More camera focus questions.

I mentioned that the gimbal lock was not on
Well RayH, these people are not too careful. Remember that I sent my P3P to DJI to replace my out of focus camera, I was very careful packing the machine to send it in. The package they used to send it back was very poor, I found the gimbal lock off, and the remote control unit was moving around the phantom. At least, all was working when I set it up back.
By the way, let us know about the focus/sharpness of the new camera. Thank you
 
  • Like
Reactions: RayH
Well, just been to pick it up and this is how the box was!!!! It looks like it's had a good drop. I've told them to get me another one as this is not acceptable. Would you have returned it?
b387a8f7e0eda4b609a4647e58de4759.jpg
d5f02c4be00abc7b5f5f9328e0e3fd28.jpg



Sent from my iPhone using PhantomPilots mobile app
 
Another 3 days until the third one arrives [emoji849]. I might just get a refund and if the price and specs of the P4 are ok I might buy that and become a Beta tester.


Sent from my iPhone using PhantomPilots mobile app
 
I received my third P3P yesterday and it looked fine.
I updated all of the firmware and calibrated the gimbal and IMU then proceeded to take some photos............All I can say is that the image quality was better but not what I expected for £1200 (including a spare battery and backpack).
I have boxed it back up and got a full refund.
I am now ordering the Phantom 4 as I have been told by Ken that the difference in image quality is noticeable and is better than the P3P which will do for me.
If the P3P was cheaper in the UK I might of stuck with it but for £22 more can get the P4.
Thanks to everyone who has helped my with sample photos and advice it has been much appreciated.


Sent from my iPad using PhantomPilots mobile app
 
...I would use a minimum of 1/250 second in flight, anything lower will introduce motion blur especially if you get lower than 1/150 second.

I looked through 5 pages of thread before anybody mentioned this. I hate to say it but this is the most obvious issue to me.

You are shooting photos from a floating camera. When I shoot many stills with my SLR I mount the camera on a carbon fiber tripod with a top tier ball head. Shooting with a long lens and I'm balanced on a gimbal. Point is I am taking every precaution to eliminate ALL chance of even the slightest motion. When I shoot video with a GoPro my mount of choice is a custom welded aluminum plate mount the the frame of my car. Same thing.... Take out all vibration.

Point is, you have to expect some softness with shots from the phantom. Let's face it, no matter how good the gimball correction is, the quad copter is constantly moving. With just a 12MP sensor, what is not likely the greatest lens in the world plus the addition of the lack of stability, you have to expect some level of blur.

I if shooting stills, I'd use no filters except perhaps a Circular Polarizer for color and reduction of glare. Then set the ISO at 100 to 200 and shoot at a good shutter speed. I'd be thinking 1/500s or faster. Then I'd shoot bursts of shots of the same subject making sure that I got a shot with the least motion. Bottom line.... If you are set for video and have the camera shooting at twice the frame rate (what I hear most say is best for video)... I'd say you're lucky if you get a sharp shot. Setup is simply not conducive to a sharp photo.
 
  • Like
Reactions: RayH
I looked through 5 pages of thread before anybody mentioned this. I hate to say it but this is the most obvious issue to me.

You are shooting photos from a floating camera. When I shoot many stills with my SLR I mount the camera on a carbon fiber tripod with a top tier ball head. Shooting with a long lens and I'm balanced on a gimbal. Point is I am taking every precaution to eliminate ALL chance of even the slightest motion. When I shoot video with a GoPro my mount of choice is a custom welded aluminum plate mount the the frame of my car. Same thing.... Take out all vibration.

Point is, you have to expect some softness with shots from the phantom. Let's face it, no matter how good the gimball correction is, the quad copter is constantly moving. With just a 12MP sensor, what is not likely the greatest lens in the world plus the addition of the lack of stability, you have to expect some level of blur.

I if shooting stills, I'd use no filters except perhaps a Circular Polarizer for color and reduction of glare. Then set the ISO at 100 to 200 and shoot at a good shutter speed. I'd be thinking 1/500s or faster. Then I'd shoot bursts of shots of the same subject making sure that I got a shot with the least motion. Bottom line.... If you are set for video and have the camera shooting at twice the frame rate (what I hear most say is best for video)... I'd say you're lucky if you get a sharp shot. Setup is simply not conducive to a sharp photo.
A simple comparison of on the ground and in the air shots would prove this.


Sent from my iPhone using PhantomPilots mobile app
 
I looked through 5 pages of thread before anybody mentioned this. I hate to say it but this is the most obvious issue to me.

