бұрын

​@@deadendboredom I have found the ACES transforms to be consistent across cameras and that is the main reason for using it. I did the same test I mentioned with all the cameras using the DaVinci color management. Using DaVinci each camera looked good, sometimes better than the ACES workflow, but they didn't all match coming in. Also the ACES outputs were more consistent. If I have a Rec709 output and switch to P3 or Rec2020 or HDR I feel like I don't have to regrade. With DaVinci I didn't feel the same. They all looked good but not exacty the same. In tiny differences. I recently did VFX on a series where it was all shot on Sony cameras and they were using S-Log3/S-Gamut3.cine for their working space. The editor used camera files and a LUT so I sent back the same Sony color space. I knew the ACES would output would be identical so I didn't have any doubts about it. I have done some heavier 16bit EXR image sequences but not often because people usually want the files upstream in editorial and prores with a lut is much easier to deal with. On another recent show I was VFX Supervisor and in general in charge of post so I could have picked a more "right" workflow but on that one it was still prores with a lut because it is good enough and the ACES input and output transformers are reversible so S-Log3 in matched the S-Log3 out. Then if people are working they can have freedom to use ACEScct or ACEScc or ACEScg but they know that S-Log3/S-Gamut3.cine comes in and goes out. Also I always use S-Gamut3 but on both projects mentioned something had been shot before I could get the information to the DP so the projects became 3.cine. 3.cine has so much outside the visible spectrum which might (probably won't) lead to virtual colors. The interesting thing is that S-Gamut3.cine is kind of in like with the old Alexa Wide Gamut 3 but the new Wide Gamut 4 is almost exactly S-Gamut3. They do a good job at pitching their new color so hopefuly I can push more DPs into the S-Gamut3 zone. www.arri.com/en/learn-help/learn-help-camera-system/image-science/reveal-color-science#tab-301418

Gehe meist den entgegengesetzten Weg. In dunklen Szenen, z.b. indoor night, gehe ich mit iso runter auf 400, oder noch weiter runter, um ein cleanes bild zu bekommen. Natürlich alles mit entsprechender Beleuchtung, was für diese cam auch unerlässlich ist. Das ist eine reine Cinema-Kamera !!! In hellen Szenen, z.b. commercial high key ohne dunkle Bereiche auf 1000. Und nur in Wechselreichen Szenen mit dunklen und hellen Bereichen, z.b. urban bei Tag mit Licht und Schatten benutze ich Native iso 800. Du achst dir echt Gedanken. Super Videos 👍

@deadendboredom

7 ай бұрын

Danke - und ich bin ganz bei dir was die entgegengesetzte ISO angeht, die ISO ist ja ein oft missverstandenes Thema.

I always thought downscaling in post was preferable than in camera, but the results in this video show otherwise? Interesting. I guess if a delivery resolution higher than 4K or potential cropping of the image in post isn't needed, you get the best overall image quality shooting downscaled-4K with this camera (with constant quality)?

@deadendboredom

Жыл бұрын

Yes, that’s why I personally record almost exclusively in 4K and Q5 with the camera. Of course this is also a matter of taste, because I like a slightly soft, analog and natural look. But I was surprised as well after I had done the tests, because I previously thought that other settings were more optimal. But this is one thing that is probably not good to be taken as a rule to other cameras. What blackmagic does in this camera with its special sensor is, as far as I know, no simple downscling. And I think that’s the real lesson from this and other tests I’ve done: each camera is different, every sensor has advantages and disadvantages and in the rarest cases you can infer from one thing to another (Even though you believe that). usually there is a peculiarity somewhere that you didn’t think of and that makes a Big difference. It may look different with the new OLPF version of the URSA 12K.

Great video, super useful! a little off topic- but how do you find 4K ProRes files via the Video assist to compare to the internal raw?

@deadendboredom

Жыл бұрын

This is a very interesting question! Just like in the video, you sometimes lose and sometimes win when it comes to file sizes: 10sec Clip 4K Q5 is between 770MB and 1.5GB vs 4K ProResHQ is constant at 950MB. The colors and contrasts are a little different, but all in all, the two are amazingly similar for color grading and white balance - RAW may be a bit better. The noise in ProRes it looks a little softer and more natural to my eye. The interesting part is Dynamic Range: in RAW you have more leeway in the highlight and can get back details that are lost in ProRes. But it looks different in the shadows, there seems to be a little more in ProRes. I just tried it without color management, with Davinci Color Management and once with ACES and everywhere there was a little more cut in RAW in the dark area. In ProRes the blacks seemed to be a little raised. I can’t quite explain the observation right now.

@jordankaiwright

Жыл бұрын

@@deadendboredom great to hear & thankyou for the detailed response! The issue with the 12k for me has been the lack of ProRes, but now knowing that I can get an external ProRes with my monitor makes it a much more suitable camera for me!

