Fuji X-T3 and X-T4 discussion

[DBounce]

Here is a quick shootout of the Fuji X-T3, Canon 1DXMk2, Canon C200 and Panasonic GH5S. Granted FCPX refused to not blowout some of the footage of camera B. It's displayed correctly in some of the head to head comparisons. I think it might have been caused by a software bug in FCPX. In any case all four cameras were filming in base ISO in controlled lighting. Opinions welcomed.

 

[webrunner5]

I thought A was the best, D second, C third B last.

[Yurolov]

28 minutes ago, DBounce said:

Here is a quick shootout of the Fuji X-T3, Canon 1DXMk2, Canon C200 and Panasonic GH5S. Granted FCPX refused to not blowout some of the footage of camera B. It's displayed correctly in some of the head to head comparisons. I think it might have been caused by a software bug in FCPX. In any case all four cameras were filming in base ISO in controlled lighting. Opinions welcomed.

 

C200 is the best, then the fuji, then the 1dx, then the gh5s (skin tones look quite bad). The 1dx is a different angle and different lighting so it is hard to tell how it compares. But i picked the gh5s instantly and knew A was the c200. 

[KnightsFan]

41 minutes ago, Yurolov said:

C200 is the best, then the fuji, then the 1dx, then the gh5s (skin tones look quite bad). The 1dx is a different angle and different lighting so it is hard to tell how it compares. But i picked the gh5s instantly and knew A was the c200. 

That is exactly what I thought, both in ranking order and that b had less flattering lighting. I didn't try to guess which was which, I just ranked based on the images. A was by far the best, c was good and b was close behind, d was clearly the worst for my taste.

[androidlad]

My take:

A - GH5S, because of the yellow skin tone.

B - 1DX II, because of the harsh highlight clipping.

C - C200, because of the smooth tonal gradation and wide DR.

D - X-T3, because the low contrast and slight colour tint scream ETERNA!!

[Yurolov]

1 hour ago, androidlad said:

My take:

A - GH5S, because of the yellow skin tone.

B - 1DX II, because of the harsh highlight clipping.

C - C200, because of the smooth tonal gradation and wide DR.

D - X-T3, because the low contrast and slight colour tint scream ETERNA!!

The gh5s should be obvious not only by skin tone but by background blur. U way off. 

[noone]

I think the background blur is not so obviously the GH5s JUST because it has greater DOF as we don't know the lenses used or aperture ETC.

Yes, D is the GH5s but it is using a seemingly wider lens and the DSLR seems to be using a longer lens as well so while it was more likely to be the smaller sensor camera, that doesn't necessarily follow.

I actually prefer the GH5s to the Fuji there as the Fuji Blue colours where quite different to the other three.

I guess if I looked at it on a few different screens I could get different preferences too.

Bottom line they all looked good to me.

[ajay]

I had the 1DXM2 and GH5S correct but had the X-T3 and C200 switched. The 1DXM2 (which I once owned) always crushed the blacks and shows in this sample video. I was happy to know that I had the X-T3 and C200 switched considering that I am so close to buying the X-T3!

Thanks!

 

[Snowbro]

I was able to tell by the autofocus alone lol. The xt3 snapped focus un-naturally & the gh5 was on a hunting expedition. Is it possible to get a smooth rack focus with the xt3s af? 

[Mattias Burling]

I liked A and C.

[Yurolov]

5 hours ago, noone said:

I think the background blur is not so obviously the GH5s JUST because it has greater DOF as we don't know the lenses used or aperture ETC.

Yes, D is the GH5s but it is using a seemingly wider lens and the DSLR seems to be using a longer lens as well so while it was more likely to be the smaller sensor camera, that doesn't necessarily follow.

I actually prefer the GH5s to the Fuji there as the Fuji Blue colours where quite different to the other three.

I guess if I looked at it on a few different screens I could get different preferences too.

Bottom line they all looked good to me.

The logical way of thinking about it is what would he have had to have done to the s35 and FF images to make them have more DOF than a MFT sensor (and why would he have done it to one and not the other two s35 & FF cameras). Unless he was willfully trying to trick us the chances are minute. So ye, you can be pretty confident about it.

His skin looked pretty sickly with the panasonic footage. Good to see fuji mathcing it with the c200 though. Both good images. 

[Emanuel]

@DBounce Thanks for your comparative : ) However, it is apples to oranges, isn't it? DOF doesn't match. You're using different sensor size camera units. You can't compare a chocolat pie with a lemon pie if you don't add the same quantity of sugar : )) Applied focal length is also crucial : -)

[DBounce]

2 hours ago, Emanuel said:

@DBounce Thanks for your comparative : ) However, it is apples to oranges, isn't it? DOF doesn't match. You're using different sensor size camera units. You can't compare a chocolat pie with a lemon pie if you don't add the same quantity of sugar : )) Applied focal length is also crucial : -)

Here is the lens breakdown:

Canon C200 - 24-70mm F2.8 ii @ 35mm F4

Canon 1DXMk2 - 16-35 F4 @ 35mm F4

Fjifilm X-T3 - 23mm F2 @ F2.6 (35mm FF equivalent)

Panasonic GH5S - 12-35mm F2.8 @ 18mm F2.8 (36mm FF equivalent)

About: I tried to get everything as close to the M43 configuration as possible as I figured it would be at a disadvantage given the smaller sensor. DOF looks different on the GH5S, because... DOF is different on the M43. I was running a $1000 lens on the GH5S that was rated at F2.8. To get shallower I would have needed to use the Leica 42.5mm, but did not have enough equivalent lenses to this focal length to accommodate all the other cameras. So 35mm was choosen.

