How Clever is Sony’s Noise Reduction?

Odyssey-Price-Promo-Banner-01-21-2016 How Clever is Sony's Noise Reduction?

Following on from the ongoing discussions about the way the noise reduction on the PXW-FS5 works I’d like to try to answer a few points that have been raised.

Noise reduction is incorporated in to almost every single video camera on the market today. Without it cameras would not be achieving the sensitivity levels that we are becoming accustomed to. Big sensors have helped increase sensitivity, but noise reduction also plays a major part. 4 years ago the typical sensitivity of a video camera was around the equivalent of 300 ISO. Today 1000 ISO is common. The laws of physics have not changed but image processing has.

However image processing has limitations. Noise reduction is a trade off as it typically introduces other artefacts including blurring or softening of the image, smearing of the image when there is motion, posterisation, blockiness or banding.

The more noise you introduce into the image the harder the NR has to work and as a result the worse, or more noticeable the artefacts will become.

Every time you add 6dB of gain (or double the ISO) of a sensor you will also double the underlying noise level, that is a very significant noise increase.

Different sensors will exhibit different noise characteristics. A sensor with a very high pixel count may exhibit finer grain, but may be less sensitive, have aliasing issues, skew issues or heat problems. A sensor with fewer pixels may have coarser grain but be more sensitive have fewer aliasing issues, fewer skew issues and be less prone to overheating. A sensor with an excess of pixels compared to the final image resolution may be easier to noise reduce as the excess sensor resolution can be used during the NR process without degrading the end image, but the high pixel count may introduce some of the other issues listed above.

So to get an image with an acceptable noise level, low skew, a desirable level of sensitivity without overheating is a very fine balancing act. Different cameras will perform differently and each will have strength’s and weaknesses. A Sony A7S has big pixels, so it’s very sensitive and only needs low levels of NR at the base sensitivity compared to say an FS7 with it’s smaller pixels. But the A7S suffers from a lot of image skew that is unacceptable for many types of video production plus there are heat issues when shooting for long periods. Meanwhile the FS7, while less sensitive has very, very low skew levels and no overheating.

But what about the new PXW-FS5? There is much discussion about the noise reduction in this camera. First of all it is a different camera to any other. So it will not perform the same as any other. It uses a different combination of sensor and processing to any other camera on the market.

Does it perform as well as an A7R in low light? Actually it’s not all that different in terms of base sensitivity. What about if you add gain? Well the image quality of both cameras deteriorates when you add gain. Both become noisier and both have more NR artefacts. The A7RII is possibly a better performer in terms of noise and artefacts, maybe due to the higher pixel count allowing the use of a lot of pixel binning. Does this mean that the FS5 is somehow broken or defective? No, it just means that the A7RII has the edge in low light. Do remember that the A7RII suffers from a lot of image skew, really appalling flash band issues and overheats on long shots. Does that mean the A7RII is broken, because after all the FS5 doesn’t have these issues problems. Of course not, these are just limitations of these two very different cameras.

Anyway coming back to the noise reduction. As we can’t actually turn off the NR on the FS5 it’s hard to understand exactly how beneficial it is, even though it does introduce some inevitable artefacts (NR will always have undesirable side effects), especially when you add more than 6dB of gain. But we can get some clues. The edge tearing artefact that can be see on vertical motion at high gain levels appears to simply be noise leaking through the temporal NR when there is a lot of vertical motion, and it’s pretty obvious that there’s actually a lot of noise being hidden.

Why does this only occur in 4K? Well that’s probably because in HD the camera has a surplus of sensor resolution, so different noise reduction processes that soften the image can be used instead of temporal NR as the softening won’t be noticed in HD.

Another clue as to how well the NR works is that if you quickly switch the gain switch from one gain level to another it takes a couple of frames for the temporal NR to catch up so for one or two frames the temporal NR level is reduced (not eliminated just reduced a bit). If you look at these frames it gives some insight into how amazingly effective the temporal NR is and as well as temporal NR the FS5 is also employing spatial NR, so this is only part of the story.

