Fujifilm X-T3 vs Canon 5D Mark IV, bokeh @ 50mm, f/1.4 (equiv.)

This is really nothing scientific nor too controlled, so please take it for what it is, more of a "for your amusement" kind of post. I did a quick comparison for myself, but thought someone else might find it interesting, too, thus sharing it here. All images are full sized, SOOC (straight out-of-camera) JPEGs. White balance set to auto ("Auto: White Priority" on Canon).

With that out of the way, let`s say I was just curious to compare Canon 5D Mark IV + Sigma 50mm f/1.4 EX DG HSM (so older one, not "Art") against Fujifilm X-T3 + Mitakon Zhongyi Speedmaster 35mm f/0.95 Mark II in terms of bokeh, but also light transmission, both wide open.

I tried to match framing and focus, but yeah, not having too much time on my hands, it isn`t really ideal... oh, well.

Anyway, for the first pair, both cameras were in "A" (aperture priority) mode, Canon auto-focused at its lens MFD (minimum focusing distance), Fuji framed to match, focused manually (eh, missed, seems front-focused a bit).

#1a, Canon 5DIV + Sigma 50mm, f/1.4, 1/40 sec, ISO 1000 - auto-focus

#1b, Fuji X-T3 + Mitakon 35mm II, f/0.95, 1/50 sec, ISO 1600 - manual focus (missed! front-focused)

From the two above, it`s clear that Fuji`s image is brighter, which is probably due to metering difference possibly even caused by a different metering mode and/or setup (didn`t really bother to check/match).

Bokeh seems pretty similar to me, might be even a bit blurrier on Fuji / Mitakon pair, but I guess that might also be caused by missed, front-focus, adding a tad more blur to the background...? Could it also be influenced by brighter exposure, making lights in the background appear bigger?

For the next pair, I switched to "M" (manual) mode, setting both cameras for same exposure as if lenses head ideal light transmission (which they don`t, I know), also paying a bit more attention to focus this time - both manual now.

Note that this time I didn`t really match the framing, not having the time to retake for the better one, here it goes as-is.

#2a, Canon 5DIV + Sigma 50mm, f/1.4, 1/50 sec, ISO 1250 - manual focus

#2b, Fuji X-T3 + Mitakon 35mm II, f/0.95, 1/50 sec, ISO 640 - manual focus

This time Fuji / Mitakon image is visibly darker, which should imply, in terms of light transmission, that Mitakon is further away from T/0.95 than Sigma is from T/1.4 - but that is something to be expected, I guess...?

So, while bokeh might be more or less comparable, ISO speed gain with APS-C Fuji / Mitakon won`t exactly be one stop, but less, giving some small theoretical edge in image quality back over to full-frame Canon / Sigma pair (all other things being the same... which is not the case, of course , but just thinking out loud).

Background this time seems blurrier on Canon / Sigma pair, though, but that could now be caused by more favorable framing of that image, being a tad closer to the subject (also a bit different perspective)...? Again, could darker exposure on Fuji / Mitakon image make the lights in the background appear smaller?

Well, that`s it for now I might do more comparisons in the future (and hopefully some more controlled ones), but hope someone find these quick ones informative anyway.

Please feel free to discuss, comment, and make your own conclusions.

Interesting, thanks a lot. It’s difficult to compare the shots regarding brightness, as the ISO values are always different. In order to judge properly between these setups, I think you should fix that as well.

hchris wrote:

Interesting, thanks a lot.

You're welcome, thank you for the interest

It’s difficult to compare the shots regarding brightness, as the ISO values are always different. In order to judge properly between these setups, I think you should fix that as well.

Unless I'm missing something, I'll have to disagree - that second image pair is exactly about comparing brightness (even though I said "exposure", which arguably may not be the most correct term, but let's leave that for another topic).

Ideally, Mitakon, being ~1 stop faster (brighter, f/0.95 vs f/1.4), should deliver same image brightness for ~1 stop slower ISO speed (or ~1 stop faster shutter speed), and that is what that second test is about.

