M2pro Sensor noise

[Ashcam]

Hi there, I'm having a bit on an issue with sensor noise, I'm finding the mavic 2 pro generally to have quite a lot of active noise especially in tonally flat areas. Plain skies are dancing with noisy artifacts, I can reduce it with noise reduction but this gets rid of all the small detail unless I have it at such a low amount that it just clumps the noise up into bigger blotches.
I thought I would do some tests today with the different profiles, to see if I could reduce it, but nothing seemed to make much difference.
Have I got a particularly bad sensor or is it just what the M2Pro does?
All frames are at 200%

 

[Jedster]

I’m having the same experience. I feel like I’m getting way more noise than normal even is well lit shots. I have seen some videos on YouTube that didn’t use noise reduction to my knowledge and I’m thinking there’s no way my footage would be that clean. Have you reached out to DJI yet? I really love the drone I’m just super paranoid it’s not performing normally.

 

[CanadaDrone]

If you examine everything at 200%, of course it's going to look bad. Also, darker areas that are less exposed will always be noisier than other, more completely exposed areas of the frame. For 200% crops (the equivalent of putting your nose up to a print several feet wide) your images don't look bad to me at all.

The M2P still uses a very small sensor compared to what many of us are used to in the photography world - 1" is still a very tiny sensor, and you are going to have all the same limitations that come along with a 1" sensor. If you're used to a 14bit RAW file from a high end DSLR or something, you're going to be disappointed every time you look at M2P footage.

I suspect you will find that normal footage (not test charts) viewed at a normal magnification (nowhere near 200%) look great, and there are simply no better alternatives if you want a folding drone, so this is as good as it gets. I think you will find that nothing is wrong with your sensor - defective sensors are incredibly rare in the photography world, a bit of sensor noise at 200% would certainly not be a symptom of a defective sensor. Make sure you don't have sharpening turned up too high either, that will exaggerate noise. Also make sure you are using the histograms and exposing properly, as bright as you can without blowing highlights (use the zebras function if it helps).

 

[brettbrandon]

This is one of my big dissapointments about the M2P's camera. There is no excuse for a bunch of noise in a well lit sky. I'm used to some noise in dark areas but never in a well lit spot.

Say what you want about the size of the sensor, but I have a cheap $100.00 point and shoot that has no noise in the sky with much less of a sensor...
I don't think this camera deserves the name Hassleblad on it. Makes me wonder if Hassleblad has been bought out and will no longer be one of the best cameras available.

 

[CanadaDrone]

This is one of my big dissapointments about the M2P's camera. There is no excuse for a bunch of noise in a well lit sky. I'm used to some noise in dark areas but never in a well lit spot.

Say what you want about the size of the sensor, but I have a cheap $100.00 point and shoot that has no noise in the sky with much less of a sensor...
I don't think this camera deserves the name Hassleblad on it. Makes me wonder if Hassleblad has been bought out and will no longer be one of the best cameras available.

Can you post some controlled testing samples of your $100 P&S outperforming the 1" Sony sensor in the M2P?

If that P&S had less noise it's because it can't even shoot RAW, and the JPEGs it's outputting have lots of noise reduction added by default (which is what all of them do for the 1/2.3" sensors). You can add a bunch of NR to the M2P images too, at the expense of detail.

Also there is nothing Hasselblad about the camera hardware, it is purely marketing. DJI owns a majority stake in the company and they are milking it. The sensor is a standard 1" Sony and the lens is nothing special. Hasselblad apparently had a hand in the image processing, but considering the brand is best known for it's medium format cameras costing as much as a new car, I agree it was quite ridiculous for DJI to use the brand in the way that they have. The camera is MUCH closer to a flying Sony RX100 than it is any Hasselblad. What DJI has done is the equivalent if me buying a Ferrari badge to put on my Civic. It's actually quite common in the camera industry - Hasselblad has tried to sell rebadged Sony cameras for absurd prices, and Leica sells rebadged Panasonic cameras for absurd prices.

So far I haven't seen any examples of noisy M2P images that have been exposed properly at base ISO. If you zoom in enough you are going to see hints of noise from virtually any CMOS sensor. You aren't going to find better image quality without moving up to the Phantom 4 Pro (or higher), so there are really no alternatives anyway.

 

[Paul2660]

The 1" Sony sensor has been around a while. I don't think it's modified much and it's not a BSI (Back side illuminated) chip. So it's best at base ISO, and can't handle much push.