@TheJ_G

9 ай бұрын

@@deadendboredomDo you find there’s any difference in setting the camera to 12K vs 4K internally when recording ProRes externally? I know this isn’t a great basis to build the question off of, but I’ve noticed with other cameras that setting the internal resolution to the best/highest quality, seems to improve external recordings. Most notably on the Sony A1, where setting the cameras internal recording to 8K, and then recording 4K externally, produced a much better 4K image externally. Or would you say it doesn’t really matter for the Ursa 12K?

@deadendboredom

9 ай бұрын

@@TheJ_G Great question! In my previous tests, I've noticed that external ProRes recording can sometimes pose color issues or deviate from the internal recording. However, I didn't observe a difference in the quality of the external recording when capturing internally in 4K or 12K. For the Sony A1, there will be a difference because the sensor is read differently depending on the internal resolution (line skipping), which in turn affects the external recording. With the URSA 12K, this is not the case. Personally, I would advise against external ProRes recording unless it's absolutely necessary for a specific reason, as it can indeed lead to color and possibly dynamic range issues, and the file sizes aren't significantly smaller. Are you required to record in ProRes for a client?

@TheJ_G

9 ай бұрын

@@deadendboredomThanks for the response! I’ve done a little testing here and there, and I don’t really seem to notice a clear difference when setting the camera to a higher resolution internally - But nothing as in-depth or amazing as the work you out into this video. It’s really been tough to find anything online that covers ProRes externally on the 12K Ursa, or really anything in-depth at all on the Ursa 12k that isn’t reviews from a couple years ago. And you guessed it, I have a rather large client for the foreseeable future that only wants ProRes, haha. I definitely get that BM built this camera for BRAW, but I wish they’d open it up to ProRes as well, even with some limitations. It would just make an already polarizing camera a little more flexible and perhaps less polarizing.

Have you found drives that will record 12k 60fps at Q0? I can't seem to find anyone who has confirmed shooting at those speeds, which makes sense since it requires a drive that can write at a sustained 3.2 gb/s. I'm just wondering if there is some solution or blackmagic was just hoping one day they will release drives fast enough for the options they offer

@deadendboredom

Жыл бұрын

No - and as far as I know, this is not even limited by hard drive. 12K60 QO exceeds the bandwidth of the physical USB-C port of the camera. I quote the BM website: „The SuperSpeed USB‑C expansion port on the rear can record to USB‑C flash storage up to 900MB/s“ Blackmagic RAW @24p is specified at Q0 with 289 to 1.156 MB/s. 60p has a mathematical maximum of 2890MB/s. If you take a very static and low-detail image, then you could get by with the minimum of 722MB/s :D This also explains why there can be interruptions even at 12K 24p Q0. The scene just has to contain enough details and movement - the bitrate shoots up and exceeds the 900 MB/s. Fortunately, I can’t think of any reason why you should record Q0.

Thanks for all that work. Color shift was very interesting. 400 or 800 iso? I shoot 400 and noise really goes down. Shoot 8k 3/1 most of the time. Maybe I should go to 5/1?

@deadendboredom

Жыл бұрын

The tests were all created at ISO 800, but exposing to ISO 400 is a good idea if you can do without the 1 stop in the highlights. As far as compression is concerned, it depends on your requirements. If it is to be small, I would even go to 12:1, otherwise maybe Q5? I would be very interested to know why you record in 8K and whether there is a reason why you prefer constant bit rate (I also did that for a long time).

@Dbj5555555

Жыл бұрын

@@deadendboredom I just hate the noise. faster read out in 8k. Constant quality is what I shoot. Q3. Sorry it was not clear. Your tests look like Q5 would be ok. I just saw you’re the actor. Great job!

@Dbj5555555

Жыл бұрын

Deliver 4K or 1080. External recording in 2k. Use that when I need ProRes fast.

@deadendboredom

Жыл бұрын

@@Dbj5555555 Yes, I would go with Q5. I couldn’t see any difference between Q5 and Q3, apart from the larger files. If the noise also bothers you so much, then think about recording it in 4K - it’s not gone, but I really find it much more pleasant. The question is also: do you need the minimum plus of details of 8K? P.s. I’m the actor?

@Dbj5555555

Жыл бұрын

I noticed in another video you put out that the human subject in the clips was you.

6:55 That's the whole point of downsampling, and therefore the whole point of this 12K camera.

@deadendboredom

Жыл бұрын

Absolutely right! But before there is a misunderstanding among readers of this comment: this does not mean that you also have to record in 12K with this camera in order to benefit from the downscaling. 12K, 8K and 4K material scaled down to 4K look almost identical in terms of details. So you already benefit from the high-resolution sensor if you only record in 4K and meanwhile save hard disk space and also surprisingly get less ugly grain. I go into the topic very briefly at 11:55

@SP95

Жыл бұрын

They looked almost identical because it was the wrong way of doing downscaling, the short mention later in the video is using the same bitrate for all three resolution which is always worth trying but fairly off topic when it comes to downscales. It is supposed to be recorded with the best bitrate reasonally possible from a 12k recording to then "squeeze" this huge file back to a 4k video (at a lower bitrate than the 12k rush) in post. Canon does this internally with that "4kHQ" recording mode and the difference compared to native 4k is significant because it comes straight from an internally downsampled 8k video.

