Thanks Chunky. Wow, that's quite a difference there. Definitely shows a lot more clearly the behaviours I saw while playing around with some comparative footage that's out there. Hard to be sure where the problem is coming from, but still hoping that some processing updates that might get us a little closer. The form factor to the Mavic is hard to resist, but not if image quality suffers this much in comparison. At least for what I do with these.
Mavic Pro-2 vs Zoom and P4P - Line Skipping + Sensor Heat?
- Thread starter Cliff_622
- Start date
Similar to what Cliff_622 showed and the same reason - pixelbinning or lineskipping in the readout.
If the next one or two updates don‘t solve the limitation I decide to return since DJI promises 4K „resolution“ and actually we are somewhat far from that.
Awesome....can you post higher resolution clip of you test results?You are correct. This is not a scientific controlled, notray confirmed test or setup to calculate the "real" resolution of each setting. It is just a privately feasible way of confirming the "gutt feeling" there is something wrong with the FOV mode. It is a little reproducible test to compare two or more settings and you can easily see which is better and which is worse. Of course you cannot judge the real resolution... It is more like a beam balance. You cannot measure the exact weight but you can clearly see what weights more.
And I do not expect it to perform like a 150K$ cam, but at least it should be better than the GoPro Hero 5 or the Mavic 1.
Edit: Added my personal pixel-peeping-beam balance test results:
View attachment 46693
Thanks!
This is fantastic.
DJI is CRUSHING the blue channel with MASSIVE amounts of crappy noise reduction. As you might know that the blue channel is commonly the noisiest of the three so maybe Ambarella is attacking that one the hardest.
This very much co-insides with what my tests show too. Phantom 4 Pro wins BIG time and beats that "Hasselblad" camera all day long
Mavic-2 Pro FOV wide resolution problems were either an "accident" that forced them to line skip....or...it was deliberately done in an effort to protect the current NEW (remember, they just released it) Phantom 4 Pro v2.0 and future Phantom 5 Pro.
CT
Roamer105
Well-Known Member
V1 LightBridge
V2 Occusync
Also (problematically for some of us needing to provide clean HDMI downlink for broadcast), the GO4 app switch to turn of OSD has been removed in V2.
Last edited:
0Kajuna0
Well-Known Member
Thanks. Interesring to see how the Mavic 1 still holds up against FOV.
Has anyone compared FOV to Raw? That should give us some more clues.
@chunky and @Cliff_622
Can you please check/confirm 2.7K and 1080p resolutions are having the same issue or they are good against P4Pro V2 and only 4K resolution is crippled? Would be great to know. Thx
Can you please check/confirm 2.7K and 1080p resolutions are having the same issue or they are good against P4Pro V2 and only 4K resolution is crippled? Would be great to know. Thx
Completely untrue. The HQ mode provides full 4k resolution.
Exactly the same as Canon just managed with their R model.
Yes, it is true 4K (whatever the lens can resolve perhaps not really 2160 lines) but the color/noise is worse than proper 5.5K to 4K downsample due to the debayering effect.
Sorry but you are both wrong - NO native sampling bayersensor results in the native resolution (!!) so even the HQ mode doesn‘t provide true 4K!
Simply compare my colored ISO testchart and see the P4Pro resolution is even higher than the M2Pro HQ mode!
The difference is between 4K imagedimensions and 4K detail information!
In real 4K every of the 3840x2160 pixels can display an individual information not influenced by surrounding pixels!
That‘s not true for FullFOV as not for HQ mode.
P4pro shows better differentiation but also doesn‘t achieve full (4K) separation, at least not in the colored charts (blue/red) where the lower red/blue pixelcount limits resolution as also colorsubsampling does.
DJI states in the manual on p.38 and in the FullFOV FAQ that the „5.5K“ sensor is downsampled to 4K „resolution“(!) - not imagesize and that‘s far from true.
Save all these information (manual and FAQ) for later reference since this may be changed as it happened with Panasonic once...
I think everybody would be fine with P4pro resolution in FullFOV mode and this should be possible!
DJI can do better as they actually show us.
Question is still why DJI is doing this limitation:
1. they want to differentiate from the P4pro - not valid since the whole FAQ and every promotion on their website states „4K resolution“ and how „professional“ the „10bit“ IQ from the „Hasselblad“ camera is.
2. the sensor overheats - also hardly believeable since the P4pro‘s camera isn’t bigger and wouldn‘t dissipate heat better.
The P4pro‘s camera may have better cooling due to better air stream from the props, maybe, than limitate full sensor readout by temperature as Sony does with the RX100V and monitor this in the app.