The Dpreview test of the RX100 IV has a good side by side comparison, and you can easily see that by 400, you are starting to get quite a bit of noise.

Sony Cyber-shot DSC RX100 VI review

The "engines" behind the P4 Pro and MP2 are different. As to how much only DJI knows for sure and they are odds are not going to say. There are slight color profiling differences between the P4 Pro and MP2, enough that I feel the processing part of each is pretty unique.

The chip in both cameras needs to be exposed to the right, as you have more highlight recover than shadow push even at base ISO. This is easy for stills, as you can bracket, but for video I agree your window of acceptable range is tight and you need to have considerable skills to be able to produce a balanced video. D-Log IMO makes it harder as you tend to see more noise in the D-log video. The P4 locks you at ISO 500 for d-log, not sure if this is the case on the MP2 as I have not used D-log on it.

Cameras being what they are, (many seem to forget the 1 2/3 sensor in the MPP and Mavic Air and now MP2 Zoom is the same sized chip that Apples uses in most of the phones they make, at least the 7, 8 and 10. Low light work with an iPhone? Even the latest iPhone IMO is weakest on low light. The 1" sensor has some advantages, but it's still small and rather limited. Everything I shoot (I mainly work with stills) is bracketed 5 shot dng and usually I can find a series that works.

The MP2 lens is it's weak link for me as I have owned 3 and still can't find one that has a good even focus across the frame, and since you are locked into 3:2 the maximum lens distortion will be seen (and there is a lot of distortion going one). I realize you are losing resolution in 4:3, but if the balance that is displayed in a 3:2 shot is ruined due to distortions and softness you might as well use a crop when taking the shot.

The other alternative now is an Inspire, with the 24MP Zenmuse. For me cost, weight and NOISE make the Inspire a no go. So I have hopes that the P5 when it's announced will have a cleaner sensor, larger sensor (20MP) is fine if they will use a larger sensor with better light gathering capabilities or use a BSI designed sensor.

I don't hold much hope in DJI adding a firmware update that will allow a better high ISO performance, as that is more than likely baked into the basic camera design.

Still love the MP2 for it's carrying weight/size and extremely low noise. I have learned ways around the lens issues to allow for images that are pleasing and hold enough sharpness.

Paul C

 

[brettbrandon]

Looks like it's Sony against Sony (and the Sony processing wins).
Here's a pic from my wifes little Sony Cybershot, and one from the M2P which is a jpg straight from the drone with it's processing, at ISO 100.
The noise in the sky is quite obvious and I can even see noise in the roof shingles. That is at 100%. The Cybershot has a little noise as well, but nowhere near as much as the M2P.
I am a photographer and went with this model because of the Hassleblad. I guess deep down inside I knew there was no way this would perform anything like a real Hassleblad. But I was hoping for at least better then what I am seeing.
I shoot in raw, but by the time you minimize the noise, you have lost all the detail...

 

[CanadaDrone]

The 1" Sony sensor has been around a while. I don't think it's modified much and it's not a BSI (Back side illuminated) chip. So it's best at base ISO, and can't handle much push.
Paul C

It is a BSI sensor but not stacked. The smaller the sensor is, the greater the benefits they get from being BSI due to the increasingly smaller real estate on the wafer available for circuitry. Being BSI or not does not determine how well a sensor can handle a base ISO shadow push. Also, every sensor is best at base ISO, no matter what it is, because that is the only time it can reach FWC.

It's actually pretty amazing for what it is and can handle a pretty significant push at base ISO - as much or more than some very expensive Canon DSLRs in fact, all the way up to full frame:

Photographic Dynamic Range versus ISO Setting

 

[CanadaDrone]

Looks like it's Sony against Sony (and the Sony processing wins).
Here's a pic from my wifes little Sony Cybershot, and one from the M2P which is a jpg straight from the drone with it's processing, at ISO 100.
The noise in the sky is quite obvious and I can even see noise in the roof shingles. That is at 100%. The Cybershot has a little noise as well, but nowhere near as much as the M2P.
I am a photographer and went with this model because of the Hassleblad. I guess deep down inside I knew there was no way this would perform anything like a real Hassleblad. But I was hoping for at least better then what I am seeing.
I shoot in raw, but by the time you minimize the noise, you have lost all the detail...

View attachment 55012 View attachment 55013

In this case, you have to compare RAW images, otherwise all you're comparing is the RAW converter in the two devices which at least on the smaller sensor, definitely includes noise reduction. The images would also have to be of the same subject matter in the same conditions and the same settings to be comparable. I'm guessing the Sony Cybershot does not allow RAW capture, but it might.