You are shooting photos from a floating camera. When I shoot many stills with my SLR I mount the camera on a carbon fiber tripod with a top tier ball head. Shooting with a long lens and I'm balanced on a gimbal. Point is I am taking every precaution to eliminate ALL chance of even the slightest motion. When I shoot video with a GoPro my mount of choice is a custom welded aluminum plate mount the the frame of my car. Same thing.... Take out all vibration.

Point is, you have to expect some softness with shots from the phantom. Let's face it, no matter how good the gimball correction is, the quad copter is constantly moving. With just a 12MP sensor, what is not likely the greatest lens in the world plus the addition of the lack of stability, you have to expect some level of blur.

I if shooting stills, I'd use no filters except perhaps a Circular Polarizer for color and reduction of glare. Then set the ISO at 100 to 200 and shoot at a good shutter speed. I'd be thinking 1/500s or faster. Then I'd shoot bursts of shots of the same subject making sure that I got a shot with the least motion. Bottom line.... If you are set for video and have the camera shooting at twice the frame rate (what I hear most say is best for video)... I'd say you're lucky if you get a sharp shot. Setup is simply not conducive to a sharp photo.
I am also a photographer (hobby) and I understand that shutter speed is the key to sharp photos but I think for the money the camera should be a lot better. Obviously the new lens on the P4 has helped with the soft and blurry images but at speed and in the wind photos will always look iffy. My tests show that even with the P3 in a static position without the motors running the images are still blurry and out of focus in some areas of the frame. It was obviously a problem with the assembly of the lens to the sensor and extremely cheep parts. The P4s lens and manufacturing has obviously improved a lot.
 
I am also a photographer (hobby) and I understand that shutter speed is the key to sharp photos but I think for the money the camera should be a lot better. Obviously the new lens on the P4 has helped with the soft and blurry images but at speed and in the wind photos will always look iffy. My tests show that even with the P3 in a static position without the motors running the images are still blurry and out of focus in some areas of the frame. It was obviously a problem with the assembly of the lens to the sensor and extremely cheep parts. The P4s lens and manufacturing has obviously improved a lot.


I have not had a chance to post shots to show the difference. Closing my quarter this week and have been nuts the past two or three.

I am sure there will be a marked difference as we adjust the shutter speed. I'll shoot some with and without filters and post shutter speed info. I'll also sharpen at different levels to see what I come up with.

As for shutter speed being the key to sharp shots..... yes and no. Sure you need to shoot at the right speed but there is no doubt the glass plays a HUGE role. I have different lenses for different styles of shots. Not all lenses product the exact same sharpness.

Having said that..... as I mentioned before.... you are paying $1K for a drone with a camera. We need to keep things in perspective. I don't expect to get a shot as sharp as I get with my top glass on my SLR. Some of those lenses are more expensive than the whole drone! Same as I would not expect my $200 point and shoot or my phone to give me shots that compare with the DSLR.

Bottom line.... we all need to set the bar at the right point.

Let me see if I can rattle off some photos and see what I come up with.
 
I have not had a chance to post shots to show the difference. Closing my quarter this week and have been nuts the past two or three.

I am sure there will be a marked difference as we adjust the shutter speed. I'll shoot some with and without filters and post shutter speed info. I'll also sharpen at different levels to see what I come up with.

As for shutter speed being the key to sharp shots..... yes and no. Sure you need to shoot at the right speed but there is no doubt the glass plays a HUGE role. I have different lenses for different styles of shots. Not all lenses product the exact same sharpness.

Having said that..... as I mentioned before.... you are paying $1K for a drone with a camera. We need to keep things in perspective. I don't expect to get a shot as sharp as I get with my top glass on my SLR. Some of those lenses are more expensive than the whole drone! Same as I would not expect my $200 point and shoot or my phone to give me shots that compare with the DSLR.

Bottom line.... we all need to set the bar at the right point.

Let me see if I can rattle off some photos and see what I come up with.
I agree with what you have said.

I think that QC should have been better at DJI as many people have complained about the out of focus images whilst I have seen perfectly acceptable images from others. I didn't expect DSLr standards but I expected the camera to perform as stated and In focus.
 
I agree with what you have said.

I think that QC should have been better at DJI as many people have complained about the out of focus images whilst I have seen perfectly acceptable images from others. I didn't expect DSLr standards but I expected the camera to perform as stated and In focus.
More than that in a 2016 piece of technology born to take shots and video that cost 1400€ i expected at least a camera comparable of a 2 years ago smartphone. Not a new one, it was enough to a 2 years old phone.
 
