How do you do video mode on the gx80?

[kye]

On 6/13/2025 at 10:30 PM, Sebastien said:

But how do you achieve that, provided it is a 4/3 lens without an aperture ring (like my olympus 17 & 45 mm 1.8 and the 12-32 kit lens)?

As I understand it, the only mode that allows for locking the shutter at 1/50 and being able to adjust the aperture via the camera is the manual mode under the movie mode - which does not allow auto iso, right? (Strangely, the M mode for stills allows for auto iso). 

From what you've said I would strongly suggest you to use Manual mode and to customise the dials individually (like @ac6000cw posted above) to configure one dial to be ISO ("Sensitivity" in the manual) and Aperture on the other.

This would give you immediate control over the ISO (which you should only raise once the lens is wide open and there isn't enough exposure) and Aperture.

This gives the advantage of the exposure of the shot not changing automatically during the shot.  This is almost always something you want to avoid.  Like I explained earlier - you don't want the exposure going up and down when bright objects come in and out of frame in the background.

In situations where you do want the exposure to be adjusted during the shot, for example if you're moving from a light location to a dark one, or if lights get turned on/off, you can adjust these things with the dials.  This will result in the exposure suddenly changing during the shot, rather than it gradually transitioning (as auto-ISO will do) but this is actually an advantage in post instead of a disadvantage.

This is because if you need to adjust exposure during a shot then in post you can just chop the clip up into a few pieces (on the exact frames that the exposure changed when you changed the dials) and can automate the it from there.  The challenge you have when using auto-ISO is that every shot where the exposure is drifting up and down will need to be adjusted with curves to compensate for what the camera did, and this can take literally hours.

I shot a rodeo once using auto-ISO and a guy fell and the bull went over him, nearly treading on him, and so my framing went: him on top of the bull, pan down to him on the ground under the bull, him on the ground after the bull has run off.  The exposure was all over the place as the elements in the frame changed and the camera "helped" me with exposure.  The exposure automations I had to use to create an exposure that looked like nothing happened were complicated and took me literally hours.  Had I shot it using one exposure then I could have just used a single exposure automation to bring up the exposure would have only taken minutes, and even if I'd have adjusted aperture or ISO during the shot it would only have added a few minutes in post to chop it up and adjust each segment individually.

As you saw from my stress tests above, there is a lot of latitude in the files, so I cannot imagine that many situations where you'd want to change exposure during the shot.

I used to be a full-auto shooter, and shot like that for years, listening to people online about how doing things manually was better.  Now we have better tools in post, I have now fully switched to manual shooting as I've been through the pain of adjusting things in post to compensate for the camera wandering around.  

On 6/13/2025 at 10:30 PM, Sebastien said:

While using standard, do you leave the "sharpness", "saturation", etc. as it is, or would you recommend to dial them down a bit?

I'll  try that, keeping in mind that the standard profile offers some flexibility to adjust the colors in post. The correction that you have shown above are impressing.

Just looked.  I have -5 Contrast, -5 Sharpness and -5 NR, and 0 Saturation.

There is a whole topic about why I have set these the way I have, but that's what I recommend.

On 6/13/2025 at 10:30 PM, Sebastien said:

I'll  try that, keeping in mind that the standard profile offers some flexibility to adjust the colors in post. The correction that you have shown above are impressing.

5600K makes things look like what they actually look like.

I shot for years using auto-WB and just couldn't make things look natural in post, they always looked like something was off in some way.  This solves that issue.  I found that when shooting in available light, even doing a custom WB on a grey-card gives a worse result than just using 5600K.

I'd encourage you to set it to 5600K and carry the camera around with you in a pocket for a day with the smallest lens you can find and just take a 1s clip of every location you can find.  Then pull them all into your NLE and see how they look.  You might be surprised at how well it works across all the different situations.  

If you can, take it to some night markets where vendors are selling food from vans and people are selling low-value items.  You will find the most incredible variety of ultra-low-quality lighting imaginable, as every vendor will buy the cheapest LED lights they could find at the time.  Trying to "correct" for these lights will be futile, but the footage should still be a representation of the environment you shot in, even if it won't look like a beauty commercial.

 

A note on testing...  I shoot in similar situations to those you have mentioned, and I have come to the conclusions I have come to via lots of experience and an incredible amount of testing.

Testing is so important to getting good results because so often you are convinced of something and then do a test to verify your opinions and find that the results are radically different than what you were expecting.  So many people online are full of opinions that are so easily proven to be false with only a few minutes of real-life testing.

I'm not sure if you've come across any cinematographers doing latitude testing of a cinema camera, but it's very telling that cinematographers (whenever possible) will do camera and lens and lighting tests prior to shooting a TV series or movie, and it seems like no-one doing videography or photography does these or talks about them.  Hell, I was looking at a lens the other day and couldn't even find anyone who published test images at different apertures to see how the lens performed wide open vs stopped down.  I found lots of photography bloggers who published lots of images and had lots to say, but testing?  Nah.... 

Professionals test, amateurs guess.

[kye]

On 6/14/2025 at 4:47 PM, projectwoofer said:

Sacrilege! 😁 How come not the "hacked" CineD? Doesn't it offer a bit more latitude?

Also out of curiosity: which variable ND are you using? I'm still using 3 and 6 stop fixed ones as I'm hesitant to go with a variable one due to potential image quality loss.

I did a bunch of testing some time ago comparing them, and they were less different than I thought they would be, and the CineD and Natural seemed to have the same latitude.  I'm not sure if somewhere along the line I got confused between Natural and Standard though, so that might be something to test.

Years on, the conclusion I've come to about colour profiles is that if you're going to colour grade in post with any kind of sophistication (and now with the Film Look Creator tool and Resolve colour management we have incredibly sophisticated tools) then it probably doesn't matter which profile you use.

If I was limited to basic tools then I might just use CineD and be done with it.

It's just preference really.

The vND I'm using is the "K&F Concept 58mm True Color Variable ND2-32 (1-5 Stops) ND Lens Filter".  They offer one with a larger range, but it isn't the True Colour one so I suspect it has more colour shifts.  I've been really happy with it and my tests didn't show any colour shifts.  I asked some professional cinematographers for advice and they said this was the cheapest one they'd recommend, so I suspect this is entry level.  The NiSi ones were also recommended, but they're significantly more expensive.

As always, do your own research, but if you're curious I might have my tests somewhere.