As for the Hasselblad brand, it is what I described earlier. There is unfortunately nothing Hasselblad about the camera hardware, it is purely marketing. DJI owns a majority stake in the company and they are using the name to generate sales. The sensor is a standard 1" Sony and the lens is nothing special. Hasselblad apparently had a hand in the image processing, but considering the brand is best known for it's medium format cameras costing as much as a new car, I agree it was quite ridiculous for DJI to use the brand in the way that they have. The camera is MUCH closer to a flying Sony RX100 than it is any Hasselblad. What DJI has done is the equivalent if me buying a Ferrari badge to put on my Civic. It's actually quite common in the camera industry - Hasselblad has tried to sell re-badged Sony cameras for absurd prices, and Leica sells re-badged Panasonic cameras for absurd prices.

 

[Paul2660]

It is a BSI sensor but not stacked. The smaller the sensor is, the greater the benefits they get from being BSI due to the increasingly smaller real estate on the wafer available for circuitry. Being BSI or not does not determine how well a sensor can handle a base ISO shadow push. Also, every sensor is best at base ISO, no matter what it is, because that is the only time it can reach FWC.

It's actually pretty amazing for what it is and can handle a pretty significant push at base ISO - as much or more than some very expensive Canon DSLRs in fact, all the way up to full frame:

Photographic Dynamic Range versus ISO Setting

Fully agree, was aware that the chip was BSI, good to know.

My point is that the chip is very good at base ISO, and can stand a bit of push maybe 1.5 stops. But it's very limited IMO once you get beyond even ISO 200, due to noise. Some of the noise can be handled in post, and Capture One IMO does a far better job, but lacks all the profile info so it takes more work. LR really has no color profile either from what I can tell. You just pick one from the options, none of them are camera related unless I missed something in the last update. The way DJI has implemented the chip, it seems to handle a push to the right, as high lights are much more capable of being brought into play (unless shooting directly into the sun) than pushing up the shadows. But bracketing allows for a good range most times. The same chip or similar in the RX100 IV can easily get to ISO 800 and give you a image to work with, but the same ISO 800 on a MP2 is pretty much hard to work with for me.

Also agree that for the price and tech, it's still a great solution, not to mention can be carried for hours in the field with a DSLR, where as with a P4 Pro, I tend to stay closer to my truck and not carry it as far due to weight and noise.

My bigger concerns are how much distortion is going on with both the P4 Pro and MP2 cameras. With the camera level, horizon's are OK, but aim up or down and the distortion is excessive. Adobe's correction IMO takes too much detail from the parts of the image being corrected. The same issue as many found on D-log video. There is also a lot of retro focus distortion towards the edges, common with all wides, but again harsh on the DJI cameras. Objects get flattened an elongated. Easy to see as you pan across a scene. One reason I prefer the 4:3 ratio on the P4 Pro even though it's costing resolution, the 4:3 keeps the worst of the edge distortion under check. This is not as big a deal for a single image series, but a huge deal when taking a 6 or 9 shot series to turn into a larger resolution final file. (9 shots, 3 horizontal shots per segment, 3 segments total).

Paul C

 

[brettbrandon]

In this case, you have to compare RAW images, otherwise all you're comparing is the RAW converter in the two devices which at least on the smaller sensor, definitely includes noise reduction.

Not in this scenario. Noise is noise. My DSLR has absolutely no noise in a clear sky, but I'm not expecting none, just alot less...
The first file is the jpg right from the Sony Cybershot with its built in processing.
The second is the jpg from the M2P direct from the camera with its built in processing.
One would think that the Hassleblad would have the same or better internal processing on the stock jpgs.
If everyone was expected to use raw, that's all that would be available.
My point is that I am very dissapointed at the picture quality for something being sold with the Hassleblad name on it.

DJI reminds me of another company, Behringer audio. He got rich ripping off other companys gear making cheap copies of them and then bought Midas, and Klark Teknik, both top dollar audio companies. Now he claims to use their engineering in his products boasting their names. Its still cheap stuff...

 

[CanadaDrone]

Not in this scenario. Noise is noise. My DSLR has absolutely no noise in a clear sky, but I'm not expecting none, just alot less...
The first file is the jpg right from the Sony Cybershot with its built in processing.
The second is the jpg from the M2P direct from the camera with its built in processing.
One would think that the Hassleblad would have the same or better internal processing on the stock jpgs.
If everyone was expected to use raw, that's all that would be available.
My point is that I am very dissapointed at the picture quality for something being sold with the Hassleblad name on it.

