Gordon's Canon R6 II vs Lumix S5 II vs Sony A7 IV video quality comparison

[John Matthews]

I thought I might share this. It would be interesting to hear some thoughts on resolution, moiré, aliasing, etc.

 

[Beritar]

2 hours ago, John Matthews said:

I thought I might share this. It would be interesting to hear some thoughts on resolution, moiré, aliasing, etc.

 

Interesting test but I own a lot of camera and there is much more to said about image quality.
The way the details are altered with noise reduction even at low ISO is also important. To my eyes, my S5II has a very bad rendering in 4K, even more with the standard profiles. Yes 6K and 6K Open Gate are better, but the image still looks more processed than on my S1.  And another issue with Open Gate (and even 6K) is the IBIS which is worse (mainly while walking) than in 4K.

 

[Ninpo33]

8 hours ago, Beritar said:

Interesting test but I own a lot of camera and there is much more to said about image quality.
The way the details are altered with noise reduction even at low ISO is also important. To my eyes, my S5II has a very bad rendering in 4K, even more with the standard profiles. Yes 6K and 6K Open Gate are better, but the image still looks more processed than on my S1. 

Hopefully with the next round of LUMIX pro flagship cameras we will get an OLPF filter choice again and we will hit a sweet spot between the older S cameras IQ and the newer S5ii cameras features. 

[MrSMW]

Watched on 16" MacBook full screen and my takeaway/opinion was:

Best detail 4k 25p = Canon

Best detail 4k 50p = Canon

Best low light 4k = Sony

Best colour = Lumix (which surprised me)

Best DR = Lumix and worst DR = Canon

Best image of the lot, Lumix 6k 25p, especially in open gate.

Does the 6k mode reduce IBIS ability slightly? Maybe but as Lumix already has arguably the best IBIS as a starting point, it's probably comparable at least with Canon and better than the Sony which most reckon is still a bit behind.

Lumix 6k 25p open gate for the win then. I prefer 30p for various reasons.

[John Matthews]

1 hour ago, MrSMW said:

Best image of the lot, Lumix 6k 25p, especially in open gate.

Does the 6k mode reduce IBIS ability slightly? Maybe but as Lumix already has arguably the best IBIS as a starting point, it's probably comparable at least with Canon and better than the Sony which most reckon is still a bit behind.

Lumix 6k 25p open gate for the win then. I prefer 30p for various reasons.

The point of IBIS cannot be understated. How many people are there who practically never use a tripod? I bet the answer would be very high. Also, what surprised me the most was the 1080p, which is what I really care about. I think the S5ii bettered the others at 25fps in FF. I'm fairly certain it's a 6k-ish downsampled image, making it the best of the bunch, but they're all fairly good at 200%.

[FHDcrew]

On 8/29/2024 at 3:49 AM, MrSMW said:

it's probably comparable at least with Canon

Man Panasonic must have utterly insane IBIS. I used a canon r5 this summer and consistently filmed with IBIS + EIS on. I found if I heel-toe walked and then applied warp stabilizer it consistently looked like I filmed with a gimbal or used gyro stabilization. If the Canok R5 is that good I can only imagine how impressed I would be if I held an s5 II or a gh7 right now. Thinking of switching to an s5 II or a g9ii paired with very fast glass for that very reason. 

[MrSMW]

S5ii is very good, especially after the I think it was April firmware.

Boost is almost tripod like. I still find I need to take a deep breath, hold it and hold absolutely still and there can be a tiny bit of movement, but I have used that method to capture impromptu 2-3 minute speeches etc. Not the holding my breath part obviously as I am not a pearl diver, but otherwise, it’s really good.

Then there is the highest level which with a bit of ninja walking is close enough for me to being gimbal-like, I managed to ditch the hated thing this year.

There is a crop though in that mode but it just means I keep a bit more distance.

The one last factor is I don’t use anything wider than 28mm for the best result.

[kye]

None of them are sharp enough to make inane subject matter into thrilling content.

I guess the engineers just need to work harder...  maybe next year?

[MrSMW]

5 hours ago, kye said:

None of them are sharp enough to make inane subject matter into thrilling content.

Indeed.

The only thing that bothered me about this test was that it was of a chart and who shoots charts in the real world?

I appreciate Gordon doing this and he did put various caveats in the piece, but real world, 90%+ of all video shot has people in it.

On that basis, I’d rather see at least a static head shot.

Outdoors for anything up to say 800 iso and anything higher indoors.

And lighting will/can create contrast which will add to at least the perception of sharpening.

But then as we always say, how much sharpening do we want on a face? Detail even?

Personally, I prefer as much as to the point of it not being over-sharpened so in post, when not desired, I can dial some of that back.

I’ve tried the ‘advised’ dial back the sharpening in camera, but the footage to me in post, looks like mush/shit, so leave my cameras at the default level.

[kye]

2 hours ago, MrSMW said:

Indeed.

The only thing that bothered me about this test was that it was of a chart and who shoots charts in the real world?

I appreciate Gordon doing this and he did put various caveats in the piece, but real world, 90%+ of all video shot has people in it.

On that basis, I’d rather see at least a static head shot.

Outdoors for anything up to say 800 iso and anything higher indoors.

And lighting will/can create contrast which will add to at least the perception of sharpening.

But then as we always say, how much sharpening do we want on a face? Detail even?

Personally, I prefer as much as to the point of it not being over-sharpened so in post, when not desired, I can dial some of that back.

I’ve tried the ‘advised’ dial back the sharpening in camera, but the footage to me in post, looks like mush/shit, so leave my cameras at the default level.

It just seems odd to me that people judge cameras or lenses by the resolution / sharpness.

If you're interested in making quality finished edits, looking at 100% crops from a 6K sensor is about as relevant as an article discussing the shape and size and edge profile of the ridges in the rubber of the hand-grip.

It's just disconnected from reality.

[Benjamin Hilton]

6 hours ago, kye said:

It just seems odd to me that people judge cameras or lenses by the resolution / sharpness.

If you're interested in making quality finished edits, looking at 100% crops from a 6K sensor is about as relevant as an article discussing the shape and size and edge profile of the ridges in the rubber of the hand-grip.

It's just disconnected from reality.

Very true. Cameras are really good these days, it really comes down to the user experience and which small niche set of features work best for you. Should you pick up a Lumix because they're cheaper, or a Canon because you own EF lenses? Do you want the overall feature set of the Sonys, or are you trying to mix and match with something else? Those the real considerations these days. 

I'm working on a show right now that we shot with an R7, S1, GH5, Gopro 7 and the DJI Air 2. Everything looks fine in post, it's just a matter of working out their individual quirks. The S1 is overall the best image, the cleanest and best DR. The R7 and GH5 look about the same DR wise, but the R7's color is a bit nicer. The gopro looks good, but a bit of over sharpened compression mush. The DJI isn't too bad, just some wonky magenta shifts that are a pain to deal with. At the end of the day though, no one would watch the final show and think a specific camera stands out, they would just judge the show to be what's interesting to them or not.