Jump to content

Panasonic GH5 - all is revealed!


Andrew Reid
 Share

Recommended Posts

5 minutes ago, webrunner5 said:

Sure it pumps, but it looks better than the GH4. It is reacting faster because it has more computing HP.  It has its own focus computer. Do we really think it was going to be like DPAF? I am afraid we have to buy a Canon C100, C300 if you want to have reasonable AF.  Panasonic cameras are not state of the art with contrast-detect only, probably never be.

I don't have one on order and I doubt I ever will. They, Panasonic has dug their heels in on their focus Mumbo Jumbo, but to be honest Canon has everyone by the ass patents wise. So I guess it is better than nothing, but it still is pretty useless in the long run for video.

You said it looks good (actually 'pretty damn good' even). It doesn't. It looks improved but still sucks. That's different. :grin: I'd rather opt not to use it at all if it's like that. Like I've been with Panasonics like since the GH2. But maybe I just have control issues and am bordering on pixel peeping disorder.

Link to comment
Share on other sites

EOSHD Pro Color 5 for Sony cameras EOSHD Z LOG for Nikon CamerasEOSHD C-LOG and Film Profiles for All Canon DSLRs
3 minutes ago, Cinegain said:

You said it looks good (actually 'pretty damn good' even). It doesn't. It looks improved but still sucks. That's different. :grin: I'd rather opt not to use it at all if it's like that. Like I've been with Panasonics like since the GH2. But maybe I just have control issues and am bordering on pixel peeping disorder.

Well I did say "So I guess it is better than nothing, but it still is pretty useless in the long run for video." You have to figure not long ago you would have your balls cut off if you even mentioned AF and Video in the same breath. I think it is coming like it or not, and if it works damn well I would Like it!

Link to comment
Share on other sites

10 hours ago, Hanriverprod said:

What do you guys think about this?

I've seen more effective GH5 AF samples than this. Not sure what they did that caused that degree of hunting. Perhaps they should not have increased focus speed as they did. There are many settings for AF, so for this type of subject matter I don't recall what the best settings were. The low light examples looked quite good.

2 hours ago, jonpais said:

For subjects moving toward the camera, it may very well be unusable. What I'm interested in knowing, since I intend to use the camera for vlogging from my home, is whether or not it can even stay focused on me for all of ten minutes. I've even seen Sony and Fuji fail this test, so I'm not getting my hopes up.

For something like that, with a stationary object sitting in front of the camera for 10 minutes, MF is the obvious choice.

Link to comment
Share on other sites

25 minutes ago, Ken Ross said:

I've seen more effective GH5 AF samples than this. Not sure what they did that caused that degree of hunting. Perhaps they should not have increased focus speed as they did. There are many settings for AF, so for this type of subject matter I don't recall what the best settings were.

For something like that, with a stationary object sitting in front of the camera for 10 minutes, MF is the obvious choice.

I agree, which is why I often ask myself why so many camera reviewers, who pretty much stay seated in the same position or standing for 30 minutes even use AF. And inevitably there is hunting going on, sometimes just a little, sometimes a lot. There must be a reason for it, it just escapes me.

Link to comment
Share on other sites

21 minutes ago, webrunner5 said:

I don't know if this has been shown on here, I don't follow everything, well sort of. :grimace:

https://***URL removed***/news/1972501860/panasonic-gh5-autofocus

So the interesting thing is their review, was that by using the appropriate AF settings, they claimed they were able to cut their out-of-focus shots by 1/2. They went from a 60% hit rate at the default setting to an 85% hit rate with their revised settings. Of course since they tend not to be 'videocentric' I didn't see anything really relating to video...unless I missed it.

Link to comment
Share on other sites

2 minutes ago, Ken Ross said:

So the interesting thing is their review, was that by using the appropriate AF settings, they claimed they were able to cut their out-of-focus shots by 1/2. They went from a 60% hit rate at the default setting to an 85% hit rate with their revised settings. Of course since they tend not to be 'videocentric' I didn't see anything really relating to video...unless I missed it.

I don't think you missed anything. I couldn't find similar tests for video AF in their review either.

Link to comment
Share on other sites

I think the biggest distraction of the AF is the reverse focus pull, aka focus confirmation rack.

I also believe it can be corrected in software.

1) DFD should calculate mount of blur, determinates focus rack. 2) focus rack smoothly starts. 3) 1/2 way through rack DFD calculates new focus rack, taking into account the movement vector of the object 4) DFD notices very little blur, racking eases off. 5) DFD notices no blurr, racking stops. Basically DFD measuring, focus motor responding depending on how far the focus is off.

The point is DFD should be used while auto racking focus.

Currently it looks like DFD is only used to determine which direction to start the focus rack, not the amount. It racks focus until DFD has noticed the rack overshoots a certain amount, then it racks back that amount.