I have a couple of frame grabs. One is a frame where the temporal NR is stabilised and doing it’s thing reducing the noise in the picture. This is with +12 dB of gain applied and to be honest it’s a pretty clean looking image for a 4000 ISO shot, imagine adding +24dB gain to an EX1 or PMW-500 to get up to 4000 ISO and what that would look like! The other frame is a frame grabbed as the camera is switching between +6dB to +12dB, so what you are seeing is about 6dB’s worth of temporal noise reduction. Just look at all that very nasty looking noise. Notice the very blocky areas in the shadows, in motion these can be seen to be fluttering from frame to frame as the NR kicks in, it’s really nasty. The noise is having a serious impact on the image resolution. On the far left table leg in the close up you can no longer see the wood grain. Now imagine what at least double if not 3 times that would look like because if you were to shoot at 3200 ISO with the standard gammas without NR, that’s what it’s going to look like. It would almost certainly overload the internal coded resulting in compression artefacts that you won’t see while shooting, only when you play the footage back when quite possibly it’s too late to do anything about it.

So next time you look at a noise reduction artefact do think about just how dreadful the image would look without any NR. We are not talking about seeing a little bit more noise and grain with reduced NR but a blizzard of noise and grain. It’s also worth remembering that all the camera manufacturers are doing this. Also consider what a typical 1/2″ or 2/3″ broadcast camera would look like with +24dB of gain added, it really would be unusable, yet the FS5 can deliver an image at 4000ISO that is really not all that bad. It’s not perfect, it does have some artefacts but really it’s quite remarkable what a camera like this can do thanks to modern noise reduction processes and large sensors.

Noise-reduced-small How Clever is Sony's Noise Reduction?
Stable noise reduction at +12dB gain.
Noise-evident-small How Clever is Sony's Noise Reduction?
Frame grab between +6dB and +12dB gain before the temporal noise reduction has fully kicked in.
Noise-reduced-1024x576 How Clever is Sony's Noise Reduction?
FS5 at +12dB gain.
Noise-evident-1024x576 How Clever is Sony's Noise Reduction?
Single frame during the transition from +6dB gain to +12dB gain before the noise reduction has fully kicked in. Notice all the extra noise, now at least double this if you want an idea of how much noise the NR is dealing with when shooting at 3200 ISO with a standard gamma.

Temporal NR: Temporal noise reduction works by taking a single pixel and measuring it’s average output over several frames. Using this average value a correction can be applied to the current frame that helps reduce the instantaneous brightness fluctuations that are seen as noise. However if there is motion in that area this can create nasty local blurring or smearing effects. On option is to include a process that detects motion and locally reduces or eliminates the noise reduction during motion to counter the smear or blurring. Either way motion in the image introduces undesirable artefacts.

 

31 thoughts on “How Clever is Sony’s Noise Reduction?”

  1. Hi Alister. I don’t want to sustain this debate unnecessary. But that is probably the point that needs discussing. Do we (professional and semi-professional) users need a over-agressive NR system that creates heavy artifacts in some situations or do we prefer a more conservative system that leaves it to the user if and how he deals with the noise created by higher gain settings. I guaranty you, most of us prefer more noise and less artifacts. For the few that actually prefer a strong in-camera NR Sony could easily add an option to control the amount of NR. There are already dozens of sometimes pretty obscure parameters to play with. Why not add this one? I would even take it as a payed upgrade…

    For the readers that don’t get the point looking at the above stills: The NR on the FS5 works beautifully in a steady image. It does on the other hand tear soft horizontal lines in vertical movements apart. The effect is rather dramatic.

    1. Absolutely. Why have it baked into the image in-camera when we can apply it more effectively in post?

      Also, I like a fine film-like grain in my low light images.

      Also, this grain helps dither between bands of different luminosity in 8bit, making for the appearance of smoother gradation.

      Additionally the grain helps the codec avoid macro blocking on areas lacking in details such as the sky, because noise itself is detail and it forces the codec to allocate more of the available file space / data rate to those areas.

      I personally think Sony should very strongly consider flicking the off switch and giving us said “off switch” in the FS5’s menus!

      1. I agree completely. I would love to see a firmware option for NR amount: “low”, “med” and “high”.

        Even better, variable amount: 1-5.

        I think that alone will stop all this FS5 anxiety on the web Sony is seeing in the past 10 days.

        I can fix noise in post pretty easily. I cant fix ragged edges. Maybe Industrial Light & Magic can,…but I cant.

        1. Alister….let me throw something at you: In the samples I have been seeing, I noticed that the BLACK colored edges seem to tear more than any colored edge?

          Are you seeing that too? (Anybody?)

          Could this NR circuit “issue” be coming from a luma channel and not the RGB chroma channels? Could a NR algorithm be processing them independently?

          Curious….if you strip the color and inspect a black and white image, how do you see it then?

