Hello Fellow Yuneec Pilot!
Join our free Yuneec community and remove this annoying banner!
Sign up

Has anyone seen a good video or pics of H+ vs H3 cameras ?

Joined
Nov 17, 2019
Messages
17
Reaction score
3
Age
33
If H+ cam and lens as good/equal to H3 then it does substantiate that the H3 was really a ploy to get mo $$/and/or introduce new approaches to some things.

any revealing vids or pic sets ??

TIA
 
Nothing of any substance (photo or video) has been posted on this forum or Youtube (to my knowledge) so as to ascertain how the H3 stacks up against either the H Plus/C23 or 520/E90.

What we do know, is that the bigger 1 inch sensors found in several different drones such as the DJI P4P, Mavic 2 Pro, Typhoon H Plus and Typhoon H 520 - are all very closely matched in terms of imagery - both photo and video. The only real differences come down to things like user interface, system qualities and aircraft capabilities. For instance, the P4P has a mechanical shutter while the Plus doesn't but - when one knows where and how a mechanical shutter has a slight advantage, it may be a small factor in the overall scheme of things.

I am in the process of making a (sort of) head to head video with an H Plus against Mavic 2 Pro and spoiler alert: ;) lets just say the imagery from the H Plus is hard to beat.

Knowing the limitations of a flying camera, and the fact that the airframe of H3 is identical to the Plus, therefore it's payload is identical, I can't imagine they've gotten a much more out of the L1 than the C23 - after all it is almost a certainty that the sensor is the same one in the Plus. Maybe a different color profile, possibly a slightly better lens but I doubt it. From this point it comes down to the periphery things such as interface and little bells and whistles such as adding back cruise control or others.

I would like to be proven wrong and perhaps one day we will have someone upload a Raw Photo, so we can pull it into our own environment and see how it compares to one of ours but so far this has not happened. I would love to get my hands on one but for me, the price tag is a bridge too far over a river of typical Yuneec mystery.
 
Great point Steve, in photo mode it is likely that we would have our drone steady in the air, framing our subject and at a somewhat high shutter speed thus making the mechanical shutter of little advantage over a rolling shutter like in a Typhoon.
 
Took my H3 out to the top of a hill, put it on the roof of my car alongside an H520 with E90 camera and took some side-by-side photos and short video.
I set the H3's ION L1 camera to Auto and then matched the E90 settings to it as near as possible and took these two photos (raw jpeg photos):
E90:
E90 a YUN_0022.JPG

ION L1:
ION L1 a YUN_0074.JPG
(just a few minutes apart - fast moving clouds!)
I am not a photographer (mech. engineer who flies drones) so I would appreciate some comments on how much better the L1 is than the E90. The detail looks the same to me. I use my H520 for surveying and I am looking for a lightweight drone to replace my Inspire 1 X5. Not sure I have found it yet.
The distortion towards the edges is different.
Where the E90 makes everything bluish, the L1 seems to make it greener.
Looking through all the photos I've taken, the L1 image is sometimes more defined and sometimes it's the E90.

I do have the RAW/DNG files which I have uploaded to a Google Drive; message me if you would like to download them and I'll share them with you.
 
Last edited:
Took my H3 out to the top of a hill, put it on the roof of my car alongside an H520 with E90 camera and took some side-by-side photos and short video.
I set the H3's ION L1 camera to Auto and then matched the E90 settings to it as near as possible and took these two photos (raw jpeg photos):
E90:
View attachment 25141

ION L1:
View attachment 25142
(just a few minutes apart - fast moving clouds!)
I am not a photographer (mech. engineer who flies drones) so I would appreciate some comments on how much better the L1 is than the E90. The detail looks the same to me. I use my H520 for surveying and I am looking for a lightweight drone to replace my Inspire 1 X5. Not sure I have found it yet.
The distortion towards the edges is different.
Where the E90 makes everything bluish, the L1 seems to make it greener.
Looking through all the photos I've taken, the L1 image is sometimes more defined and sometimes it's the E90.

I do have the RAW/DNG files which I have uploaded to a Google Drive; message me if you would like to download them and I'll share them with you.
I noticed that the church steeple in the far distance (on the right) is straight up/down in the E90 photo and leaning inward in the L1 photo. Also, the L1 sky has a hint of banding. Agree with you about the green and blue tints.
 