If Panasonic can calculate DFD 240 times a second, I think they can calculate subject distance, relative speed, and to smoothly rack focus to the distance that is calculated 240 times a second from noticing the change in blur. Overshooting and racking back is ok for slow individual photos, but NOT for video. Actually when I think of it, I rather have 1 DFD focus rack estimate and rack and stop, then have rack stop and reverse. Man I would love to see how this software is designed. I think amping up the amount of measurements is never going to prevent the focus confirmation rack.

Link to comment
Share on other sites

31 minutes ago, Cas1 said:

I think the biggest distraction of the AF is the reverse focus pull, aka focus confirmation rack.

I also believe it can be corrected in software.

If Panasonic can calculate DFD 240 times a second, I think they can calculate subject distance, relative speed, and to smoothly rack focus to the distance that is calculated 240 times a second from noticing the change in blur. Overshooting and racking back is ok for slow individual photos, but NOT for video. Actually when I think of it, I rather have 1 DFD focus rack estimate and rack and stop, then have rack stop and reverse. Man I would love to see how this software is designed. I think amping up the amount of measurements is never going to prevent the focus confirmation rack.

Well, there is good video about the GH5 AF on YouTube :

 

On the AF speed issue:

gh4: 240 times a second but only each 6th frame for AF. This results in 40 times a second only

gh5: 480 times a second and each frame

-> it's 40 vs 480 times a Second for AF confirmation. 

The GH5 is the first Pana that has the next generation DFD built in.

 

B

 

Link to comment
Share on other sites

1 hour ago, Borbarad said:

gh5: 480 times a second and each frame

-> it's 40 vs 480 times a Second for AF confirmation. 

The GH5 is the first Pana that has the next generation DFD built in.

Yes in photos but in VIDEO:

24P - 24 AF samples/second

30P - 30 AF samples/second

60P - 60 AF samples/second

Link to comment
Share on other sites

 

55 minutes ago, Vesku said:

Yes in photos but in VIDEO:

24P - 24 AF samples/second

30P - 30 AF samples/second

60P - 60 AF samples/second

 

Makes no sense.

in return this would mean the camera shots 480 FPS and it doesn't. It does 9 FPS but reads the sensor still 480 times for AF confirmation.

So why shouldn't  the GH5 shoot photos at 4K resolution at 24/30/50/60 FPS and still reads the sensor at 480 times?

B

Link to comment
Share on other sites

11 minutes ago, Borbarad said:

Makes no sense.

in return this would mean the camera shots 480 FPS and it doesn't. It does 9 FPS but reads the sensor still 480 times for AF confirmation.

So why shouldn't  the GH5 shoot photos at 4K resolution at 24/30/50/60 FPS and still reads the sensor at 480 times?

B

With photos the camera reads sensor differently before final exposure. It may read only the focus box area very fast or with lower resolution or with less color for DFD AF.

In video the sensor works in some frame rate and I dont think it does different reading modes between video frames for AF. I dont know it for sure but if someone knows better it would be nice to hear. I may be wrong. Seems still that GH4 and GH5 auto focuses faster at 60P than at 24P at same resolution.

Link to comment
Share on other sites

28 minutes ago, Dimitris Stasinos said:

Contrast analysis has nothing to do with fps. It is a separated task and in many cameras there is a dedicated chip for this. This is like bonding fps with EVF's refresh rate. Different things.

But there is no other sensor for AF. How a sensor can read different fps for video and for AF. What if the shutter angle is 360, it uses the whole time exposing video frame.

Link to comment
Share on other sites

17 minutes ago, Vesku said:

But there is no other sensor for AF. How a sensor can read different fps for video and for AF. What if the shutter angle is 360, it uses the whole time exposing video frame.

In simple terms a Video is nothing more than single photos stacked together . So what the sensor does it takes single images at a the reduced 4K resolution (about 8MP) between 24 and 60 FPS with an electronic shutter or it takes full resolution (about 20MP) at max.  9FPS via electric or mechanical shutter. that's all.

AF confirmation is not a single frame like a photo taken and recorded, it's in simple terms like checking the status of a pixel or pixel area (I.e. what does the pixel see; or like asking "are you there?" 480 times per second)

B

Link to comment
Share on other sites

11 minutes ago, Borbarad said:

AF confirmation is not a single frame like a photo taken and recorded, it's in simple terms like checking the status of a pixel or pixel area (I.e. what does the pixel see).

B

This makes sense. The sensor in video is illuminated all the time and the rolling shutter picks lines for video frame. The camera still sees the sensor all the time and can pick other than actual video frame data and with different speed and area. I thought the camera analyses only the video data. Thanks.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

  • EOSHD Pro Color 5 for All Sony cameras
    EOSHD C-LOG and Film Profiles for All Canon DSLRs
    EOSHD Dynamic Range Enhancer for H.264/H.265
×
×
  • Create New...