DJI reminds me of another company, Behringer audio. He got rich ripping off other companys gear making cheap copies of them and then bought Midas, and Klark Teknik, both top dollar audio companies. Now he claims to use their engineering in his products boasting their names. Its still cheap stuff...

I'm sorry you're not happy with the output, but it is literally impossible to do a meaningful and objective image comparison outside the parameters I listed earlier. Further to this, viewing a 20MP image at 100% and comparing it to a 16MP image at 100% is also a pointless comparison - you would need to downsample the 20MP image to 16MP, and in doing so I think you would find that it looks significantly better regarding the areas you are unhappy with, as one of the things that does is reduce visible noise. 16MP on a 1/2.3" sensor like the one in your P&S is the equivalent of approximately a 90MP full frame camera - the 1" 20MP Sony sensor is equivalent to approximately a 54MP full frame camera, just to give you an idea.

I also agree that using the Hasselblad name is ridiculous, but 30 seconds on Google would provide enough information for any prospective buyer to know that it is not exactly following in Hasselblad's footsteps ($30K+ medium format cameras) and is simply a marketing ploy (A consumer drone company purchasing a majority stake). In your second and third lines you acknowledge that all you've done is compare RAW converters on two completely different images, which is what I am trying to explain. That 1" Sony sensor is producing far less read noise than any 1/2.3" sensor, all else equal. Noise it not "noise" in the way you are describing it, as there are many ways to change the noise profile, for example with JPEG processing, NR, poor exposure, over exposure, resizing, etc etc. There are objective ways to measure it, and the very first step is not looking at JPEGs. By the sounds of it you simply prefer the JPEGs out of your P&S, which is fine, but doesn't provide any information regarding the sensors at hand.

Every image starts out as RAW - it's not an end use format. The difference between the crappy internal JPEG processing and what you could get out of that image in a proper RAW editor are miles apart. A manually processed RAW is always better, but not everyone wants to process their images after the fact, so of course there are other options and those options come with compromises.

There is no fine detail at all in the shot you posted from the point & shoot - and the trees in the background look like mush at high magnification due to what looks like the high level of NR being applied even at base ISO. If you applied a similarly aggressive NR profile to the M2P shot, you would probably get a result closer to your liking. It is also very easy to apply selective NR to the sky in a few clicks if you are doing any post processing. Another thing to keep in mind is you would never see that fine grain in a print or in any other normal viewing scenario - examining the image at 100% or beyond is the equivalent of putting your nose up to a print several feet wide. In any normal usage scenario that little bit of grain would be invisible. A little bit of grain + much better detail is always preferable to smearing details with NR, whether it be for print or online viewing.


Here is an objective comparison between that Sony sensor and a typical tiny point & shoot sensor if you are curious:

Photographic Dynamic Range versus ISO Setting

 

[brettbrandon]

I'm going to add this and be done with it.
What I am talking about is finished product. I just grabbed the first two photos I saw with sky in it to show a general difference. My apologies for not researching more alike photos.
I have been shooting and working with photos for over 12 years. When working with photos, even someone elses photos, I don't need to know the equiptment to see the quality of photos I am working with. You see what you have to work with and what you are able to do with it. And with that, the photos are rather noisy, even in brighter areas.

Now to answer the original posters question, "Have I got a particularly bad sensor or is it just what the M2Pro does? "
It is a normal sensor, thats just what the M2Pro does.

 

[CanadaDrone]

I'm going to add this and be done with it.
What I am talking about is finished product. I just grabbed the first two photos I saw with sky in it to show a general difference. My apologies for not researching more alike photos.
I have been shooting and working with photos for over 12 years. When working with photos, even someone elses photos, I don't need to know the equiptment to see the quality of photos I am working with. You see what you have to work with and what you are able to do with it. And with that, the photos are rather noisy, even in brighter areas.

Now to answer the original posters question, "Have I got a particularly bad sensor or is it just what the M2Pro does? "
It is a normal sensor, thats just what the M2Pro does.

It doesn't sound like you understand that you are just comparing RAW converters, and you are able to change the look of RAW files (which is the point of RAW). That is all I am trying to get at here, because it is a critical distinction. All you have done is show us that you prefer the look of the sky in one image over another based on completely unknown JPEG processing parameters. You can't come to any conclusions surrounding sensor performance with that information whatsoever.