  • Like
Reactions: RayH
OK.... So here's the comparison shots.

These were all taken today. I gave you guys shots with a variety of textures. Some smooth lines, some trees, some grass, some pavers, reflections, ect.

All the shots were shot in RAW format (NEF files from the Nikon DNG files from the DJI). All the shots were at the full resolution the camera can produce in each case. All the shots here were process in a general process (ie I did not go in to edit pixels) with Lightroom. I adjusted the white balance at the same point for all the photos (a neutral point on the car's tire) and all had adjustments made to saturation, contrast, vibrance, etc. They were all sharpened after all other adjustments were made. The Nikon photos can take more agressive sharpening because of the fact they are 36MP images. The DJI files can handle less because they are smaller. Also, I cropped into the DJI shots slightly to try to match the framing as much as I could. All the shots have been reduced to 1600 pixels on the longest axis. This would benefit the drone shots because they require less reduction.

I am sure that I am forgetting something... but if you have a question, ask me.... no worries.

These two shots are from a Nikon D800 with a 24-70mm, f2.8 Nikon Lens. The lens was set to a focal length of 28mm. I took one shot with a polarizer and one without. The camera was on a solid tripod with a ball head and the lens axis was at about 6 feet off the ground. In short, this is the control.

This shot's data: ISO 100, 1/200s, f/8.0, Circular Polarizer on the lens.

Drone_Sharpness-1_zps53asb9kg.jpg


This second shot: ISO 400, 1/2000s, f/8.0, no polarizer on the lens.

Drone_Sharpness-2_zpsn2p06hsl.jpg


Between these two shots you should see very little difference. The main difference will be with the reflections/flare coming off the red car's quarter panel and the black car's quarter panel (in the background). That is the effect of the polarizer. Since both shots where taken from a locked down, tripod mounted camera, the shutter speed is irrelevant except for, perhaps, the leaves in the trees moving about in the wind. This shots should appear very sharp and you can use them to compare against the Phantom 4K camera shots.

Here you will see the shots moving about a bit. Reason is that the drone was bucking about in a breeze and I can only get it "close" to the same spot as the tripod mounted camera.

These are the drone photos.

The first shot is with only the polarizer in place. Figure about a 1.5 stop difference. (slower shutter)

Shot Info: ISO 100, 1/400s, f/2.8 (aperture cannot be changed)

Drone_Sharpness-3_zpskjlt1emn.jpg


The next shot has no polarizer and the shot info is: ISO 100, 1/750s, f/2.8

Drone_Sharpness-4_zpsg9tj4hxj.jpg


To increase the shutter speed, I bumped up the ISO. Since we cannot change aperture, it is the only option.

Shot info: ISO 400, 1/2000s, f/2.8

Drone_Sharpness-5_zps0k13vhjl.jpg


In these shots you will see some differences. I'll highlight a few for you that will show sharpness differences.

- In the first two drone shots, check the reflection in the rear polished wheel. If you look at the reflections, you will see the brick pavers reflections are less sharp.
- In the first two drone shots, you should also see less detail in the grass than in the latter shots and/or when compared to the DSLR shots.
- Look at the detail in the trunk of the two palm trees directly behind the red car. You'll see more detail in the DSLR shots
- Compare the red car cover from one shot to the next. Look at the creases. NOTE HERE... don't mistake more contrast (sun comes in and out) because of shadows with more sharpness. Look at the hood. where it is always evenly lit.
- Look at the pavers in the foreground and the lines. You can see the curvature in the tops of the pavers in the DSLR shots but it goes away a bit in the slower shutter drone shots.

Other things that showed up.... The increase to ISO 400 on the drone added a significant amount of noise in the photo. You can see that in the red paint. If you look at the ISO 400 shot from the drone, you will see a snow/noise effect on the paint. If you look at the DSLR shot, the color is dead smooth. Note that neither of the DSL shots show noise regardless of the ISO. This is basically a factor of the sensor size. It will typically show most in red...hence the reason I shot the red car.

Bottom line......You will see some differences in the shots. I took these all at fairly high speeds. 1/200s is on the fence really for a sharp shot. If I had it do do again, I'd drop a ND32 filter on there and shoot at ISO 100. Then you will see shutter speeds in the 1/60 to 1/120s and that will make the blur more obvious.

Hope this helps.....
 
OK.... So here's the comparison shots.