          Is the tearing stronger?

          Is the chroma covering up the tearing?

          Is it possible that the chroma channels are clean? (or cleaner?)

          Dont killl me,..I’m just throwing out wild guesses here.

          CT

      2. You complain about the weakness of the codec and artefacts in the already stressed codec and then suggest that the solution is to stress the codec still further?

        If you put more noise into the codec you will have not only the original noise to deal with but also a ton of extra compression artefacts to deal with.

        Don’t forget: gain does not make a camera more sensitive, it just makes the image brighter. Why do we normally prefer to add gain in camera and not in post? Because it’s almost always preferable to do the image processing before the weakest link which is the codec. Furthermore it must be remembered that the H264/XAVC codec always includes a noise reduction process, so some NR will always be present.

        You do have the option to shoot with no gain and add extra gain and NR in post. The majority of the problems people see with this cameras NR are only when using more than 6db gain.

        1. The ragged edges are predominantly just edges in an other wise clean image where you can now see the noise again, the noise making them look ragged, when everything around them is still largely noise free. I don’t think most people comprehend just how much noise is getting taken out of the images.

  2. I’m very curious to see what these ragged edges look like just before the NR circuit locks on to it. You say that it’s about a 2 frame delay?

    Is there any way to shoot an a contrasty edge that you know will distort, shoot it with light motion and a fast shutter speed. You can switch a the gain a few times from +9-+12db and hopefully spot what the frame looks like just before it gets hit by the NR algorithm. Y’now, a “before and after” A/B inspection.

    I know that all cameras, sensors are different and it’s not fair to reasonable to compare them against each other. I do agree with you on that Alister.

    I’m just sooo glad I’m not seeing this issue on any of my other Sony cameras. I’m perfectly happy to deal with the “regular” noise reduction problems on those models. 😉

  3. But the NR is actually surprisingly good in most normal shooting situations. Most motion within a video frame is horizontal. Pans are more common than tilts, objects passing through the frame etc are all most common in the horizontal axis. So I suspect Sony have tuned the NR to allow for this. The end result is some very effective NR that holds together well even when you pan the camera or something passes horizontally through the shot. The NR is aggressive because the amount of noise being added is huge.

    The FS5 is not a super low light machine. The A7S is better, but that has issues of it’s own such as overheating and a bucket load of image skew no matter what the gain.

    If you reduced the NR you would almost certainly overload the codec and run into a whole new world of issues as you move the artefacts from the NR process to the compression process (and you will still have artefacts, they are not going to go away). Only now you won’t see them while your shooting, your only going to see them later in post when it’s too late to do anything about it, a very dangerous situation indeed. Why is it so hard for people to understand that this is a very delicate balancing act and you can’t remove something from one side of the scales without it affecting the balance. Maybe Sony will allow the NR to be reduced in a later firmware release, but again don’t expect miracles, Sony can only change the firmware, they can’t change the laws of physics. The NR is masking one hell of a lot of noise.

    It also should not be forgotten that there is always the option of shooting at 0dB then adding any additional gain and NR in post. Remember gain does not make a camera more sensitive, all it does is make the image brighter, sensitivity is governed by the sensors efficiency at converting light into electrons and gain does not change that. So there is nothing stopping you from adding the extra gain in post and doing your own NR. I doubt it will look any better.

    It’s very sad that these days the performance of a camera is judged by “lets see how bad we can make it look” rather then “lets see how good we can make it look”.

    1. I’d rather have the option reduce the NR, get more detail and fewer weird artifacts and to bring in more noise as a consequence. I’m a big “Neat Video” user in post. As far as codec being overloaded, I’m sure ProRes can handle anything that XAVC-L cant. (PIX E5 and Shogun)

      Speaking of XAVC Long GOP. I must say the h.264 100mbp/s in the .mp4 wrapper (XAVC-S) that the A7S-II and RX10-II uses is pretty decent. It seems to handle the noise that the little RX10-II produces pretty well. It doesn’t block up (macro-block) or freak out when that image gets gritty.

      I’m actually very impressed with the little RX10-II in high gain. It’s SLOG-2 forces 800 ISO but that codec seems to hold up OK when you stretch it back to rec709 levels.

