JustName Posted at 10-29 02:35
I have now a second OA3 here and defocused again. In direct comparison to an OA2, the OA3 delivers clearly washed-out images.

Does this also happen in Ultrawide without stabilization? Can you post an example image/screen grab?

I think you already posted a comparison between the two cameras on this thread, but I assume it was with your previous OA3. Can you see a difference between the two OA3 cameras?

Iancraig10 Posted at 10-29 00:32
“This is because DJI has committed a flaw by declaring an unrealistic specification, "Focus range: 0.3 m to ∞", blatant lie.“

My A3 was bought for me by a shop considering selling them. I was making a video about the camera for them to sell. Once he saw the footage, even on a tripod, he decided not to sell them.

In fact, it all revolves around that randomly shot specification on their site.
If you don't put it then no one can dispute it, or couldn't claim an unstated specification anyway.
"We did not declare hyperfocal," question closed.
But if you write it down and it doesn't the issue is totally different.
On this alone revolves my disappointment, for everything else I like the OA3.

So at 28cm and zoomed into 140% . You will notice in focus but close focus also in the front edges you can read visa on the CC in the wallet also keys , for this camera it looks fine and acceptable I don’t ever see the need to be focusing this close with this camera but it is important that focus is reasonable from front to back, also shot in relatively low light. Resolution less than 720p for this platform. Higher resolution on YT if you choos to watch on that platform.

Action 3 all auto settings shot in 4k 50fps 4.3. FOV Wide. I shot it on my Lap just to show I wasn’t further than advertised parameters.  I reduced to 1080p for YT and it auto reduces to less than 720p for this platform

hallmark007 Posted at 10-29 04:08
So at 28cm and zoomed into 140% . You will notice in focus but close focus also in the front edges you can read visa on the CC in the wallet also keys , for this camera it looks fine and acceptable I don’t ever see the need to be focusing this close with this camera but it is important that focus is reasonable from front to back, also shot in relatively low light. Resolution less than 720p for this platform. Higher resolution on YT if you choos to watch on that platform.

Nice, please share the source and the settings.

fish sticks Posted at 10-29 02:49
Does this also happen in Ultrawide without stabilization? Can you post an example image/screen grab?

I think you already posted a comparison between the two cameras on this thread, but I assume it was with your previous OA3. Can you see a difference between the two OA3 cameras?

Yes i did choose the settings recommended by DJI on both cameras. Ultrawide, 100-iso and good light conditions.
For me it looks like my first and second OA3 has the same wrong optics calibration. Like a wrong setting in the building machine.

I will post some new comparison images later.

MarcoR Posted at 10-29 04:12
Nice, please share the source and the settings.

Action 3 all auto settings shot in 4k 50fps 4.3. FOV Wide. I shot it on my Lap just to show I wasn’t further than advertised parameters.  I reduced to 1080p for YT and it auto reduces to less than 720p for this platform

hallmark007 Posted at 10-29 04:24
Action 3 all auto settings shot in 4k 50fps 4.3. FOV Wide. I shot it on my Lap just to show I wasn’t further than advertised parameters.  I reduced to 1080p for YT and it auto reduces to less than 720p for this platform

It's incredibly great result, post the source, thank you!
And if you can, do exactly the same video with the OA2, at the same settings, distance, etc. etc....

MarcoR Posted at 10-29 04:29
It's incredibly great result, post the source, thank you!

When you say source you’re referring to what?

hallmark007 Posted at 10-29 04:31
When you say source you’re referring to what?

The file generated by the cam.
You still had RockSteady enabled, correct? It's too still the image and too sharp, so also the colors in Natural.

MarcoR Posted at 10-29 04:32
The file generated by the cam.

Why would this make any difference.

hallmark007 Posted at 10-29 04:39
Why would this make any difference.

Of course there is!
It's kind of like when you rightly objected to me putting crops and images here, I always posted the sources, that's what we make assessments on.
Even a single frame extracted from your video is enough.

And please also try to take a picture, again from that distance, thank you.
In the movie digital sharpness comes into play, in the photo it does not.
You might have a doped OA3, lucky you!

