I bought the Action 3 and sold the Action 1 just before. Now I have a unfocused/unsharp image.
What can I do?

I would be more than happy to get this fixed?
Is there a way I can help?

All the best
Martin

Found this video and image seems good to me? Please deny the torture parts but also good tests to prove its a solid cam

https://youtu.be/q-dVnIFAoNQ

TNQ Posted at 10-12 11:02
Found this video and image seems good to me? Please deny the torture parts  but also good tests to prove its a solid cam

He cut between so many cameras, but it is really soft you can clearly see around 1:39 or so

osmonauta Posted at 10-3 03:23
Here are 2 photos, Action 3 vs iPhone 7. Both of them were taken on a table-top tripod 50cm from the book. The Action 3 was using Pro mode with ISO set to 100-200 (all other settings are default). I don't have a "studio" so I did cropped the photos and they were both saved as JPG at 100%.

(Granted, I bought the iPhone 7 brand new in 2019 because it was on sale for $360, so basically they both were in the same price range.)

TNQ Posted at 10-12 11:02
Found this video and image seems good to me? Please deny the torture parts  but also good tests to prove its a solid cam

https://youtu.be/q-dVnIFAoNQ

You don't actually see much of of the OA3 footage, but the footage which is shown looks pretty decent to me
I think I've seen some several good and sharp recordings too already. So I guess there're some good and bad batches around. How manny cams would have been sold already now? Thousands or maybe ten thousands? Just making a guess here, but wouldn't be more people complaining by now then?

I'm still waiting for my OA3 to arrive. It'll probably be mondy.
But Somehow, after reading and seeing some stuff here I'm afraid of recieving a bad camera too, and maybe their just all the same.
But when I see comparisons with GoPro's etc. Then I feel like the GoPro's sometimes look the same when talking about "softness". I've even saw some comparisons where to my feeling the OA3 looked sharper.

But it's hard to tell since a lot of people record different things from different distances in different lightning conditions. Maybe it would have been a good thing to create some universal test. I saw some tests around with some test markings set on different distances, and I feel something like this is actually a good universal test. From that it's better to say if camera's perform differently of if they're all out of focus. And everyone would be able to print the same test sheet at home.

Rockel83 Posted at 10-12 18:48
You don't actually see much of of the OA3 footage, but the footage which is shown looks pretty decent to me
I think I've seen some several good and sharp recordings too already. So I guess there're some good and bad batches around. How manny cams would have been sold already now? Thousands or maybe ten thousands? Just making a guess here, but wouldn't be more people complaining by now then?

I like the idea of a common used test chart. As you say it's hard to compare different cameras with different objects in view. I'm going to print this ISO12233-chart in this link to see how it turns out:
Test pattern here

Hi, just one question, first time writing here. Generally, is it just close focus problem (30cm) or is it that whole image is off focus? Tnx

---DELETED---

---DELETED---

What do I have to do so that they diagnose my camera that way, too?

aliveathome Posted at 10-12 21:29
I like the idea of a common used test chart. As you say it's hard to compare different cameras with different objects in view. I'm going to print this ISO12233-chart in this link to see how it turns out:
Test pattern here

Something like a chart like this yes. Let's say when everyone uses the same (a lot of people will be able to print) then it would be much better to compare, since everybody somewhat has the same test results.

Only need to find a good distance to test from. And maybe it's good to use multiple charts on different distances (let's say 30/40/50cm).

---DELETED---

---DELETED---

osmonauta Posted at 10-13 03:49
Chat with DJI...?

German support does not know anything about this, it seems...

osmonauta Posted at 10-13 03:51
The only issue I have with that test pattern is that it doesn't come with a printer...  I can, however, display it on my laptop full screen.

I don't know if testing with a screen is a good universal testmethod. Because of diversity in resolutions, refreshrates, differences in colors etc.

Maybe one of these can be helpfull (they're all in PDF, so printable)?
- https://treasurevalleycameraclub.com/wp-content/uploads/2020/01/focus_test_chart_edited.pdf
- http://www.squit.co.uk/photo/files/FocusChart.pdf
- http://www.bealecorner.org/red/test-patterns/ISO_12233-reschart.pdf

I actually like the 2nd one myself. Not too much ink to print, but the combination of vertical  and diagonal test looks/seems practical. It takes a little bit more effort to "build" but after it looks like a good way to make a good and universal test with because it will be easy to put the camera and the test subject at horizontal line, at an specified distance from eacht other. Not much what could go wrong here. And everybody should be able to get the same test (angle) setup like this.