      I have no complaints about the noise reduction on the RX10-II or even the X70. I think they really got the NR algorithms right on those 1inch-type cameras. (Of course the A7s-II is nothing to ever complain about…that camera is beautiful even with his rolling shutter)

      I’ll be testing the FS5 on my own in a short while. I will post comparisons with all of them together soon. (will also pitch the FS5 against the FS700’s 4k raw to ProRes too)

      CT

      1. Sorry but the X70 suffers from a whole host of NR issues. It’s very smeary has a lot of fixed pattern noise that smears across the image and the resolution drops when you add gain. The RX10 and X70 (like the A7RII) are able to make more use of spacial NR as they are oversampled in all modes which does give them an NR advantage. Something the FS5 can’t do in 4K.
        Lets also not forget that the X70 has nowhere near the dynamic range of the FS5. These means that Sony probably have a lot more flexibility as to where they choose to place the 0dB point as they don’t have to cater for a big highlight range. This means they can focus much more on getting the best noise performance from the sensor.

        As you increase the dynamic range the window over which you can operate the sensor becomes much narrower, sometimes forcing you into uncomfortable noise compromises. It’s far easier to make a low noise camera when you only have a narrow dynamic range. For every stop less of DR you wish to capture you have around a 6dB noise advantage.
        Consider that at 0dB if you shoot the same scene with S-log and then with a standard gamma and expose both at the correct levels, after grading the S-log will look noisier because you have to record the mid range in a lower, nosier range to make room for the extended highlight range. It’s the same sensor, same camera, same gain, but when you go to a wider DR you get forced into a less than ideal corner.

        It is likely that Sony could re-engineer the FS5 so that the base ISO’s for the cinegammas and standard gamma could be lower, maybe 500 ISO (effectively -6dB) as we don’t need to utilise the full sensor range. This would give an even cleaner image at 0dB (although I feel it’s already acceptable at 0dB). BUT it wouldn’t make the issues go away because so many people would just crank up the gain agin, back up to 3200 ISO so they can shoot in dark holes and it would look just the same as it does now at 3200 ISO.

        There are all kinds of things at play here. Light doesn’t travel in beautiful smooth waves, it travels in little randomised packets and burst. As light levels drop the randomness in these packets start to become visible and the very light we are trying to capture has a degree of noise built in.

        1. Speaking of the dynamic range of that 1 inch-type RX10 and X70 sensor. The RX10-II is NOT a low dynamic range camera. I compared it’s SLOG-2 vs. the A7s (original model) in a crazy high dynamic range scene and the two cameras were roughly the same in DR! I almost feel out of my chair. The noise difference was big but the dynamic range and detail were the same. That little RX10-II with SLOG-2 captured the exact amount of shadow information as the A7s. (RX10-II wide open, 1/30 shutter, A7 much high shutter speed to equal the shots)

          You can see the shocking test results here:

          http://www.dvinfo.net/forum/sony-rx-cybershots-cx-series-camcorders/529250-rx10-ii-matches-a7s-dynamic-range.html

          What do you think? Sony is now getting ALLOT of DR out of tiny image sensors.

          So this would put that little 1 inch-type sensor at the same 12 stops that the A7s had. At least between highlights and shadows. That light was REALLY bright. The RX10-II was at 800 ISO and A7S at 3200 ISO.

          For the FS5, I’m complain less about the codec and more about having a better HDMI output. When I dont care, I’ll shoot XAVC-L. When I care, I’ll shoot ProRes externally. The additional noise wont hurt ProRes.

          Anyway, I realize that everything I have seen in the past two weeks is stuff that other people have shot. I can’t put too much faith into that right now. I really need to shoot stuff myself and judge from my own tests. I’ll spend maybe two weeks comparing it to the RX10-II, X70, EX1r, A7S-II and FS700r. (all extremely different cameras) Who knows? Maybe I will love it as it is today. Maybe I will cry and moan about wanting variable NR firmware settings. But if it’s overall UHD quality sits at the bottom of the pile?…it’s probably going back to the store. I really Do expect to keep it though. That electronic variable ND is a pretty damn attractive component.

          On the FS5, I’ll shut up for a while now! 😉

          CT

          1. Sorry to disappoint you but in the examples you have linked to there is a least a 2 stop DR difference between the RX10 and A7 example, you can see a lot, lot more shadow detail in the A7 image, they are not even close. It’s worth noting that the bottom one or two stops of DR are hard to pick out even with a correctly set up test chart such as a Zyla, so if you can clearly see more picture information in the shadows of a couple of pictures like this, that’s a least a 2, more likely 3 stop difference. I don’t have an RX10II but when I did have the X70 the most DR we could squeeze out of it was 10.5 stops, which is actually a fairly respectable DR for such small pixels and conventional gamma.