These were all taken today. I gave you guys shots with a variety of textures. Some smooth lines, some trees, some grass, some pavers, reflections, ect.

All the shots were shot in RAW format (NEF files from the Nikon DNG files from the DJI). All the shots were at the full resolution the camera can produce in each case. All the shots here were process in a general process (ie I did not go in to edit pixels) with Lightroom. I adjusted the white balance at the same point for all the photos (a neutral point on the car's tire) and all had adjustments made to saturation, contrast, vibrance, etc. They were all sharpened after all other adjustments were made. The Nikon photos can take more agressive sharpening because of the fact they are 36MP images. The DJI files can handle less because they are smaller. Also, I cropped into the DJI shots slightly to try to match the framing as much as I could. All the shots have been reduced to 1600 pixels on the longest axis. This would benefit the drone shots because they require less reduction.

I am sure that I am forgetting something... but if you have a question, ask me.... no worries.

These two shots are from a Nikon D800 with a 24-70mm, f2.8 Nikon Lens. The lens was set to a focal length of 28mm. I took one shot with a polarizer and one without. The camera was on a solid tripod with a ball head and the lens axis was at about 6 feet off the ground. In short, this is the control.

This shot's data: ISO 100, 1/200s, f/8.0, Circular Polarizer on the lens.

Drone_Sharpness-1_zps53asb9kg.jpg


This second shot: ISO 400, 1/2000s, f/8.0, no polarizer on the lens.

Drone_Sharpness-2_zpsn2p06hsl.jpg


Between these two shots you should see very little difference. The main difference will be with the reflections/flare coming off the red car's quarter panel and the black car's quarter panel (in the background). That is the effect of the polarizer. Since both shots where taken from a locked down, tripod mounted camera, the shutter speed is irrelevant except for, perhaps, the leaves in the trees moving about in the wind. This shots should appear very sharp and you can use them to compare against the Phantom 4K camera shots.

Here you will see the shots moving about a bit. Reason is that the drone was bucking about in a breeze and I can only get it "close" to the same spot as the tripod mounted camera.

These are the drone photos.

The first shot is with only the polarizer in place. Figure about a 1.5 stop difference. (slower shutter)

Shot Info: ISO 100, 1/400s, f/2.8 (aperture cannot be changed)

Drone_Sharpness-3_zpskjlt1emn.jpg


The next shot has no polarizer and the shot info is: ISO 100, 1/750s, f/2.8

Drone_Sharpness-4_zpsg9tj4hxj.jpg


To increase the shutter speed, I bumped up the ISO. Since we cannot change aperture, it is the only option.

Shot info: ISO 400, 1/2000s, f/2.8

Drone_Sharpness-5_zps0k13vhjl.jpg


In these shots you will see some differences. I'll highlight a few for you that will show sharpness differences.

- In the first two drone shots, check the reflection in the rear polished wheel. If you look at the reflections, you will see the brick pavers reflections are less sharp.
- In the first two drone shots, you should also see less detail in the grass than in the latter shots and/or when compared to the DSLR shots.
- Look at the detail in the trunk of the two palm trees directly behind the red car. You'll see more detail in the DSLR shots
- Compare the red car cover from one shot to the next. Look at the creases. NOTE HERE... don't mistake more contrast (sun comes in and out) because of shadows with more sharpness. Look at the hood. where it is always evenly lit.
- Look at the pavers in the foreground and the lines. You can see the curvature in the tops of the pavers in the DSLR shots but it goes away a bit in the slower shutter drone shots.

Other things that showed up.... The increase to ISO 400 on the drone added a significant amount of noise in the photo. You can see that in the red paint. If you look at the ISO 400 shot from the drone, you will see a snow/noise effect on the paint. If you look at the DSLR shot, the color is dead smooth. Note that neither of the DSL shots show noise regardless of the ISO. This is basically a factor of the sensor size. It will typically show most in red...hence the reason I shot the red car.

Bottom line......You will see some differences in the shots. I took these all at fairly high speeds. 1/200s is on the fence really for a sharp shot. If I had it do do again, I'd drop a ND32 filter on there and shoot at ISO 100. Then you will see shutter speeds in the 1/60 to 1/120s and that will make the blur more obvious.

Hope this helps.....
Awesome post and very well explained without being boring. Thanks [emoji106]


Sent from my iPhone using PhantomPilots mobile app
 

Members online

Forum statistics

Threads
143,066
Messages
1,467,354
Members
104,933
Latest member
mactechnic