4K DSLR with no jittery / stutter video ? (Canon vs Fuji vs Sony)

[kaylee]

is this a troll

[DeesserLord]

 

not my video, maybe not 1/50 but just look how it looks when she moves quickly.

who ´s the troll ? Anyway, I’m out.

[KnightsFan]

13 minutes ago, DeesserLord said:

not my video, maybe not 1/50 but just look how it looks when she moves quickly.

That's certainly not 1/50. The stutter in that video is from a very fast shutter speed. Is this representative of what you are experiencing? If so, then your shutter speed is too fast.

[frontfocus]

pretty easy to spot, there is no motion blur with her hair, so that's just a fast shutter speed. That's a user error if you wanna call it an error, maybe the user wanted that kind of effect, who knows. 

I think it's pretty safe to say now, that there is no problem with the X-T30 or the A6300, but that some users just need to learn the basics

[thebrothersthre3]

I don't notice anything jittery in that video, just high shutter speed and (probably 60p as well?).  

Hard to notice things without a comparison. 

[DeesserLord]

Except it still  has the same unnatural high shutter speed look at 1/50, with rolling shutter making things worse, and you get banding under some led lights which never occurred with the canon.

Some users have actual EXPERIENCE with said cameras while some just endlessly read reviews and forums, just follow the hype and think they help. Yes it’s a great camera for you tubers, for a video look, but not for those aiming an analog film look.

Show me your « filmic » footage with analog like 24p motion shot on the xt30 and A6000 if you want to prove me wrong .

[frontfocus]

2 hours ago, DeesserLord said:

Show me your « filmic » footage with analog like 24p motion shot on the xt30 and A6000 if you want to prove me wrong .

Why should we proof you wrong? It's your turn to proof your point and show clips you shot yourself showcasing the problem. 

 

[Inazuma]

3 hours ago, DeesserLord said:

Except it still  has the same unnatural high shutter speed look at 1/50, with rolling shutter making things worse, and you get banding under some led lights which never occurred with the canon.

Some users have actual EXPERIENCE with said cameras while some just endlessly read reviews and forums, just follow the hype and think they help. Yes it’s a great camera for you tubers, for a video look, but not for those aiming an analog film look.

Show me your « filmic » footage with analog like 24p motion shot on the xt30 and A6000 if you want to prove me wrong .

I have been shooting with the XT30 for 2 months and haven't had these issues except when shooting at high shutter speeds. Please do upload some videos on youtube so we can see it for ourselves  Today I'm going to be shooting a side by side comparison between the XT30 and Canon M6 II to see if the motion differs.

[Inazuma]

Here's a video comparing the XT30's 4k at 1/50th and 1/1600th shutter speed. 

 

[Papiskokuji]

I had the Fuji X-T20 and I always found the motion rendition a bit weird, stuttering... Or also like the OP describes, like the shutter speed is higher than it actually is. The X-T20 also had the fuji waxy skin effect. Now with the X-T3, everything looks great ! Eventhough sometimes those skins flirt with waxyness (to emulate film I guess, deeper, thicker skin tones, not always desirable).

[DeesserLord]

On 11/10/2019 at 4:49 PM, Inazuma said:

Here's a video comparing the XT30's 4k at 1/50th and 1/1600th shutter speed. 

 

Thanks. Rolling shutter is worse than I thought at 1/50, might be why the motion looks strange. 

Skin tones can look weird indeed in Provia or Astia, that’s why I used F-Log

[no_connection]

Rolling shutter is the same regardless of shutter speed.

Also highlight rollof/contrast will change the apparent shutter speed or motion blur, take the roof for example, it looks like a sharp(ish) edge while the fence is more blurry, the red object have a bit of a gradient while the poles are moving blocks etc.

[KnightsFan]

34 minutes ago, DeesserLord said:

Thanks. Rolling shutter is worse than I thought at 1/50, might be why the motion looks strange. 

Skin tones can look weird indeed in Provia or Astia, that’s why I used F-Log

It's still possible that there is something else going on, but it is sounding like rolling shutter is the issue.

I am not aware of any measurements of the XT30's rolling shutter and haven't used it myself, but the Sony a6xxx cameras have terrible rolling shutter in 4k. If RS is indeed what you are seeing, then the XT3 is the best 4k APS-C option, aside from the modern FF cameras in their APS-C crop mode (Z6, S1, A73). Most 1080p modes have better RS, so it's probable that the t6i is faster and would explain why you liked that camera's motion better. Did you try comparing 1080p vs. 4k on your XT30 and seeing if that solves the motion issues?

If you are okay with HD video, then the Samsung NX1 can be found very cheap. It's still a top-class APS-C photography camera, though unfortunately a dead system.