Printed the test pattern at this link on an A4 paper (PDF-version):  ISO12233 test pattern
The camera is on tripod and the center of lens is placed on the same height as the center of the pattern (10.5cm). FOV: Standard (Dewarp), Exposure: Auto, White Balance: Auto.

Result:

10cm


20cm


30cm


40cm


50cm

Johannes Posted at 10-13 05:30
German support does not know anything about this, it seems...

"Wir haben es nicht gewusst"

Sorry, bad joke

aliveathome Posted at 10-13 06:39
Printed the test pattern at this link on an A4 paper (PDF-version):  ISO12233 test pattern
The camera is on tripod and the center of lens is placed on the same height as the center of the pattern (10.5cm). FOV: Standard (Dewarp), Exposure: Auto, White Balance: Auto.

Oh cool!

Well from here behind my screen the distances 30cm and up seem to be quite clear. But hadn't expected it would "look that far away".
Didn't find a way here (yet) to download the immages and zoom in on it to see if the lines are sharp (but also this zooming in needs te be don in a way without loosing quality) so not sure yet how to achieve that now haha

aliveathome Posted at 10-13 06:39
Printed the test pattern at this link on an A4 paper (PDF-version):  ISO12233 test pattern
The camera is on tripod and the center of lens is placed on the same height as the center of the pattern (10.5cm). FOV: Standard (Dewarp), Exposure: Auto, White Balance: Auto.

Original files here on Google Drive:  Original photos

I'm not sure if they get compressed when uploaded here.

Rockel83 Posted at 10-13 06:50
Oh cool!

Well from here behind my screen the distances 30cm and up seem to be quite clear. But hadn't expected it would "look that far away".

Yes, they got pretty small. Also printed the pattern in original size on A3. Not everyone has this possibility to print this size though but here are the results. Original files on Google Drive here in case they are compressed when uploaded here:  Original files here

A3 10cm


A3 20cm


A3 30cm


A3 40cm


A3 50cm

aliveathome Posted at 10-13 06:50
Original files here on Google Drive:  Original photos

I'm not sure if they get compressed when uploaded here.

Thanks for sharing!

I've tried to download them, open them with Photoshop and zoom in to all of them with 100%/150%. But it's hard to tell like this too.

It seems shrp without zooming, but when zooming in, grain is becomming vissable and makes it therefore hard to say if it's sharp or blurry.
Maybe I'm doing things wrong here.
"I'm not sure if they get compressed when uploaded here."
Looking to the file size, I suppose they're an exact copy?


(BTW, you're posting images with a box with an adress on it to the left, not sure if it's your adress ofcourse, but maybe something to keep in mind when sharing on the internet)

Rockel83 Posted at 10-13 07:27
Thanks for sharing!

I've tried to download them, open them with Photoshop and zoom in to all of them with 100%/150%. But it's hard to tell like this too.

Yes, I did the same and agree with you, it's hard to tell if its bad or good   Would be interesting to see someone elses photos of the same pattern. Maybe my camera is fine after all?

Yes I know, no it's not my adress, its a company a bought some stuff from a long ago. I kept it there just to get the text in the picture. Thanks for the heads up anyway

aliveathome Posted at 10-13 07:59
Yes, I did the same and agree with you, it's hard to tell if its bad or good   Would be interesting to see someone elses photos of the same pattern. Maybe my camera is fine after all?

Yes I know, no it's not my adress, its a company a bought some stuff from a long ago. I kept it there just to get the text in the picture. Thanks for the heads up anyway

Well, I thought it could have been a good neutral test to test like this. So everybody will be able to perform the same test.
But did not expected that it would look that far away.

Zooming in seems to be not that great option too, because it seems to get more grainy fast then. And not all the software will behave the same whe zooming in. I've used a few fotoplayers before to arrange some holliday foto's and they all looked different when zooming in. So therefore I thought that Photoshop would have been a reliable option here.