MarcoR Posted at 10-29 04:44
And please also try to take a picture, again from that distance, thank you.
In the movie digital sharpness comes into play, in the photo it does not.
You might have a doped OA3, lucky you!

Again I don’t need to take a photo. If the camera focuses from less than .30  with video then surely that covers what the blurb says.

MarcoR Posted at 10-29 04:44
And please also try to take a picture, again from that distance, thank you.
In the movie digital sharpness comes into play, in the photo it does not.
You might have a doped OA3, lucky you!

I have searched high and low but I don’t see anywhere I asked you for source. Its sort of becoming ridiculous if something is in focus its in focus, if its out of focus you cannot get it into focus . But when Im home ill post link to orignal video. It also has become apparent that those posting in focus videos are seen almost as the enemy, and that smacks of real buyers remorse.

---Deleted---

Iancraig10 Posted at 10-29 05:40
That is a lot better than what I’m getting on mine. Maybe I have a faulty one. It was one of the first released.

Absolutely not, why enemy? Lucky!
If you get paranoid and troll then yes you go back to the predecedent level.
Never mind the original video, never mind, thanks anyway.
If you really want to make yourself useful take out the OA2 that you have it, put them side by side and make a comparative video with the same settings, then we can understand something.
That would be the 9 proof.
And if you don't want is the same, no enemies here.

MarcoR Posted at 10-29 05:45
Absolutely not, why enemy? Lucky!
If you get paranoid and troll then yes you go back to the predecedent level.
Never mind the original video, never mind, thanks anyway.

https://drive.google.com/file/d/ ... view?usp=share_link

Orignal video cropped video above was cropped in luma fusion

MarcoR Posted at 10-29 05:45
Absolutely not, why enemy? Lucky!
If you get paranoid and troll then yes you go back to the predecedent level.
Never mind the original video, never mind, thanks anyway.

The reason enemy is simple. If I post a blurred out of focus video everyone is so glad they’re all over it like a rash. Post something reasonable or good and no interest or you must be cooking it up.
I can say originally I did believe I had some kind of problem, but with new FW it improved but dji then posted internal parameters , so I contacted them went through a full factory reset followed by reinstall up to date Fw followed by horizon calibration. I did question the need for the latter, they said this could affect focus I said it couldn’t but did it anyway and for video use I’m not having a problem. My need is the camera outdoors is reasonably in focus from front to back and to the outer edges and I’m satisfied it is.

Regarding photos. Its a horrible stills camera and is really only possible to get useable photos in very good to good light.
I will if the light is good tomorrow post some good light photos as well as some half lit good and bad photos, focus won’t be the issue light will , bad light causes extremely soft focus including mushy spots all over the photo. Focusing too close leads to distortion which can also look like areas are out of focus or not focusing from the centre out but off centre you will see parts of the image in sharp focus including when cropped to 100%, this looks to be caused by the distortion of a wide angle lens and this should be expected if you are adamant on shooting within minimum distance. Can dji improve this and overall photo quality I expect they can, but almost no one should buy this for shooting stills their forever best holiday.

hallmark007 Posted at 10-29 06:17
The reason enemy is simple. If I post a blurred out of focus video everyone is so glad they’re all over it like a rash. Post something reasonable or good and no interest or you must be cooking it up.
I can say originally I did believe I had some kind of problem, but with new FW it improved but dji then posted internal parameters , so I contacted them went through a full factory reset followed by reinstall up to date Fw followed by horizon calibration. I did question the need for the latter, they said this could affect focus I said it couldn’t but did it anyway and for video use I’m not having a problem. My need is the camera outdoors is reasonably in focus from front to back and to the outer edges and I’m satisfied it is.

I performed the same operation on 2 OA3s, same bad result, identical.
The only difference is that I recorded at 30 fps.
We will see what DJI will say when they get the replacement.
Thanks for the video.

Iancraig10 Posted at 10-29 05:40
That is a lot better than what I’m getting on mine. Maybe I have a faulty one. It was one of the first released.

If this camera is used in the right conditions for a camera with such a small sensor then good footage is quite possible. If people continue to get hung up with slightly out of focus issues from incredibly close distances then they will never reveal the possibilities of the camera.
We all accept that iphone footage goes through a tremendous amount of artificial editing process including amazing amounts of sharpening and the biggest array of pleasing and saturated colouring to churn out its videos. Yet what I see is everyone comparing footage straight out of this camera with fully processed iphone footage.