Of course, as lens light transmission is not ideal (T-stop not the same as f-stop), Fuji / Mitakon pair image ended up darker, which could (should?[1]) mean Mitakon is less than ~1 stop faster than Sigma in regards to light transmission - in other words (and more importantly), it is not allowing full stop ISO (or full stop shutter speed) advantage in comparison to Canon / Sigma.

But it may be interesting to see brightness difference for the same shutter speed / ISO with both wide open, where Mitakon is expected to be brighter, I agree. Also, maybe a comparison with different ISO and brightness to match, loosely measuring light transmission advantage.

Thanks for your thoughts.

[1] I'm questioning this as camera manufacturers might opt to brighten certain lenses in-camera, as I've read somewhere for Fuji doing with (some of) its own lenses, so lens light transmission couldn't really be compared purely by end image brightness (for otherwise same setup).

pluton wrote:

Personally, I wouldn't assume that ISO 'X' on the Canon equates to ISO 'X' on the Fuji, or between any two cameras, unless they are the exact same model.

Pretty sure they differ at least .6 of a stop (difference in ISO between Fuji and others has been discussed to death on this forum)

keesenzo wrote:

pluton wrote:

Personally, I wouldn't assume that ISO 'X' on the Canon equates to ISO 'X' on the Fuji, or between any two cameras, unless they are the exact same model.

Pretty sure they differ at least .6 of a stop (difference in ISO between Fuji and others has been discussed to death on this forum)

Hmm, isn't this known "Fuji ISO variation/difference" more about RAW, being originally darker, while JPEG is still "corrected" by in-camera brightening on internal processing?

From what I understand, this "brightness correction" factor amount is recorded inside RAW and should be applied on JPEG developing, something not all RAW processors out there are complying to.

These are out-of-camera JPEGs, but different camera having different ISO interpretation / implementation is still a valid point in general, though, and something I didn't reason about enough, thanks both for bringing it up.

Here`s a bit more controlled comparison this time, though still not a perfect one, of course For Canon, framing is a bit off (which could have influenced Sigma bokeh balls shape a bit, making them more round in comparison to Mitakon, as being farther from the edge of the frame)... and I forgot to set base ISO.

Otherwise, custom white balance used this time on both cameras, and also attempted to match brightness. Shot off the tripod, 2 sec delay timer. Again, straight out of camera, full sized JPEGs.

#3a, Canon 5DIV + Sigma 50mm, f/1.4, 1/50 sec, ISO 400 - manual focus

#3b, Fuji X-T3 + Mitakon 35mm II, f/0.95, 1/20 sec, ISO 160 - manual focus

And a side by side comparison of two camera/lens pairs producing above mentioned images:

#4, Fuji X-T3 + Mitakon 35mm II (left, ~ 1kg), Canon 5DIV + Sigma 50mm (right, ~ 1.4kg)

Both cameras have a tripod mount attached at the bottom (thus raised a bit above the floor, casting a shadow below).

I took lens hood off Sigma to make comparison a bit more even as Mitakon doesn`t have one, though I usually do use Sigma with the hood, and Mitakon without...

Also note Canon having a silicon cover case, adding 1-2mm to the perceived overall bulk, but I just couldn`t be bothered to take it off for a single picture only, putting it back on again afterwards

Very fun comparison! Just enough to see that there are differences, but not so dry that I stop caring.

The shape of the bokeh balls from the Mitakon is interesting.

Did you find it difficult to focus the Fuji wide open when the subject is that close?

j0llysnowman wrote:

Very fun comparison! Just enough to see that there are differences, but not so dry that I stop caring.

The shape of the bokeh balls from the Mitakon is interesting.

Did you find it difficult to focus the Fuji wide open when the subject is that close?

I don`t think Mitakon bokeh shape is really that much different than Sigma, just that framing doesn`t really do it any favors here. Looking at the very first post of the thread, small green and orange LED lights bokeh balls seem similarly "cat`s eye" oval shaped for both lenses, being in a more similar position in the scene.

