5Dmk4 vs A6300 vs X-T2 vs GX85 (GH4) 4K Detail & Lowlight Comparison

Aug 11, 2016
46
0
4,881
https://www.youtube.com/watch?v=3n7Qvu1KMt8
:-\

jv6xdf.jpg
 
One of the reasons for my buying a 5DIV was that my 5Ds R occasionally gives Moire when
I am so close to birds that the feather barbs give some Moire. It's only happened a couple of times in 20,000. But, the incredibly sharp 400mm DO doesn't blur enough. Otherwise, the 5Ds R is stunning in terms of sharpness.
 
Upvote 0
I know nothing much about video, so apologies if this is a stupid question, but surely the AA filter works on too low a level to show up in video as softness? Given that video is already downsampled, or whatever the correct term is? Even 4K?

Like, if you took a photo with the 5Ds and the 5DsR, and downsampled them both to 4K resolution, any difference in 1:1 sharpness in the original images would be swamped by the reduction in output size? Am I missing something?
 
Upvote 0
scyrene said:
I know nothing much about video, so apologies if this is a stupid question, but surely the AA filter works on too low a level to show up in video as softness? Given that video is already downsampled, or whatever the correct term is? Even 4K?

Like, if you took a photo with the 5Ds and the 5DsR, and downsampled them both to 4K resolution, any difference in 1:1 sharpness in the original images would be swamped by the reduction in output size? Am I missing something?

4K from the 5D IV isn't being downsampled. It's a 4096*2160 crop of the centre of the sensor, which is why some people are unhappy about the crop factor.
 
Upvote 0
davidj said:
scyrene said:
I know nothing much about video, so apologies if this is a stupid question, but surely the AA filter works on too low a level to show up in video as softness? Given that video is already downsampled, or whatever the correct term is? Even 4K?

Like, if you took a photo with the 5Ds and the 5DsR, and downsampled them both to 4K resolution, any difference in 1:1 sharpness in the original images would be swamped by the reduction in output size? Am I missing something?

4K from the 5D IV isn't being downsampled. It's a 4096*2160 crop of the centre of the sensor, which is why some people are unhappy about the crop factor.

Is that right? I thought it was a roughly 1.6x crop. 4096x2160=8.8MP. The 5D4's sensor is 30MP. Crop factor is a linear measure, right? So is the calculation not roughly ((√30MP)/1.6)²? Which is over 11MP. I'm not great at maths though :-\
 
Upvote 0
I'm beginning to think that there is something in the setup that was missed. I find it really hard to believe the image can be that soft. Also, it could be a software issue between Canon files and editing software.
So many sample videos at 720P that's it's hard to really tell.
 
Upvote 0
How about using a canon lens with this camera?
After all people shooting a 5d m4 are probably going to use canon L glass or eqivalent

We all know glass is king, so pair up a kingly lens to the kingly camera for a proper comparison
 
Upvote 0
njene said:
How about using a canon lens with this camera?
After all people shooting a 5d m4 are probably going to use canon L glass or eqivalent

We all know glass is king, so pair up a kingly lens to the kingly camera for a proper comparison
He is testing the 4k movie quality rather than len quality, that's why same sigma APSC len is used.
 
Upvote 0
Hi,
I think he should double check his settings or try using an EF lens to verified the result again.

Anyway, if I were a reviewer and I found some big issue with a high profile camera, I'll make sure I done all I can to verified the result include getting a 2nd unit or use other lens on it before I post the result... credibility is very importance and once lost, it hard to gain back.

Have a nice day.
 
Upvote 0
raptor3x said:
scyrene said:
Is that right? I thought it was a roughly 1.6x crop. 4096x2160=8.8MP. The 5D4's sensor is 30MP. Crop factor is a linear measure, right? So is the calculation not roughly ((√30MP)/1.6)²? Which is over 11MP. I'm not great at maths though :-\

You also have to take the aspect ratio difference into account, hence the 1.76x crop.

Ooh of course. I knew there was something I was missing!
 
Upvote 0
pokerz said:
njene said:
How about using a canon lens with this camera?
After all people shooting a 5d m4 are probably going to use canon L glass or eqivalent

We all know glass is king, so pair up a kingly lens to the kingly camera for a proper comparison
He is testing the 4k movie quality rather than len quality, that's why same sigma APSC len is used.

This is where I had concerns...since the 5D mark IV doesn't downsample the entire sensor...it seems to me like individual pixel sharpness would matter to a much greater degree than on competing cameras. If you get one pixel of blur on the 5D Mark IV that would show up...if you got it on a competing camera, it would potentially be masked out by the downsampling routine.

I don't know why Canon doesn't downsample and I'm not video expert...but it does bring into question if that's the best lens to use on the 5D.
 
Upvote 0
pokerz said:
njene said:
How about using a canon lens with this camera?
After all people shooting a 5d m4 are probably going to use canon L glass or eqivalent

We all know glass is king, so pair up a kingly lens to the kingly camera for a proper comparison
He is testing the 4k movie quality rather than len quality, that's why same sigma APSC len is used.

Doesnt matter what hes testing - the lens used makes a difference
The test method is flawed
See also: http://www.canonrumors.com/eos-5d-mark-iv-compatibility-warning-from-sigma/
 
Upvote 0
I went chasing Golden Eagles in the mountains yesterday. There was a young eagle hovering nearby. Then suddenly I saw a buzzard starting to skydive. The following 4 pics were taken at a distance of approx. 200 m just in the original size (just a 100% crop of 600x400 pixels - just to say, how small the bird on the original photo is! ;) ). I think it is still amazing to see how the skydive begins. And last two shots of the eagle as well. I am not sure about the last picture. Looks like the eagle has swallowed a ball... :) Maybe these are the remainings of a groundhog.
 

Attachments

  • 911_2550s.jpg
    911_2550s.jpg
    136.7 KB · Views: 785
  • 911_2553s.jpg
    911_2553s.jpg
    134 KB · Views: 795
  • 911_2556s.jpg
    911_2556s.jpg
    134.8 KB · Views: 792
  • 911_2558s.jpg
    911_2558s.jpg
    135.1 KB · Views: 782
  • 911_2273s.jpg
    911_2273s.jpg
    433.8 KB · Views: 162
  • 911_2297s.jpg
    911_2297s.jpg
    465.9 KB · Views: 174
Upvote 0