            While I do fully understand the desire to test camera against camera and in some cases it is useful to compare one against another benchmark. These kinds of tests typically tend to be good at highlighting deficiencies but less good at finding benefits, unless they are conducted extremely carefully under tightly controlled conditions. People will readily spot a problem area and then become focussed on that one problem and then forget to look at the bigger picture.

            There are currently tests circulating that talk almost exclusively about the high gain tearing issue of the FS5, comparing it with other cameras, but fail to also note that overall the noise increase as you raise the gain is not proportional to the gain added. IE, add 6dB gain but the image does not become twice as noisy.

            I think it is quite possible that you will be disappointed in the FS5 if your key role for it is to shoot in the dark. The A7SII is very much the king of that arena. But as an all round camera with great ergonomics, very useful sensitivity (lets not forget that 1000 ISO base is pretty sensitive anyway and the “issues” don’t really start to show up until you’ve added at least 6dB of gain). Low cost media, lightweight, excellent battery life, streaming etc all add up to a very capable camera that can make some really beautiful pictures.

          2. Wow, your eyes are definitely more fined tuned for dynamic range than mine. I see them pretty darn close on 4 different monitors. The noise difference is certainly dramatic.

            The biggest point in the RX10-II vs A7s was the “David and Goliath” fight. Goliath obviously wins but little, tiny and heavily disadvantaged David does deliver a very noble performance.

            The X70? you have that little camera a break and cut it some slack. It’s sensor is small. It’s pretty amazing that it’s not as noisy as it “should” be for what it is.

            I dont at all expect the FS5 to go toe to toe with my A7s-II in low light. That’s another unrealistic “David and Goliath” battle. I’m buying the FS5 for it’s ergonomics, convenience and cool ND filter. I want to roll UHD for 2 hours straight without needing my external recorder. (which will be glued to my A7s at the time) The FS700r needs my Shogun for 4k, that’s a drag to setup. So the FS5 fit’s a good general purpose roll. My X70 is nice all around little guy but I want my FS5 to be more substantial. I do have 12 lenses that the FS5 will make good use of too.

            My noise expectations are very “Super35 sensor” grounded. Nothing more, nothing less. I do certainly expect the FS5 8bit HDMI to look similar to the FS7’s 8bit HDMI given that they have the same sensor. I dont want to see Sony artificially “dumb down” it’s 8bit output more than necessary. I HATE when companies do that kind stuff for “business reasons”.

            Alister, there is a 90% I will keep the FS5 even with it’s flaws. (Like I do with my other Sony cameras)

  4. Just out of curiosity, what effect, if any, will the FS5’s forthcoming ability to output RAW images have on this issue? If the NR is applied is upstream of the codec, presumably none.

    1. I wish I could answer that with certainty, but I can’t. The NR is upstream of the codec, but normally with raw things like NR are bypassed. The camera simply takes the de-bayered signal from the sensor, bypassing any gain, gamma or NR stages and passes it to the raw output.

      So fingers crossed, it should by pass everything.

  5. I was thinking that too. FS5 12bit raw will give you true 10bit ProRes SLOG-3. However, you better have “Neat Video” installed. You will need it.

    Thankfully, FS700 raw works well with the Shogun.

    I still have high hopes that Sony will make on-board NR firmware changes that will get this FS5 to look more like other Sony cameras do today with +6db, +9b or +12db settings. (Yes, even though it’s a very different camera than all the other Sony models)

    CT

  6. A little noise is easily cleaned up in post, and some add noise in post to emulate film grain. NOISE is not the problem here.

    As stated above NR works on a steady image but tears soft horizontal lines in vertical movements apart and the effect is rather dramatic – this is the ‘PROBLEM’ users are trying to communicate.

  7. Apparently edge tearing has been observed in the Cinegammas and Japan was indeed notified. Japan just returned to work on Tuesday after break.

    Good its being looked into 🙂

  8. Thanks Alister for the good writing and sound advice.

    You mentioned that “A sensor with a very high pixel count may exhibit finer grain, but may be less sensitive, have aliasing issues, skew issues or heat problems. A sensor with fewer pixels may have coarser grain but be more sensitive have fewer aliasing issues, fewer skew issues and be less prone to overheating.”