And focusing seems pretty easy with Mitakon (and it`s my first manual lens, even), where Fuji`s both focus peaking and dual display mode help a lot. Though obtaining focus at such shallow depth of field is one thing, while actually taking the shot in focus could be a different matter, especially if target is not really static. I`ve written about it in more details here.

Thanks for looking!

DominikT wrote:

These are good examples. But unless Fuji Bring out a 35mm f0.95 Auto focus lens, we won’t have an exact equivalent, to be able to say “ FF doesn’t matte”. Because we see in your test, it does matter, when you want shallow depth of field and auto focus. You just won’t get it with a crop censor.

That's a fair point, and if you need auto-focus, there is no comparison, I agree.

But the main reason I moved to Fuji is to downsize, so accepting manual focus experience was something I was more then willing to do for the sake of it, too, keeping the output/result (image) somewhat the same being a more important goal.

Adding auto-focus into the whole design would unavoidably make the lens bigger and heavier, in which case one could/should just stick with full-frame and 50mm f/1.4 again...

For an example, just look what happened to Fuji's own 33mm f/1 attempt, later scrapped for 50mm f/1 instead to keep the bulk reasonable, and not requiring a lens foot to handle, lol (where neither 50mm f/1 is a small lens by any means).

Even though I had no previous manual focusing experience, I've adapted more than fine (finding it superior for some applications, even), having me wish Fuji would release a set of fast but small manual-focus-only primes, with electronic contacts (for EXIF, for example) - in fact exactly what Cosina Voigtlander has been doing as of lately.

DominikT wrote:

WeirdSheep wrote:

DominikT wrote:

WeirdSheep wrote:

DominikT wrote:

These are good examples. But unless Fuji Bring out a 35mm f0.95 Auto focus lens, we won’t have an exact equivalent, to be able to say “ FF doesn’t matte”. Because we see in your test, it does matter, when you want shallow depth of field and auto focus. You just won’t get it with a crop censor.

Well that's not exactly true, you forgot the clarifying comment of 'at 35mm with autofocus'. You can get shallow dof very easily on APS-C, with AF, with plenty of lenses, so it's not exactly an issue for most people, especially as manual focus isn't exactly difficult these days.

I’m not sure you get my point. Yes you could use for example the Fuji 56 f1.2 to get the same depth of field as a 50mm 1.4 on full frame. But you’ll need to move way back. You can’t stand the same distance as with the 50mm on FF. I like to shoot with a 35 50 and 85mm on full frame. So I’m always looking at the equivalent on the Fuji crop sensor. And unfortunately there just aren’t any. Or they’re so big, there is no point in having the APSC system then, as you get the same thing with a smaller FF lens.

No l do get your point, firstly there are manual focus lenses that do what you want, but you want AF, but then the lenses are too big for you, so just use FF and be done with it, problem solved. I could pick any format and come up with a similar scenario, they're all compromises one way or another, that's life l'm afraid.

I’m not trying to start anything here. I love my X-Pro2 and have sold off my canon gear because the Fuji system is more fun to use, with its manual dials and rings. It brings me closer to the film experience.

However, what gets me is when people say that FF has no advantage over APSC. To me it does. Because you can get a 50mm 1.2 AF lens now for all the FF systems, but you can’t get a 35 f0.95 lens with AF for Fuji. Nor a 23mm one. And zoom lenses are even worse, nothing below f2.8 for Fuji. Which is an f4 on FF.

Yes there are MF third party lenses, the Voigtlander lenses look amazing, and I don’t mind using manual focus, (I have two vintage lenses I use very often on my Fuji) however, a lot of these lenses are just very soft wide open, almost to the point that you need to stop down to at least f2.8 to get a decent image.

Here are my requirements

-Love fujifilm dials and the design language

-Love fujifilm 1.4 primes, the aperture ring

-Love fujifilm film simulation (even though i shoot raw)

-I want thin DOF

-I want thin DOF zoom lenses (FF 2.8) that aren’t too heavy (e.g Tamron 20-40 2.8)

-I am also on a budget and need 3rd party options (Samyang/tamron/sigma) etc

-I want “reliable” AF-C when shooting at thin DOF

-I use NR for all my raw files (DxO PL + LR Cc)

My Solution:

-Why not have both (Fujifilm + FF)? if there are two different worlds that have their own set of strengths keep then separate.