@deadendboredom

Жыл бұрын

@@SP95 I find the topic very interesting, but I can’t quite follow your reasoning. In order for us to be on the same basis, it must be clear that I am Talking about the URSA 12K. Other cameras handle their resolutions and recording methods differently (line skipping, binning, etc.). So when you talk about a Canon camera that in one mode uses over sampling and in the other mode line skipping or binning, then of course it is clear that the result is different. To my knowledge, the URSA 12K uses a sensor technology that Kind of samples the complete sensor down to 8K and 4K and I have been able to see this in all my tests: whether 12K, 8K or 4K, in a 4K timeline all show comparable details. If you have an idea for a concrete test setup that shows other results: bring it on, I’m happy to test it!

What media are you recording to? If using an SSD, what USB-C cable (Thunderbolt 3?) are you using?

@deadendboredom

Жыл бұрын

I use both CFast 2.0 and SSDs. I always use CFast cards when the bit rate is enough and I don’t have to record high frame rates. SSD only when it is necessary, as the connection via cable is a little more unwieldy. Among other things, I use the Samsung T7 with the supplied USB-C cable. Special Thunderbolt cables should not be necessary. I have also tried other SSD variants, including m.2 SSD in an external housing, but there the results were rather mixed and I would not recommend that.

@odaimanea

9 ай бұрын

Thank you, but what’s the best thing to record best at 12k?

@deadendboredom

9 ай бұрын

You can’t go wrong with a Samsung T7

Thanks for your video but i have a question who can seems stupid. Why using a 12k camera if it's for using it in 4k with the lower quality? I mean, i understand your test and that you find that Q5 4K was the more balance but why don't using directly a 4K camera like the BMPCC 4K or the KINEFINITY TERRA 4K? Do you think the Q5 4K on the ursa will perfrom better? Or its the ergonomic of the camera that you like? Many thanks

@deadendboredom

8 ай бұрын

There can be various reasons for this, but the underlying idea is indeed correct; for most individuals, a standard 4K camera is sufficient. A crucial factor is the sensor's visual characteristics. Different cameras exhibit distinct looks and, for example, varying dynamic ranges. The Pocket Camera 4K provides an entirely different image, dynamic range, sensor size, and more. Another factor is the ability to capture in higher resolutions, coupled with the camera's capacity to record, for instance, 4K at 240 frames per second. Ergonomics naturally also play a role, including the capability to change the lens mount.

@boogeyb

8 ай бұрын

@@deadendboredom Thank you for your answer. Have you noticed variance in dynamic range with the ursa between 12k and 4k which could impact the quality of the picture?

@deadendboredom

8 ай бұрын

@@boogeyb No, I haven't noticed any difference in dynamic range between the 4K and the 12K URSA. By the way, I forgot to mention that rolling shutter is another factor, and in comparison, the URSA performs very well in that aspect.

@boogeyb

8 ай бұрын

@@deadendboredom Thanks again. Definitively tempted by this cam. Maybe the new OPLF will behave a bit differently tho

@deadendboredom

8 ай бұрын

@@boogeyb Is there a specific reason why the camera tempts you?

Did you make a black balance after each shot?

@deadendboredom

Жыл бұрын

No, i did not. To what extent do you think this could have an impact on the results?

@odaimanea

Жыл бұрын

You made a great and very important video, I hope you do the Calibrate Sensor after shooting each of the models that shot in front of the lens. I hope you make an extra video to benefit from all of them. best wishes

@deadendboredom

Жыл бұрын

@@odaimanea I’m afraid that won’t be possible, because fun fact: The URSA 12K no longer has the possibility to perform a black calibration. Allegedly, this is no longer necessary with the new sensor and the possibility is no longer available anywhere in the menu.

What iso and f stop did you shoot at?

@deadendboredom

Жыл бұрын

Hi, the iso was set to the native 800 during the test and the f stop was slightly higher between 4.0 and 5.6 To better assess sharpness and details.

@bedenegative6450

Жыл бұрын

@@deadendboredom I'd be interested to see the false colour of the shot. The camera doesn't seem to do very well in low light, don't forget that 5.6 is pushing the evolope for 12k becuase of defraction. I wonder how the olpf will do.

@deadendboredom

Жыл бұрын

@@bedenegative6450 Unfortunately, I can’t serve with the falsecolor image, but I can tell you that all shots were exposed with an exposure meter. You’re right, the camera has weaknesses in low light and generally in darker areas (Compared to some other cameras). I have in mind that defraction occurs faster at high resolutions, but now that you say it, I should perhaps test when it Starts to be actually visible at 12K. What would you suspect? I would also be very interested in the real differences to the new OLPF version.

@bedenegative6450

Жыл бұрын

@@deadendboredom hard to say... I always operate under the assumption of adding some noise reduction in post with blackmagic. I can see a lot of colour noise which surprises me tbh...

@deadendboredom

Жыл бұрын

@@bedenegative6450 The strong colored noise also surprised me and disappointed me a little, especially in the high resolution. That’s the main reason why I record almost exclusively in 4K.

A cheese something