    With respect to aliasing, isn’t the other way around? I would tend to think that a sensor with a very high pixel count may exhibit less aliasing issues compared with one with fewer pixels, since the former is capable of adequately resolving the fine details of a potential scene (this assuming a full read out of the sensors).

    1. It depends on what the camera has been set up for. A DSLR set up for stills will have an excess of pixels when shooting 4K or HD video and if the downsampling isn’t done extremely well it will alias. Even a full pixel readout may alias if any form of pixel binning is used to drop the resolution. Excess resolution causes aliasing when you down convert to a lower resolution unless you have very good processing. The best is always to carefully match the sensor to the optical filter.

      A sensor with too few pixels should not normally alias as you would normally have an optical low pass filter in front of it to reduce the resolution of the incoming light to match that of the sensor.

      1. Funny as Alister is always telling that more noise will break the h264 codec and it seems that no one is listening and everyone wants less noise reduction or no noise reduction.
        For those who are hoping to bypass the issue by going raw there is another problem:
        With Sony’s FSRAW you don’t have the full dynamic range of 14 stops as the raw is linear 12bit which is not sufficient for 14stops (the fs7 is having exactely this problem (google!) and probably the fs5 will also have, the fs700 doesnt have this problem as the dynamic range is not that high and 12bit linear is enough. But not for 14stops!)

        1. Sony use floating point math to record 14 stops with 12 bits on the FS7, FS700 and FS5. It’s the same sensor in all 3 cameras. The FS7 has slightly different processing and it wasn’t fully optimised at first. The latest firmware update has improved this.

    2. Assuming the sensors and OLPF’s are correctly designed it depends on how well the downscaling/downsampling is done. HD video down converted to SD will often be full of aliasing, similarly 4K video derived from a +20MP sensor may (or may not) have more aliasing than 4K derived from an 8MP sensor. The 8K sensor (assuming proper design) should always have minimal aliasing at 4K. High quality downscaling needs a lot of processing power, especially if it at high dynamic ranges as you need 16 bit processing for 14 stops.

  9. Greetings- Sony PXW-FS5 – The ISO/GAIN button next to white balance and shutter in manual mode will not allow me to adjust gain/ISO when pressed. I have a choice when pressing between auto and whatever it was previously set to. I can manually adjust gain / iso by pressing function 5 on the hand grip zoom then navigating joystick. Is this something Sony needs to fix? White balance and shutter works but not the iso / gain. What if I do not use hand grip then do I need to go into menu and change gain / iso? Is anyone else having this issue?

  10. Alister, since the FS7 now has low/med/high (and off) NR options in CineEI, would you be prepared to make a recommendation for NR level for shooting 1000EI in a controlled – studio lit environment (with not much post beyond the editor applying my favoured LUT) please?

  11. Presumably, if the sensor is fed with lots of light NR won’t do anything? If the sensor is fed with less light and gain is applied NR will smooth out the noise. Therefore is there any disadvantage in leaving NR set to high? Is Sony’s NR that clever?

  12. When shooting any gamma different than sLog, at 1000 iso or slightly more the image is clean. When it comes to sLog, even on a bright sunny day and by deliberately overexposing 1 – 1,5 stops the image is always noisy.
    Would it help reducing the base sensitive to 2000 iso like the Fs7?

    1. If you are finding it more noisy in log when exposing at +1.5 stops then there is something up with your workflow.

      Both 1000 ISO standard gamma and 3200 ISO S-Log are 0db gain. No added gain, so the noise levels will be the same. But if you were to use a light meter to set the exposure, 3200 ISO would give an aperture 1.3 stops more closed than 1000 ISO. Thus when you shoot at the base recommended levels, log will typically have about 9db more noise as there is less light falling on the sensor. But by shooting at +1.5 stops you will be putting a little more light on the sensor, so will have a tiny bit less noise (more accurately you have a bigger signal and this improves the signal to noise ratio, the noise of the sensor remains constant).

      How you handle the footage in post is absolutely critical. You need to de-log the material either with an S-Curve or LUT or use a workflow such as ACES to really get the best from the footage.

      1. Alister thanks for your answer. I’m sure you know what you’re talking about but I don’t understand why I should see noise everywhere in the frame even from the viewfinder when magnifying the image to check focus and why I don’t see it with any standard gamma if they are both 0db gain.
        Thanks

          1. It’s all down to exposure. You will never be entirely noise free, but the ONLY thing that changes the cameras true signal to noise ratio is how much light you put on the sensor.

Leave a Reply

Your email address will not be published. Required fields are marked *