DominikT wrote:

Raymond L wrote:

DominikT wrote:

WeirdSheep wrote:

DominikT wrote:

WeirdSheep wrote:

DominikT wrote:

These are good examples. But unless Fuji Bring out a 35mm f0.95 Auto focus lens, we won’t have an exact equivalent, to be able to say “ FF doesn’t matte”. Because we see in your test, it does matter, when you want shallow depth of field and auto focus. You just won’t get it with a crop censor.

Well that's not exactly true, you forgot the clarifying comment of 'at 35mm with autofocus'. You can get shallow dof very easily on APS-C, with AF, with plenty of lenses, so it's not exactly an issue for most people, especially as manual focus isn't exactly difficult these days.

I’m not sure you get my point. Yes you could use for example the Fuji 56 f1.2 to get the same depth of field as a 50mm 1.4 on full frame. But you’ll need to move way back. You can’t stand the same distance as with the 50mm on FF. I like to shoot with a 35 50 and 85mm on full frame. So I’m always looking at the equivalent on the Fuji crop sensor. And unfortunately there just aren’t any. Or they’re so big, there is no point in having the APSC system then, as you get the same thing with a smaller FF lens.

No l do get your point, firstly there are manual focus lenses that do what you want, but you want AF, but then the lenses are too big for you, so just use FF and be done with it, problem solved. I could pick any format and come up with a similar scenario, they're all compromises one way or another, that's life l'm afraid.

I’m not trying to start anything here. I love my X-Pro2 and have sold off my canon gear because the Fuji system is more fun to use, with its manual dials and rings. It brings me closer to the film experience.

However, what gets me is when people say that FF has no advantage over APSC. To me it does. Because you can get a 50mm 1.2 AF lens now for all the FF systems, but you can’t get a 35 f0.95 lens with AF for Fuji. Nor a 23mm one. And zoom lenses are even worse, nothing below f2.8 for Fuji. Which is an f4 on FF.

Yes there are MF third party lenses, the Voigtlander lenses look amazing, and I don’t mind using manual focus, (I have two vintage lenses I use very often on my Fuji) however, a lot of these lenses are just very soft wide open, almost to the point that you need to stop down to at least f2.8 to get a decent image.

Here are my requirements

-Love fujifilm dials and the design language

-Love fujifilm 1.4 primes, the aperture ring

-Love fujifilm film simulation (even though i shoot raw)

-I want thin DOF

-I want thin DOF zoom lenses (FF 2.8) that aren’t too heavy (e.g Tamron 20-40 2.8)

-I am also on a budget and need 3rd party options (Samyang/tamron/sigma) etc

-I want “reliable” AF-C when shooting at thin DOF

-I use NR for all my raw files (DxO PL + LR Cc)

My Solution:

-Why not have both (Fujifilm + FF)? if there are two different worlds that have their own set of strengths keep then separate.

Yes my requirements are very similar to yours. I’ve been using a Canon FF system since the original EOS 5D. All the way up to the EOS R, (and still use the R5C for work) but about 6 years ago I picked up a used X-E2, and more recently an X-Pro2. And the Canon cameras and L series lenses ended up sitting in my cupboard, and more recently packed away in storage. I just stopped using them. I was hoping that Fuji would eventually release a FF rangefinder style camera. But that’s probably won’t happen. So my options are to get a Sony A7 series with some nice primes with aperture rings and to pretend it’s like using a Fuji. Or get the GFX 50R and hope for some fast primes from Fuji. Or use my X-Pro2 and see what happens in the long term.

You were using Canon for many years, then Fuji. Both systems are well known for their excellent color science, especially skin tones. You rally wonna sacrifice it and spent like 3x times more on editing, when trying to fix crappy Sony colors?