Setup: In some cases physical proximity to the cameras was different. Because there's was only so much space to set everything up. 

Honestly, I was surprised when comparing these cameras side by side how much less shallow the M43 image appeared. With the S35 sensor the Fuji looked pretty similar to the Canons.

Colors: Color wise the C200 was easiest to work with. It requires next to nothing to look good. Shoot at base ISO for the log profile you intend to add and you are done. It looks great all day long.

The Fuji was shoot in Eterna... reflecting back perhaps Flog would have been easier to tweak? 

The 1DXMk2 was shoot in neutral picture profile.

The GH5S was shot in VLogL. The standard Panasonic LUT was then applied. 

FCPX: The project was set to wide color gamut. I had read that this is desirable when editing Raw footage. But doing this seemed to cause issues with the 1DXMK2 image. It looked fine in the editor,  but rendered with blown highlights. I believe FCPX did not know how to handle its 8 bit image with the WCG project setting. The other cameras were all shot at 10 bit or higher. I tried several approaches to try to regain the highlight detail of the 1DXMk2, but nothing seemed to work. It continually looked great in the editor,  only to render blown out and harsh. I believe davinci is the better editor when dealing with a mix of 8 bit and higher. Unfortunately, my free version of resolve does not output 4k DCI... so that was out of the picture. 

Fruity chocolate: I don't think it's Apple's to (insert here) as they say. I used the lenses I own. This is a real world test. Question: Does Panasonic make a M43 equivalent to the Canon 24-70mm? Well,  Panasonic would tell you, "yes... it's the 12-35mm". So that is what I used.  And frankly I'm not alone in this thinking. While anything can be jerryrigged together, that leaves you compensating for something the cameras lacks.

This of course applies to all M43 cameras,  not just the GH5S. You are at a disadvantage regarding DOF. It's a limitation you get used to... or choose to find workarounds for. While I have a MSB, I seldom use it.  My Canon glass runs better natively... so why use it on an M43 body? Also doing so negates any size advantage.

 

4 hours ago, Yurolov said:

The logical way of thinking about it is what would he have had to have done to the s35 and FF images to make them have more DOF than a MFT sensor (and why would he have done it to one and not the other two s35 & FF cameras). Unless he was willfully trying to trick us the chances are minute. So ye, you can be pretty confident about it.

His skin looked pretty sickly with the panasonic footage. Good to see fuji mathcing it with the c200 though. Both good images. 

I wanted to get the lenses close to each other. So everything was set to match the smallest sensor. The full frame cameras running at F4 while the X-T3 was at F2.6 IIRC.

[Emanuel]

Despite I appreciate your answer, you should use a faster lens for a fair comparison with MFT as for instance or will become misleading as happened so. Take a look on that FF sample there. Not a tragedy but you could remake it as you intend an accurate outcome. Sorry mate for this time (E : -)

[mercer]

You could always do your own comparison and set the parameters yourself. 

3 minutes ago, Emanuel said:

Despite I appreciate your answer, you should use a faster lens for a fair comparison with MFT or will become misleading as happened so.

DBounce did a test based on what he had and what he felt was fair. I think the results were interesting, although not surprising.

[DBounce]

2 minutes ago, Emanuel said:

Despite I appreciate your answer, you should use a faster lens for a fair comparison with MFT or will become misleading as happened so.

I slowed down the other lenses to f4... none were wide open, save for the M43 camera. 

Consider the following: 

GH5S = $2500 - body only + $1000 lens.

Fuji X-T3 = $1500 body only + $350 lens.

Tell me, how much should I cripple the much less costly Fuji so that the Panasonic can be competitive? At some point it becomes unfair to handicap the other cameras. By right handicapping should not even happen, as I know of no one that purposely slows down their lenses out of respect for those with m43 sensors.

 

[mercer]

And it’s barely a one stop difference, so whatever. I think the takeaway here is how good the X-T3 is and with all of this FF talk, S35 isn’t dead yet.

[KnightsFan]

11 minutes ago, Emanuel said:

Despite I appreciate your answer, you should use a faster lens for a fair comparison with MFT as for instance or will become misleading as happened so. Take a look on that FF sample there. Not a tragedy but you could remake it as you intend an accurate outcome. Sorry mate for this time (E : -)

but this isn't a sensor size comparison test, he's just doing a quick real world comparison between several cameras. I didn't even look at dof when judging, I only looked at skin tones, and it was clear to me which looked better. It was very good in that regard.

If anything, it also highlights the fact that you need prohibitively expensive glass to get equivalent dof on m43.

[Emanuel]

Glenn, give your tolerance a break : ) There's nothing wrong on constructive criticism. I only find that test is not the fairness itself ; ) It surely is a private test. I just think my observation is rather pertinent. That test sould also have its own thread BTW, not in this one here.

[Mattias Burling]

I don't understand why camera tests should be "fair". Its strange to me. Its like the popular "everything else equal". It never is