The GH5 and especially the GH5s have good RS, but you'd need a speed booster for APS-C framing. They have nice chunky grips and are comfortable to hold. There really aren't many good APS-C hybrids out there, to be honest. XT3 is awesome, but Canon and Sony continue to have high RS, and the NX1 is simply outdated.

[ac6000cw]

There is a rolling-shutter comparison list here - http://www.dvxuser.com/V6/showthread.php?303559-Measuring-rolling-shutter-put-a-number-on-this-issue!&p=1986288177&viewfull=1#post1986288177

[DeesserLord]

10 hours ago, KnightsFan said:

It's still possible that there is something else going on, but it is sounding like rolling shutter is the issue.

I am not aware of any measurements of the XT30's rolling shutter and haven't used it myself, but the Sony a6xxx cameras have terrible rolling shutter in 4k. If RS is indeed what you are seeing, then the XT3 is the best 4k APS-C option, aside from the modern FF cameras in their APS-C crop mode (Z6, S1, A73). Most 1080p modes have better RS, so it's probable that the t6i is faster and would explain why you liked that camera's motion better. Did you try comparing 1080p vs. 4k on your XT30 and seeing if that solves the motion issues?

If you are okay with HD video, then the Samsung NX1 can be found very cheap. It's still a top-class APS-C photography camera, though unfortunately a dead system.

The GH5 and especially the GH5s have good RS, but you'd need a speed booster for APS-C framing. They have nice chunky grips and are comfortable to hold. There really aren't many good APS-C hybrids out there, to be honest. XT3 is awesome, but Canon and Sony continue to have high RS, and the NX1 is simply outdated.

Thanks. I didn't try the X-T30 at 1080p (i was really hoping to switch to 4K). It looks good on dpreview https://***URL removed***/reviews/fujifilm-x-t30-review/8 . But i had other issues with the camera : i thought the auto-iso realtime transitions were too harsh / stepped, the kit lens was noisy (the canon STM are completely silent), the surprising lack of custom user "picture profile", which forced me to use FLOG + a custom LUT, the not so great user interface, still  much better than sony (awful) but i prefer the canon one, and most importantly i prefer Canon's color science...  

So i think i'm going to stick to 1080p and try a Canon M50 with a (discontinued) 18-55 STM lens (not a fan of 15-45). Its 1080p seems sharper than the T6i. We'll see.

This whole thing made me realize we all trash Canon for their lack of real upgrade in the video department for too long, but their stuff still holds up when you really look in the details. 

Hopefully things will evolve in a few year and 4K will mature a bit. Right now unless you spend a lot it seems like it's always gonna be a compromise.

[Inazuma]

@DeesserLord What custom picture profile do you use on Canons? I'm finding with my M6 that the environment looks better/closer to real life but skin can tend to look too red or have harsh contrast compared to my XT30. 

Also in the video I posted above there is no stutter issue and personally I find the rolling shutter to be only noticeable when doing very fast pans.

[DeesserLord]

I use VisionColor (not free) with contrast -2, sharpness to minimum, saturation at zero. 

my issue with the motion on the xt30 might not be rolling shutter (though it surely doesn’t help), I just can’t describe it properly, my English has its limits.

 

[fuzzynormal]

On 11/6/2019 at 9:55 AM, thebrothersthre3 said:

Have you tried lowering the shutter speed? You should be able to do 1/48 shutter. I know back in the GH2 days a lot of people liked shooting at 1/40 shutter to get a more film look. 

This.

It's kinda weird how there's all these wild suggestions to affect the cadence of the how a camera plays its video...but you can easily change it, and dial in a more personally favorable look, by simply adjusting the shutter speed.  

The "180" notion should be considered a suggestion, not a rule. 

[KnightsFan]

7 hours ago, DeesserLord said:

This whole thing made me realize we all trash Canon for their lack of real upgrade in the video department for too long, but their stuff still holds up when you really look in the details.

Whatever looks and works best for you, and yeah, for a lot of people that is Canon.

7 hours ago, DeesserLord said:

Hopefully things will evolve in a few year and 4K will mature a bit. Right now unless you spend a lot it seems like it's always gonna be a compromise.

Thats fair, but the XT3 is under $1200 used and is phenomenal in 4k.

1 hour ago, DeesserLord said:

my issue with the motion on the xt30 might not be rolling shutter (though it surely doesn’t help), I just can’t describe it properly, my English has its limits.

Thats why i would love to see a sample! So much easier than describing it. It could be as simple as auto shutter speed (especially if the video you posted was indicative of your issue) or it could even be a one-in-a-million bug with your camera. But if you are happy with the M50 then the issue is moot.

[thebrothersthre3]

I am pretty sure the XT3 and XT30 has the same rolling shutter. Same sensor and processor. I could be wrong of course.