No one seems to feel the need to spend any time editing or trying to improve the file they get their footage from.

Photos are not good unless taken in good light and this is similar with all these cameras and trying to create outstanding photos from .5 metre is just not a thing unless you shoot macro photography, I believe you can get focus at this distance but distortion will determine where the focus is. But if take some normal everyday shooting of video you will appreciate this camera is in focus from very close in front to the back of your frame and pretty good to the edges, it doesn’t have great DR so plenty of dark spots that we see in every 8 bit camera.

Two pics.. one with the Action 2 and the other with Action 3. Same conditions, same settings etc.

JustName Posted at 10-29 06:58
Two pics.. one with the Action 2 and the other with Action 3. Same conditions, same settings etc.

[view_image]

---Deleted---

Iancraig10 Posted at 10-29 07:45
I get the same kind of result when I compare my A2 with A3. That’s why I’m starting to wonder whether I (and you) might have a faulty camera.

Especially after seeing Halmark007’s video, although it still doesn't look great on a large screen tbh. However, I think to be expected in those lighting conditions.

Reply to JustName: it is very difficult that halmark007 is here one of the very few lucky who has a well-calibrated one, but it is not impossible.
The pictures you posted show the difference there is, but it would be better if you took two videos with the same settings.

MarcoR Posted at 10-29 05:45
Absolutely not, why enemy? Lucky!
If you get paranoid and troll then yes you go back to the predecedent level.
Never mind the original video, never mind, thanks anyway.

You’ve lost me Marco. Don’t understand.

Iancraig10 Posted at 10-29 07:54
You’ve lost me Marco. Don’t understand.

I think you don't have a failed cam because on average that is the comparative result you get with OA2, other users have shown it, including on video.

MarcoR Posted at 10-29 07:53
Reply to JustName: it is very difficult that halmark007 is here one of the very few lucky who has a well-calibrated one, but it is not impossible.
The pictures you posted show the difference there is, but it would be better if you took two videos with the same settings.

I think there are certain reasons why the halmark doesn't post pictures, but only creepy videos.

Of course, on a video, the difference is not so noticeable, because here are significantly more factors and the image is rather with the OA2 worse, than with the OA3 better when a video is recorded.

I use the ActionCams to make during diving not only videos, but also photos.  Currently the OA3 ist not the best choice for this.

JustName Posted at 10-29 08:14
I think there are certain reasons why the halmark doesn't post pictures, but only creepy videos.

Of course, on a video, the difference is not so noticeable, because here are significantly more factors and the image is rather with the OA2 worse, than with the OA3 better when a video is recorded.

---Deleted---

JustName Posted at 10-29 08:14
I think there are certain reasons why the halmark doesn't post pictures, but only creepy videos.

Of course, on a video, the difference is not so noticeable, because here are significantly more factors and the image is rather with the OA2 worse, than with the OA3 better when a video is recorded.

What creepy photos/videos have I posted? I have posted 100s of my own photos here from my drones I dont remember posting a single one from any action camera I own or an osmo pocket and I own 7 in total. These are not cameras for stills photography . You mention diving well if you want a better small camera for photos the pocket 2 is the best of them.

JustName Posted at 10-29 08:14
I think there are certain reasons why the halmark doesn't post pictures, but only creepy videos.

Of course, on a video, the difference is not so noticeable, because here are significantly more factors and the image is rather with the OA2 worse, than with the OA3 better when a video is recorded.

Previously I posted two frames taken from movies, not photos, comparing OA3 and GoPro HERO 6, making sure that the central image with the paper coming to the sensors occupies the same pixels, and the difference can be seen quite a bit.
But they are two different cams, this would be a test to do with OA2, which I don't own or I would have already done.

hallmark007 Posted at 10-29 08:47
What creepy photos have I posted? I have posted 100s of my own photos here from my drones I dont remember posting a single one from any action camera I own or an osmo pocket and I own 7 in total. These are not cameras for stills photography . You mention diving well if you want a better small camera for photos the pocket 2 is the best of them.