It's very important for people to understand that, especially if investigating a possible sensor issue, because comparing a pair of JPEGs that were not even equalized for their respective resolutions let alone taken in the same conditions/settings and coming to any conclusions based on that information is completely useless. So to make a comment like "that's just what the M2Pro does" is very misleading, because what you're referring to is merely the JPEG engine with whatever particular settings you used (again these can be changed which will change the image). A person could change the JPEG settings to whatever they want and make the image look horrible, and that would have nothing at all to do with the M2P's capability even though it might seem like it does to someone who does not understand how the image processing pipeline works. That is the point, and that is why you cannot compare JPEGs unless the sole objective is to simply compare JPEG engines. The quality you can extract out of a RAW image is completely different. I am sorry if anything I said earlier was unclear.

 

[brettbrandon]

I understand completely. I'm not talking about a senser shootout, I am talking about what you end up with. I shoot in raw, process it in lightroom, and have problems with noise even in well lit areas of the photo. You can reduce the noise in post, but with the loss of clarity.
The two jpgs show the internal processing of a pocket camera versus what is supposed to be far superior Hassleblad processing. You would think with their superior processing you would have less noise with even a worse sensor.
I'm afraid you are missing the point. As I said, what he is seeing, is what you get. I can't remove the noise in post without messing up the rest of the photo or without doing a whole lot of isolated part work.
Heres a couple of pics with lots of processing. These would not be possible with the noise of the M2P. I don't need a sensor comparison, the original question was is this noise typical, and that's just what the M2Pro does...

 

[Bussty]

I hear what you are saying about the MP2 files. If I take an MP2 file and try and develop the best possible TIFF image out of the RAW file using a RAW convertor I can only get to a certain point. The RAW files are "thin" if you like I stole that from this guy who seems to know what he is doing

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

From purely a noise and detail perspective though if you Develop a RAW MP2 file to a default TIFF then run it through AI Gigapixel it evaporates noise really well BUT also increases detail. Something I haven't seen in other upsizing programs. I'm still experimenting to see if I'm better to edit fully using a RAW convertor then run through AI Gigapixel or if I'm better to develop the RAW image maybe adjusting for dynamic range only then running that TIFF through AI Gigapixel then editing. Important thing for me is in terms of final image quality it's restoring some faith using that approach and gives you an image you can put on the wall in larger sizes.

 

[brettbrandon]

I would keep in mind that you can't do as much with a tiff file as compared to raw. As long as major processing is done before converting to tiff you should still have what you need after noise reduction.
I will have to have a look at this "AI Gigapixel". I have noticed a thread about it bet never read it. Even DSLR's get some noise in dark areas....

 

[CanadaDrone]

I understand completely. I'm not talking about a senser shootout, I am talking about what you end up with. I shoot in raw, process it in lightroom, and have problems with noise even in well lit areas of the photo. You can reduce the noise in post, but with the loss of clarity.

If you do it properly and apply it selectively you can get rid of it without touching the subject's detail. At any rate, a tiny bit of grain will never show up in a print nor a web-sized image, which is 99% of use cases for digital images these days. Viewing 100% crops to scrutinize noise is the equivalent of putting your nose up to a 3-4 foot print and would never be noticeable under any normal circumstances. We're all guilty of pixel peeping like that but it's important to consider the end use of any photo at the same time.

The two jpgs show the internal processing of a pocket camera versus what is supposed to be far superior Hassleblad processing. You would think with their superior processing you would have less noise with even a worse sensor.

They don't show us anything because the resolutions were not equalized - the way it was presented you are exaggerating the M2P noise due to viewing a higher pixel count at exactly the same magnification as the lower pixel count. The photos were also not taken in the came conditions at the same settings, nor do you know what processing is being done by each - therefore no objective comparison can be made. I could process a file horribly, save it as a JPEG, and completely misrepresent the output of any camera and say that what you see is what you get - it just doesn't work like that. I'm not saying that's what you're doing, but it's why JPEG comparisons are useless when evaluating image quality. It would be like judging a fitness competition, but with all the contestants wearing snow suits - you could rate them on what color you liked the best maybe, but not much else.

The same principal applies to much higher resolution cameras, which is why my 46MP D850 files, for example, downsized to 24MP, actually look as good or better than any native 24MP camera. You get the best of both worlds.