Another pair with slightly different settings:
E90:
View attachment 25143

ION L1:
View attachment 25144
I'm a mechanical engineer too. But I've gotten pretty picky about images. In this pair, the L1 may have slightly greater sharpness (distant trees) but the sky/clouds don't look as good as the E90. L1 sky looks a little ruddy when zoomed in. Both images must be very wide angle because of the similar distortion. Jpg images may have fisheye distortion corrected.
 
  • Like
Reactions: Whisperit
PM Sent.

JPGs are tuff to judge because we don't know how much compression, sharpening and so forth are done in the camera whereas a raw file will let us see and extract every ounce of detail and color available.

Right off the batt, just looking real quick at those JPG's full size - there is not a lot of difference between them - in fact if you put up three pics by each camera and didn't label them I doubt anyone could correctly tell which photo came from which camera.
 
So, I set both cameras to their "Ultra" picture quality (whatever that actually is, why the childish labels?). I did use the Distortion Correction option on the E90.

You can download the JPGs and view the EXIF data. (I don't understand most of it, but the data is almost identical between the L1 and E90 - does that mean the sensors might actually be the same??)

If you want the 38MB RAW files then PM me a gmail address and I will share them with you.
I have some brief video too.

If anyone wants a particular camera test done, I could try to do that. Problems flying in the UK at the moment with our COVID lockdown, but I'll see what I can do.
 
  • Like
Reactions: Mike Irish
Hello all. It has been a while since I have been on.
Would the E90 pictures be @ the same as the c23?
 
The E90 and C23 are generally considered to be the same camera. I don't actually know if the image processing software is the same.
 
I want to thank @Whisperit for shooting me the raw DNG files from the images he posted above, I've had a chance to bring them into Photoshop and have a very close look at just how the two camera's (E90/C23 and Ion L1) Stack up.


First and foremost, the images are shot with roughly the same composition and lighting, though the differences are enough to keep me from making a conclusive determination about a few aspects such as barrel distortion and chromatic aberration - each of which seems to appear in at least one of the cameras while the other did not exhibit these features or; to a lesser degree. In addition, it is impossible to see if the focal length/angle of the cameras are the same but I would say they are very close. A true side by side would require that the two aircraft are booted up and framing the foreground and background precisely the same and both shutters being hit simultaneously.

However I have seen enough to make some determinations about how the cameras stack up. While I can't make as good a presentation here in photos as I could in a video, for those that want to see the process I use to import, process and compare photos see This Video it will show what I am doing to bring out the full depth of these images.

I was give two photos from each camera as shown above, I selected the one set of photos that was more closely composed. Below are three sets of photos - 1 from each camera with a description of what is shown. Obviously we can't see the full size image so I am presenting them as follows:

---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------


SET 1 - Both Images are Raw DNG's reduced down using Bicubic reduction to reduce to 1000 x 666 pixels, no color or lighting corrections to the E90 and only White Balance to the L1 image to get the Temperature and Tint of the pics to look the same. It should be noted that when I copied and pasted these two values from the E90 Image to the L1 - the colors instantly matched. My suspicions are and have been that the L1 almost certainly was given a different color profile so as to separate it from previous Yuneec Cameras.

Note: The different angle of the L1 (more sky) has a slight effect on exposure, especially to the right where the light is coming from but overall the overexposed areas tend to be very close in both cameras though the E90 appears to be just slightly more exposed but not by much.



E90 - Raw DNG converted to PNG, no corrections, color or sharpening, reduced to 1000 x 666
E90_1NoFix.png



Ion L1 - Raw DNG converted to PNG, no corrections or sharpening other than White Balance matched, then the image was reduced to 1000 x 666
L1_1WBMatch.png

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------


SET 2 - Both Images as above, but this time I have also reduced the exposure and added fill light in the shadows to bring out the full depth of the captured image. Note the detail in the clouds while maintaining the detail in the shadows. In addition, clarity and viberance were bumped about 25% to these images


E90

E90_2.png


Ion L1

L1_2.png


------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Set 3 - In this set of images both are as above, but to see the detail, the images are left at full size and an area of roughly 1000 x 756 was cropped out and exported as a PNG so as to show how these images compare. I cropped both of these images right at where they were overexposed and at great distance to demonstrate the cameras capabilities in the most difficult areas. Note: Look at the furthest horizon way at the back, notice the detail on the land, particularly the very tall antenna and other structures from each image. Also compare the details in the cloud formations.