As first please learn to read..  

MarcoR Posted at 10-29 08:54
Previously I posted two frames taken from movies, not photos, comparing OA3 and GoPro HERO 6, making sure that the central image with the paper coming to the sensors occupies the same pixels, and the difference can be seen quite a bit.
But they are two different cams, this would be a test to do with OA2, which I don't own or I would have already done.

I have been testing and will post some samples tomorrow, but these two cameras are very similar shooting photos, both are dismal in lowlight, lets see if we get good light. But even with good light trying to focus at very close distance with A3 will result in focus off centre because of the very wide FOV taking photos at reasonable distance in good light will give you ok photos. Lets wait and see.

JustName Posted at 10-29 09:02
As first please learn to read..

Apoligies, so what creepy videos did I post ? It is almost Halloween maybe I struck it lucky…

hallmark007 Posted at 10-29 09:03
I have been testing and will post some samples tomorrow, but these two cameras are very similar shooting photos, both are dismal in lowlight, lets see if we get good light. But even with good light trying to focus at very close distance with A3 will result in focus off centre because of the very wide FOV taking photos at reasonable distance in good light will give you ok photos. Lets wait and see.

Sure, but I mean two movies with the same settings, since even the OA2 seems to me to have the same FOV and Ultra-Wide, or am I wrong?
Very welcome this if you make it man, thank you!

MarcoR Posted at 10-29 09:27
Sure, but I mean two movies with the same settings, since even the OA2 seems to me to have the same FOV and Ultra-Wide, or am I wrong?
Very welcome this if you make it man, thank you!

Its wide but not so wide.

MarcoR Posted at 10-29 09:27
Sure, but I mean two movies with the same settings, since even the OA2 seems to me to have the same FOV and Ultra-Wide, or am I wrong?
Very welcome this if you make it man, thank you!

There is exactly the same FOV on OA2 and OA3.  155°, f/2.8

MarcoR Posted at 10-29 08:54
Previously I posted two frames taken from movies, not photos, comparing OA3 and GoPro HERO 6, making sure that the central image with the paper coming to the sensors occupies the same pixels, and the difference can be seen quite a bit.
But they are two different cams, this would be a test to do with OA2, which I don't own or I would have already done.

I have seen the videos, but the two cameras have totally different field of view, so the sharpness will be different already because of that.

Should it be bright and dry tomorrow, I will try to make two comparison videos.

JustName Posted at 10-29 10:33
I have seen the videos, but the two cameras have totally different field of view, so the sharpness will be different already because of that.

Should it be bright and dry tomorrow, I will try to make two comparison videos.

Much appreciated, thank you!

In another topic a user shared two comparative movies made with A2 and OA3, with the cams set the same way.
I extracted two frames from the movies and made some comparisons.
The difference is not little, I post only a few crops, here the source.
He also posted a video made with Pocket 2, and there the level is far beyond obviously, but it is not an action cam of course.

Also from those two videos a fairly central subject in the shot, roughly about 2 meters away from the cams.
Less obvious, but something different is there.
I would like to know the difference on the lenses of A2 and OA3, when I believe the sensor is the same, because imho it doesn't just a digital increase in sharpness on A2 because the problem is exacerbated on neighboring subjects.
Shot in the dark: did DJI saved a little too much on the OA3's lenses quality to make it as competitive as possible?
Hasselblad could say "almost certainly"!

MarcoR Posted at 10-30 05:45
Also from those two videos a fairly central subject in the shot, roughly about 2 meters away from the cams.
Less obvious, but something different is there.
I would like to know the difference on the lenses of A2 and OA3, when I believe the sensor is the same, because imho it doesn't just a digital increase in sharpness on A2 because the problem is exacerbated on neighboring subjects.

Same thing with these comparisons: they are not enough to show a hardware issue with the lens. The FOV is significantly different on the two images and the videos. We saw in this thread that if you use the widest FOV on the OA3 and A2 you can get a matching result in terms of detail/sharpness.

You could argue that "the default settings" of the OA3 lead to decreased image quality, or that it is very easy to select settings that reduce detail, but this is different than some other camera samples where the image degradation was quite severe.