Mission 1 Pro: Can GoPro Labs completely disable noise reduction? #1684
Replies: 15 comments 16 replies
-
|
NR01=x - Noise Reduction control. e.g. $NR01=50 range 1-100, 100 being 100% active (default), so 1 is the noisest. Mainly if you want noise reduction complete off, or lower than Low. For the GP13 ( custom log curve has not being implemented in Mission 1 yet) : NR01=x - Noise Reduction control, which should be combined with the LOGB control. e.g. oMNR01=50 range 1-100, 100 being 100% active (default), so 1 is the noisest. When pushing the dynamic range of any camera, shadow details may be suppress by noise reduction. Increasing the logbase will enhance the shadows, but will also reveal noise reduction artifacts tuned for a different encoding curve (the more consumer friendly Rec709.) |
Beta Was this translation helpful? Give feedback.
-
|
Thanks for the explanation, but I think we're talking about two different use cases. I'm not using LOGB or any custom log curve. My question is specifically about the standard Mission 1 Pro image pipeline. What I'm trying to understand is:
The reason I'm asking is that on previous cameras (Hero 9 / Hero 11), reducing NR produced a much noisier image. On Mission 1 Pro, even with NR01=1, the footage still appears surprisingly clean. So I'm trying to determine whether this is due to a better sensor/pipeline, or because some noise reduction remains active outside the NR01 control. |
Beta Was this translation helpful? Give feedback.
-
|
This might help #1217 (comment) |
Beta Was this translation helpful? Give feedback.
-
|
Thanks, but that discussion refers to previous HERO cameras. My question is specifically about Mission 1 Pro and whether the NR pipeline behaves the same way on this new platform. Mission 1 Pro appears noticeably cleaner than HERO cameras even with NR01=1, which is why I'm trying to understand if the behavior is identical or if there have been changes to the image processing pipeline. |
Beta Was this translation helpful? Give feedback.
-
|
It turns off all the denoising at the GP3 level, unless there is a bug a something remains on. I just re-ran a test, shooting 6400 ISO with a darkish scene, and NR01=1 is noiser than just DNSC LOW. So it is somewhat working. I agree that is not as noisy as H13, a lot as this sensor is way clearer. NR01 scales the denoise parameter of the UI setting. So NR01=50 for DNSC LOW will half the Low setting. However I also agree there is a lack of pixel randomness. I will investigate further in the next weeks. |
Beta Was this translation helpful? Give feedback.
-
|
so you say NR01 scales the denoise parameter of the UI setting. NR01=1 is denoising off on LOW? and NR01=1 is denoising off on HIGH? so both look the same? |
Beta Was this translation helpful? Give feedback.
-
|
I intended to use NR01 to reduce the smearing effect of GP3 on dark details. Unfortunately, when I turned on NR01, the dark areas in 8K 30p recording did not exhibit random noise, but rather algorithmic smearing. |
Beta Was this translation helpful? Give feedback.
-
|
After careful comparison, I feel that NR01 doesn't seem to work in 4K, as the noise remains in 8K. |
Beta Was this translation helpful? Give feedback.
-
|
4K is pixel binned. Noise will be less prominent because of this. |
Beta Was this translation helpful? Give feedback.
-
|
Beta Was this translation helpful? Give feedback.
-
|
After more testing, I think Night Mode is extremely well tuned for high ISO shooting. In my tests, it provides a result that I cannot easily reproduce using the normal mode with DNSC LOW and additional noise reduction in DaVinci Resolve. Someone clearly spent a lot of time tuning this mode. However, I would strongly advise against using Night Mode in good lighting conditions (low ISO). The loss of fine detail is quite significant compared to the normal mode. At low ISO, the normal mode preserves noticeably more texture and micro-detail. So far, my impression is: Low ISO → Normal mode I have not used any GoPro Labs noise reduction controls for these tests. Everything was tested with the standard camera settings. What I am still trying to determine is where the crossover point is. Based on my current tests, Night Mode definitely starts to make sense around ISO 1600 and above, but I need more testing. I am also curious whether completely disabling noise reduction through GoPro Labs could provide any real advantage over DNSC LOW, especially with the new sensor. The sensor already appears much cleaner than previous HERO cameras. One more question: we know Mission 1 Pro uses dual gain. At what ISO value does the HCG/LCG transition occur? Is there a specific crossover ISO where the sensor switches to high conversion gain? |
Beta Was this translation helpful? Give feedback.
-
|
That smearing looks terrible on the 4K NR01=1 example. Long time ago David made a beta firmware with noise reduction off for the Hero 11. Im still using that FW. Can you guys upload somewhere short sample clips for some pixel peeping? GP-Log2 8K30/4K30 NR Off. |
Beta Was this translation helpful? Give feedback.
-
|
However, with 8K30p (240mbps), especially 8K60p (360mbps), a significant amount of random noise remains when NR01 is turned on. I believe this effect is quite similar to that of the previous GoPro 12/13. |
Beta Was this translation helpful? Give feedback.