As I said, what he is seeing, is what you get. I can't remove the noise in post without messing up the rest of the photo or without doing a whole lot of isolated part work.
Heres a couple of pics with lots of processing. These would not be possible with the noise of the M2P. I don't need a sensor comparison, the original question was is this noise typical, and that's just what the M2Pro does...

You can remove noise selectively in a few clicks if you know what you're doing, and plugins like NIK DFine can do it for you with even less effort (i.e. just the sky or just one color or whatever you want).

You need a RAW image comparison (done in a controlled and objective manner) otherwise we aren't really learning anything at all about the M2P's output potential.

We also have no idea if the photos you posted would be possible with the M2P or not because you did not take them with a M2P and process them identically for comparison. Also at extremely low web resolutions, most photos from cell phone or a $10K DSLR setup are indistinguishable, which admittedly makes it very difficult to present things to support our points on forums. I don't see anything there that I couldn't get out of my RX100 VA, which uses a similar sensor to the M2P.

All I am saying is that if you are going to make comparisons or conclusions abut any camera's output (M2P or otherwise, JPEG or RAW) you need to do so with a proper objective comparison, otherwise nothing is gained. Just looking at random images under completely different scenarios/settings and offering an opinion is 100% fine, but does not provide any facts with regards to evaluating M2P output. I hope that makes sense and I do not mean any offense.

 

[LapetusOne]

I just love how people want to hate on the M2P.
I'm a commercial director and photographer and I literally make my living from videos and photos for major clients all over the world and I couldn't be happier with this little drone. It's a $1,300 toy and it does better work than my Inspire 1 Pro X5 did just 3 years ago. I just can't imagine what you could be doing work wise that this thing is a disappointment to you. If it sucks that bad, then go buy an Inspire 2 X7.

This drone would be military spec hardware just 5 years ago and all I see here are people trying to dissect all of it's problems like it's somehow preventing them from doing good work. The creativity and ambition of your work is more important than the noise level, pixel binning, etc of a toy drone. You are the creative source, not the drone. Nothing is holding you back.
I just think everything is awesome and I'm disappointed that so many people aren't happy. Let's just go outside and create something, we'll be happier. The truth is 99.9999% of your clients and fans will never know the difference.

 

[brettbrandon]

At any rate, a tiny bit of grain will never show up in a print nor a web-sized image, which is 99% of use cases for digital images these days. Viewing 100% crops to scrutinize noise is the equivalent of putting your nose up to a 3-4 foot print and would never be noticeable under any normal circumstances.

Being a photographer who has had wall size prints made, you don't know what you are talking about. Noise does show up in a print. If you are like what you call most people, you are viewing the photos on a phone and even ones taken with the phone look good there.

I could process a file horribly, save it as a JPEG, and completely misrepresent the output of any camera and say that what you see is what you get - it just doesn't work like that. I'm not saying that's what you're doing, but it's why JPEG comparisons are useless when evaluating image quality.

That is why I posted jpg's processed by the cameras themselves untouched.

You can remove noise selectively in a few clicks if you know what you're doing, and plugins like NIK DFine can do it for you with even less effort (i.e. just the sky or just one color or whatever you want).

FYI I do know all about noise reduction, I don't need any help from you. Once again you are so far off the question of the thread.

You need a RAW image comparison (done in a controlled and objective manner) otherwise we aren't really learning anything at all about the M2P's output potential.

I have worked with many MP2 raw files and am frustrated by all the noise. They are advertising this as a Hassleblad. I answered the poster that what they are seeing is typical of this unit. I don't need a stupid sensor shootout to see what is on my computer monitor.

All I am saying is that if you are going to make comparisons or conclusions abut any camera's output (M2P or otherwise, JPEG or RAW) you need to do so with a proper objective comparison, otherwise nothing is gained. Just looking at random images under completely different scenarios/settings and offering an opinion is 100% fine, but does not provide any facts with regards to evaluating M2P output.

I have spent enough years processing photos to tell you I don't care what the sensor is, I can see what I can and can't do with it. The MP2 is noisy period.
Back to the posters question, "Have I got a particularly bad sensor or is it just what the M2Pro does?"
And my response is yes "that is just what the M2Pro does".
I will not be responding to you again as you seem to think you need a scientific experiment to be able to see noise in a photo. Try using your eyes or maybye quit viewing it on a phone iPad/tablet...
I apologize to the original poster for this post going way south.

PS Try looking at the video above, he did it just the way you wanted it and found the same thing I did, go figure......