E90

E90_3.png


Ion L1

L1_3.png

-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------




As mentioned, there is a lot we can tell from these images but, there are still just a few questions that may still be answered but, in my estimation, (and this is just my opinion), ;) there is not much between these two cameras. Here are some things that stand out to me.

Distortion - I definitely see a slight barrel distortion in the Ion L1 – in the last set of photos look at the fence line and the large tree and compare this area on both images and you’ll see that the L1 has a lean. Also look at the tree line in the second set of images – the Ion L1 tree line has a greater curve the E90. This is partly due to the camera be positioned at a greater angle, but it is there.

Color Profile – The color settings are different in the L1 but this is in the programming – once we even out the White Balance the cameras are almost indistinguishable from a color perspective.

Exposure – With only a couple photos to work with it is hard to make a snap call on this but the seem pretty equal though I felt like the E90 held a little more detail in the highs and lows – not a lot but it was there. However, I would want to look though several photos that were staged as mentioned in the opening to make a final call.

Chromatic aberration – The lighting of this composition (light in front of but to the side of the cameras) sets up conditions that can cause CA and here again with only this photo to go on I see just a tinge more from the Ion L1 – look at the big tree in the images from Set 3. Note I looked at the other images with the gate and came to the same conclusion.

Noise - Both images are well lite so noise/graininess is very low, however the Ion L1 does tend to exhibit more noise than does the E90 but I would point out that most noise reduction filters when properly set would be more than enough to handle all but the most difficult condition such as night time.

Overall Image – The size of the images in terms of Pixels are IDENTICAL and there file size is almost so – only kilobits separate them so we can see that the sensors are almost certainly the same.



Conclusion

I think both are very good consumer cameras considering they are slug under an aircraft. The E90 is (in my estimation) ever so slightly ahead of the L1 in terms of image quality, but only slightly and possibly only in certain conditions. Of all the drawbacks I could find there is only one that concerns me the most and that is the distortion. Anything else is up to conditions, lighting and what ones does with the final image. Indeed, a filter can probably be found or created to fix the distortion so that might not be a deal breaker.

What We have seen so far is only what the maximum of the cameras can produce in terms of single images. From experience this tells us that the video side would have similar results but that is undetermined at this point but most likely the video from these cameras would be comparable to one another.

As we are only looking at the cameras, I’ll leave the discussion as to how the cameras and platform price figure in to the whole thing for a later time

I now am anxious to see how the C23 stacks up against these cameras but from my experience with the imagery from my C23, I would say that it is almost certainly a re-labeled E90.
 
Last edited:
And just as a comparison to show the detail that can be brought out of a DNG, here I have stacked image 1 and 2 from each camera together to see the difference. On a 4K screen and the whole image at full size, this level of detail is very good.

E90 The Top image is no corrections and shows slightly more detail than the JPG would, The second is using the full depth of the sensor to cut the overexposed areas while maintaining detail - look at the back face of the wooden fence in the foreground.
E90_1NoFix.png
E90_2.png




Ion L1 Even though there is no foreground elements, the L1's DNG captures many details in the over and under exposed areas.
L1_1WBMatch.pngL1_2.png
 
Last edited:
Now I'm annoyed I didn't snap the photos at exactly the same time!

I think from Ty Pilot's analysis though, it looks like it is not worth selling your Hplus/C23 to buy an H3/L1.

What did Leica actually do?!? Loan an engineer for a few hours to tweak the settings, and take a fee from Yuneec to let them use the Leica label for marketing?? I really was expecting the sensor to be better, and the lenses too.

If the H520 had sensible gimbal control rates on the cameraman's controller in Team-mode I wouldn't bother with the H3. (Using Team-mode on the H520/E90, it is impossible to get smooth gimbal pan footage, the controls are too sensitive.)
 