-
|
Can you show more samples? Even with normal/Cinematic color? Im also curious about the loss of detail in base ISO range. Hero 11-13 was good between ISO100-800 with NR off, and terrible above ISO100 with NR on. M1P should be good in ISO100-3200 range. This is very important due to how Gyro compensated exposure works. |
Beta Was this translation helpful? Give feedback.
-
you can check this screen shot 8k60 with iso 6400 nr01 enable， currently I only have 8k footage, 4k results may update later.
|
Beta Was this translation helpful? Give feedback.
-
|
Most of the remaining noise reduction is only happening in 4K24-60. 8K are some have found completely disables the reduction with NR01=1, any non-randomness at 8K will be the limitation of HEVC at the bitrate microSD can currently support. Noise is hard to compress, so 300Mb/s 8K60, is like 75Mb/s 4K60 which is only so-so for noisy data. Now for the good news, there is a control today for the noise reduction in 4Kp24 to 60 modes. BTNR=1, Bypass Temporal Noise Reduction. On HERO, the temporal NR was disabled by NR01. This will be fixed in the next release so NR01 will do it all. In the meantime try: $NR01=1$BTNR=1 |
Beta Was this translation helpful? Give feedback.
-
|
Interesting, so Temporal NR and Spatial NR on HERO series were linked thru NR01? Are there more details on BTNR as a separate command and the allowed range of values? i.e. if I want to disable Spatial NR, but keep Temporal NR (to a low value), or vice-versa for scenes with more movement. |
Beta Was this translation helpful? Give feedback.
-
|
No, it doesn't work that way. Spatial denoising blocks, do allow for the mixing of source and output streams, so NR01=1 is 100% source. Whereas the temporal NR is binary, at least in the way it is implemented in Labs. |
Beta Was this translation helpful? Give feedback.
-
NoiseControl.mp4This is a very clean camera, I really had to gain up a lot to see good random noise. |
Beta Was this translation helpful? Give feedback.
-
|
hi david，
When I set the bitrate to b300 or b400 for the current 8k30p, the maximum bitrate is only 240mbps, which I believe affects the preservation of noise. Is there a way to increase the bitrate of 8k30p to 300mbps or higher?
Additionally, I tested that 8K60p can achieve a bitrate of 360mbps. When setting the b400 option, I hope 8K30p can also achieve a higher bitrate than the same setting, to preserve more details and noise.
…---- Replied Message ----
| From | David ***@***.***> |
| Date | 06/10/2026 04:31 |
| To | gopro/labs ***@***.***> |
| Cc | freefreer ***@***.***>,
Comment ***@***.***> |
| Subject | Re: [gopro/labs] Mission 1 Pro: Can GoPro Labs completely disable noise reduction? (Discussion #1684) |
https://github.com/user-attachments/assets/a9f4a365-7396-4571-bfda-14fcbda2aa33
—
Reply to this email directly, view it on GitHub, or unsubscribe.
Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS and Android. Download it today!
You are receiving this because you commented.Message ID: ***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
Higher bitrates are possible, these are examples using 8K24 open gate, with *NR01=1. You can see that asking for higher and higher bit-rates have negative results. What is happening: if the SD card falls behind, the bit-rate is dropped significantly, then ramps again, making it less efficient. This take some testing on your camera and SD card media to optimized. 370Mb/s is the highest I've ever achieved. |
Beta Was this translation helpful? Give feedback.
-
|
I'll also test the bitrate of OpenGate 8K 24p. Currently, on my SD card, I can achieve approximately 360Mbps bitrate for 8K 60p 16:9, but for 8K 30p in 16:9, it only achieves 240Mbps. My settings are B400 NR01. I don't know why the 8K 30p in 16:9 is always around 240Mbps. I hope you can help me test it, and I will try further!
…---- Replied Message ----
| From | David ***@***.***> |
| Date | 06/12/2026 00:22 |
| To | gopro/labs ***@***.***> |
| Cc | freefreer ***@***.***>,
Comment ***@***.***> |
| Subject | Re: [gopro/labs] Mission 1 Pro: Can GoPro Labs completely disable noise reduction? (Discussion #1684) |
image.png (view on web)
Higher bitrates are possible, these are examples using 8K24 open gate, with *NR01=1. You can see that asking for higher and higher bit-rates have negative results. What is happening: if the SD card falls behind, the bit-rate is dropped significantly, then ramps again, making it less efficient. This take some testing on your camera and SD card media to optimized. 370Mb/s is the highest I've ever achieved.
—
Reply to this email directly, view it on GitHub, or unsubscribe.
Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS and Android. Download it today!
You are receiving this because you commented.Message ID: ***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
|
Yes there is something weird with the 16x9 modes being forces to 240Mb/s. A bug. Shoot 4x3 is you can. |
Beta Was this translation helpful? Give feedback.
-
|
Okay, I understand. Thank you for your explanation. I hope this issue can be fixed in the future.
…---- Replied Message ----
| From | David ***@***.***> |
| Date | 06/12/2026 00:50 |
| To | gopro/labs ***@***.***> |
| Cc | freefreer ***@***.***>,
Comment ***@***.***> |
| Subject | Re: [gopro/labs] Mission 1 Pro: Can GoPro Labs completely disable noise reduction? (Discussion #1684) |
Yes there is something weird with the 16x9 modes being forces to 240Mb/s. A bug. Shoot 4x3 is you can.
image.png (view on web)
—
Reply to this email directly, view it on GitHub, or unsubscribe.
Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS and Android. Download it today!
You are receiving this because you commented.Message ID: ***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
|
The next version will allow 8K 16:9 to go to 400Mb/s, I got that much working. |
Beta Was this translation helpful? Give feedback.
-
|
Thanks. I think 400 Mbps should offer better protection for details and random noise. My Lexar 1066x SD card can achieve write speeds of around 500 Mbps, so 400 Mbps video bitrate should be fine. I look forward to the development of this feature.
…---- Replied Message ----
| From | David ***@***.***> |
| Date | 06/12/2026 04:47 |
| To | gopro/labs ***@***.***> |
| Cc | freefreer ***@***.***>,
Comment ***@***.***> |
| Subject | Re: [gopro/labs] Mission 1 Pro: Can GoPro Labs completely disable noise reduction? (Discussion #1684) |
The next version will allow 8K 16:9 to go to 400Mb/s, I got that much working.
—
Reply to this email directly, view it on GitHub, or unsubscribe.
Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS and Android. Download it today!
You are receiving this because you commented.Message ID: ***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
|
hi david，
When will we have 400Mbps 16:9 8K? Do you think video resolution can be improved at 400Mbps? Currently, I think 360Mbps video has a significant improvement in sharpness compared to 240Mbps. I'm really looking forward to high bitrate GoPro videos.
…---- Replied Message ----
| From | David ***@***.***> |
| Date | 06/14/2026 00:04 |
| To | gopro/labs ***@***.***> |
| Cc | freefreer ***@***.***>,
Comment ***@***.***> |
| Subject | Re: [gopro/labs] Mission 1 Pro: Can GoPro Labs completely disable noise reduction? (Discussion #1684) |
No, it doesn't work that way. Spatial denoising blocks, do allow for the mixing of source and output streams, so NR01=1 is 100% source. Whereas the temporal NR is binary, at least in the way it is implemented in Labs.
—
Reply to this email directly, view it on GitHub, or unsubscribe.
Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS and Android. Download it today!
You are receiving this because you commented.Message ID: ***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
|
hi david，
I noticed that in Flat mode with NR01 enabled, I see color noise and luminance noise retained in the image, while other modes only show black and white noise. I would like to ask why this is the design? Is it possible to disable color noise reduction in other modes?
…---- Replied Message ----
| From | David ***@***.***> |
| Date | 06/14/2026 00:04 |
| To | gopro/labs ***@***.***> |
| Cc | freefreer ***@***.***>,
Comment ***@***.***> |
| Subject | Re: [gopro/labs] Mission 1 Pro: Can GoPro Labs completely disable noise reduction? (Discussion #1684) |
No, it doesn't work that way. Spatial denoising blocks, do allow for the mixing of source and output streams, so NR01=1 is 100% source. Whereas the temporal NR is binary, at least in the way it is implemented in Labs.
—
Reply to this email directly, view it on GitHub, or unsubscribe.
Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS and Android. Download it today!
You are receiving this because you commented.Message ID: ***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
|
What res and fps? GPLog2 not behaving the same way? |
Beta Was this translation helpful? Give feedback.
-
|
Beta Was this translation helpful? Give feedback.
-
|
shot in 8k60p,I rechecked the gplog mode and it had a similar effect, with some color noise, but the natural mode didn't have this noise. 
|
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
Hi David,
I have a question regarding noise reduction on the Mission 1 Pro.
Could you explain how the Low, Medium and High noise reduction settings work internally? Are they simply different strengths of the same algorithm, or do they enable different processing pipelines?
I have also been testing the GoPro Labs parameter
NR01. According to the documentation, values from 1 to 100 control noise reduction strength. However, after comparing footage recorded withNR01=1and higher values, I see much less difference than expected.This makes me wonder:
NR01=1truly disable noise reduction, or does it only reduce part of the processing?NR01value?NR01interact with the built-in Low / Medium / High noise reduction settings?From my tests, it appears that some level of noise reduction is always active, even with
NR01=1, but I would like to understand whether this is expected behavior or if I am misunderstanding how the feature works.I have one more question regarding Night mode.
The results are surprisingly good, especially in low light, and I am very impressed by the image quality. Could you provide some insight into how Night mode works internally? For example, does it use temporal noise reduction, multi-frame processing, frame stacking, or any other advanced techniques?
I am also wondering whether it is possible to achieve even better image quality by recording with minimal in-camera processing (using GoPro Labs settings) and then applying noise reduction and enhancement later in DaVinci Resolve.
Or does Night mode already have access to sensor data and processing stages that cannot be replicated afterwards from the recorded H.265 file?
From my tests, Night mode performs exceptionally well, which makes me wonder whether post-processing can realistically improve upon it, or whether the camera is already doing the optimal processing before compression.
Thank you for your time and for all the work that has gone into Mission 1 Pro and GoPro Labs.
Beta Was this translation helpful? Give feedback.
All reactions