Before temperature gets to high simply make a fall back to binning. Sony‘s RX100V does up to 5mins. in full sensor readout without (!!) active cooling - not to bad for many things and not to „professional“ - I „could“ live with that.
3. the sensor maybe doesn‘t allow 10bit full sensor readout at 30p - true from the sensorspecs we „know“ (only 26p possible up to 16:9), 24/25p would be also fine „for me“.
Also bring an option for 8bit full sensor readout as an option with D-Cinelike.
DJI, you‘re reading, most Mavic 2 Pro users now know there is „sand in the gearbox“.
Be loyal, your current specs are not true, tell us what you think to do to come out of this, please!
1. they want to differentiate from the P4pro - not valid since the whole FAQ and every promotion on their website states „4K resolution“ and how „professional“ the „10bit“ IQ from the „Hasselblad“ camera is.
2. the sensor overheats - also hardly believeable since the P4pro‘s camera isn’t bigger and wouldn‘t dissipate heat better.
The P4pro‘s camera may have better cooling due to better air stream from the props, maybe, than limitate full sensor readout by temperature as Sony does with the RX100V and monitor this in the app.
Before temperature gets to high simply make a fall back to binning. Sony‘s RX100V does up to 5mins. in full sensor readout without (!!) active cooling - not to bad for many things and not to „professional“ - I „could“ live with that.
3. the sensor maybe doesn‘t allow 10bit full sensor readout at 30p - true from the sensorspecs we „know“ (only 26p possible up to 16:9), 24/25p would be also fine „for me“.
Also bring an option for 8bit full sensor readout as an option with D-Cinelike.
DJI, you‘re reading, most Mavic 2 Pro users now know there is „sand in the gearbox“.
Be loyal, your current specs are not true, tell us what you think to do to come out of this, please!
Last edited:
So just to be clear. In HQ mode the sensor is properly used (no binning, line skipping) and it also uses the true amount of pixels needed to fill up a 4k image, but due to the debayering effect it loses resolution. The P4P however uses the whole 5.5K sensor and properly down samples to achieve a 4k image with slightly less than true 4k resolution.
If heat isn't a problem here but processing power, DJI could therefor add an 8bit mode with proper down sampling to achieve a similar resolution as the P4P. So lets hope DJI is actually willing to implement additional recording modes than.
From blade-strike @ RCG:
"Engineering response to Binning / line skipping
Mavic 2 Pro uses a different processor than P4P and has two 4K modes, as stated on the product page, which are FOV mode and HQ mode. FOV mode has a wider view but less details than HQ mode. FOV mode's image processing requires raw subsampling, which scales 5.5K raw into 4K raw before feeding it to the image processing pipeline.
Raw subsampling is different from line skipping; it uses the full area (of the 16:9 portion) of the image sensor so the signal-noise-ratio is much better than line skipping. Raw subsampling is also different from pixel binning. It is an optimized filtering & sampling process which is much more advanced than the simple average mechanism of pixel binning. Raw subsampling causes some loss in the finest details. However, this is not as visible in real scenes (like aerial filming) as it is on charts.
DJI looked into many solutions for implementing FOV and found that the current FOV mode is the best compromise for the Mavic 2
hardware. Please remember the Mavic 2 is an extremely complex and compact piece of hardware. The current solution represents the best possible way to enable filming using the full 1” sensor wide angle area FOV on the Mavic 2 hardware."
"Engineering response to Binning / line skipping
Mavic 2 Pro uses a different processor than P4P and has two 4K modes, as stated on the product page, which are FOV mode and HQ mode. FOV mode has a wider view but less details than HQ mode. FOV mode's image processing requires raw subsampling, which scales 5.5K raw into 4K raw before feeding it to the image processing pipeline.
Raw subsampling is different from line skipping; it uses the full area (of the 16:9 portion) of the image sensor so the signal-noise-ratio is much better than line skipping. Raw subsampling is also different from pixel binning. It is an optimized filtering & sampling process which is much more advanced than the simple average mechanism of pixel binning. Raw subsampling causes some loss in the finest details. However, this is not as visible in real scenes (like aerial filming) as it is on charts.
DJI looked into many solutions for implementing FOV and found that the current FOV mode is the best compromise for the Mavic 2
hardware. Please remember the Mavic 2 is an extremely complex and compact piece of hardware. The current solution represents the best possible way to enable filming using the full 1” sensor wide angle area FOV on the Mavic 2 hardware."
Similar threads
DJI Drone Deals
New Threads
-
-
Air 3 My Year in Europe with the DJI Air 3- Started by VegardBerland
- Replies: 0
-
DJI Mavic air 2 needing second activation after repair(issue)
- Started by ASD7
- Replies: 2
-
-
Forum statistics