On March 14, Whisperit mentioned that he used Distortion Correction on the E90. Is that option not available on the L1? How about the C23? In Ty Pilot's photos (above), distortion seems to be the biggest and most noticeable issue. I sure like how Ty color corrects the images. Thanks for giving us these comparisons.
To evaluate video quality, I have used screen captures from a 4K monitor. Pick a scene where the camera and subject are relatively still. Better yet, do the test with the drones sitting on top of your car or some other high perch.
 
Now I'm annoyed I didn't snap the photos at exactly the same time!

I think from Ty Pilot's analysis though, it looks like it is not worth selling your Hplus/C23 to buy an H3/L1.

What did Leica actually do?!? Loan an engineer for a few hours to tweak the settings, and take a fee from Yuneec to let them use the Leica label for marketing?? I really was expecting the sensor to be better, and the lenses too.

If the H520 had sensible gimbal control rates on the cameraman's controller in Team-mode I wouldn't bother with the H3. (Using Team-mode on the H520/E90, it is impossible to get smooth gimbal pan footage, the controls are too sensitive.)

One thing I would add to this conversation (and it is coming from the photographer side of me rather than the drone flyer) ;)

I have observed not only the expansion and improvement of drone over the years but also that of cameras. I would point out that a 20 megapixel - 1 inch sensor has been (and still is) sort of a plateau. 12 mega pixels was the standard for small cameras like on drones such as the Typhoon 480 (and still is for modern phones) and the next step is the Sony 1 inch sensor. I believe it first showed up on the Inspire 5 or six years ago and since then we have seen it added to the phantom line in the Phantom 4, then about the same time the Mavic 2 and H 520 got it then the Plus and now (or soon the Autel brand puts their 1 inch bird out.

The point is - there is still nothing out there (in this price range of around 1500 - 2500 bucks) that will surpass this level of camera. I have a Mavic 2 and while it's camera is pretty good and has a far superior interface, adjustable aperture and more - the final image ain't any better than my Plus. For this reason I have hung on and heralded the H Plus/ 520 and now H3 as the apex of the Typhoon line.

The reason that I don't think will see a 40 MP sensor stuffed into a sub 2500 dollar drone is that is would obsolete the higher end stuff over night. So as you point out they do things like branding the H3 as 'Leica' equipped to compete with the Hasselblad of the Mavic. Then Autel drops their 48 MP - 8K camera that is just a 12 MP sensor with some processing that currently doesn't add a lot to the party because it is still in it's infancy. However Adobe has some new software that does take a digital image and does a 400% boost and it is showing promise.

Oh and in regards to the camera man's control rates. :cool: we found that on the Plus, in CCC Mode and with the pan knob in global - giving camera control to the right stick - just like in team mode - the rate switch changes the yaw and pitch rates of the camera so we can can dial in just the right speed of movement. I have never tested team mode on a 520 but the next time you do, have your camera man adjust the rate slider of his ST-16S and see if it does change the camera control rates.
 
(or soon the Autel brand puts their 1 inch bird out.

The point is - there is still nothing out there (in this price range of around 1500 - 2500 bucks) that will surpass this level of camera. I have a Mavic 2 and while it's camera is pretty good and has a far superior interface, adjustable aperture and more -
I bought the Autel EVO II Pro with 1" sensor and adjustable aperture (cost $1600 to $1800). I'm very happy with the video quality, especially in H.265, 4K/30, LOG where you get 10-bit color and 120 Mbps. Also, there is very little, if any, residual image distortion. To take full advantage of the image quality, you need a powerful computer and editor. DaVinci Resolve 16 free version or Studio version ($300) works well once you learn how to use it. I'd always wanted a Typhoon Plus for that wonderful C23 camera, but bought the Autel because it is more compact. I apologize for discussing the non-DJI alternative brand here. I like this forum for the in-depth technical discussions.
 
Like you, I had gotten to the point where I just needed a compact drone to take to places I had difficulty packing the H Plus to, and at the time, there was really only one choice as Autel hadn't released that second one with the big sensor or I may have just gone that route too - I am no fan of the dark side, ;)
 

New Posts

Members online

No members online now.

Forum statistics

Threads
20,955
Messages
241,593
Members
27,286
Latest member
lahorelaptop