Maybe the use of a "universal" chart is usable. But maybe we will need some refference points to compare it with? Like if someone could perform the exact same test with the same chart with other camera's> Let's say the GoPro's or even the OA1?

Maybe it can be also usable to use a less detaild chart? What do you think of the 3 options I've put in one of my previous message? I believe the last one is exactly the same as that you're using here.
But again, your tests seem to be pretty sharp!

When I check this camera on Amazon.de, then this camera gets a 4,5 out of 5 (almost 7.000 reviews).
I.m not sure if this are only German reviews, but 7.000 is massive already. So worldwide should have been much more.

Let's say that all the camera's went through the same production process, then more people should have noticed problems out of these 7.000+ ?
Although can it be hard to tell if you don't have a refference ofcourse.

---DELETED---

---DELETED---

osmonauta Posted at 10-13 10:10
Sadly, Amazon is known to have a lot of fake/paid reviews so it's best to be careful. I always look at the 1-2 star reviews. The rest I don't care about.

Oh I didn't knew that.
But I was actually checking the 1 and 2 star reviews too. To see if something mentions something about focus. But I could not filter much out of this particular problem.

---DELETED---

Kickbacks galore.  Has been going on for years.

https://www.ft.com/content/bb03ba1c-add3-4440-9bf2-2a65566aef4a

Check new fw 01.01.05.40

I posted some comparison photos here taken before and after the update:

https://forum.dji.com/forum.php? ... 055&pid=2877282

aliveathome Posted at 10-13 23:46
I posted some comparison photos here taken before and after the update:

https://forum.dji.com/forum.php?mod=redirect&goto=findpost&ptid=277055&pid=2877282

It looks sharper, for sure!
But I'm wondering if they've just increased sharpness (same as you can achieve on postfocus) or if there was really a focus problem.

I hope not, but compensating with sharpness does ofcouse not solve the focus problem.

But your images look pretty good so far I guess!
I believe your images were already verry clear compared to images other people showed. So that get me thinking there're already diffrent kind of batches on the market.

More Tests here: https://youtu.be/Qrfu4jn5CdE


You can use a diopter to correct the bad focus, hopefully there is a better solution from DJI

Yesterday i had my copy in a real world test. Outside temperature aprox. 20°C
I filmed at 2,7K 60fps - Camera overheated and stopped filming. I was Not even replay was possible...!!!
After seeing SO MANY (i guess well payed) videos of how great this camera should be, i was shocked!

Than i turned the resolution down to 1080P and 60FPS and the camera overheated AGAIN!!!
Latest version of FW was installed on 14.10.22. via Mimo App.
I HATE thate i have to get an APP to "activate" an action camera!?!?!?! WHAT???? WHY DJI, why???

Tomorrow i will film another Overheating test. The blury image is the second problem...
So after the test tomorrow i will have to send my copy back because the adventure pack is to expensive to keep it and not use it.

I can not belive how a company that produced this GREAT DJI MIC that i own (and LOVE),  can bring the Action 3 in a somewhat unfinished condition to the market.
DJI, did you have to rush?

So my action 3 is blurry. But I love everything else. This Youtuber at 1:48 used a +4.0 diopter lens and the image was clear. DJI if you are reading this, Please just make a + 4.0 diopter protective lens that we can replace with the original one. And please give it to those who activated/registered the device for free. that way we don't need to return the product. Win win.

djiuser_DAIKvCfgBJ30 Posted at 10-14 11:44
More Tests here: https://youtu.be/Qrfu4jn5CdE
https://youtu.be/Qrfu4jn5CdE

is there a way we can start a petition to make DJI give us a + 4.0 diopter protective lens?

andy741231 Posted at 10-16 18:35
is there a way we can start a petition to make DJI give us a + 4.0 diopter protective lens?

Only solves part of the issue unfortunately. What about when using ND-filter, then you can't have the protective lense on.  On the other hand you often use ND-filter to achieve motion blur so maybe the built in blur doesn't matter then?

aliveathome Posted at 10-16 21:34
Only solves part of the issue unfortunately. What about when using ND-filter, then you can't have the protective lense on.  On the other hand you often use ND-filter to achieve motion blur so maybe the built in blur doesn't matter then?

You can still use pgytech nd filter. It sits on top of the protective lens. For other brands, there can be a male to male protective lens. It's not compilecated.