In the mirrorless market, Fuji and Sony are two of the key players, with Sony lifting the entire market as a whole, and photographers switch to either system in droves for their own perspective reasons. The Sony A7 III and Fujifilm X-H1 are two of the latest cameras from the photographic giants.
Both cameras are fantastic in their own right – until you start to compare them to each other. Priced similarly the feature set includes magnesium alloy construction and weatherproofing, and IBIS. The X-H1, however, is heavier and slightly larger than the Sony. The Sony, which houses a larger sensor in its magnesium skeleton, has superior ISO performance when it comes to video – or that is what Max Yuryev discovered in his latest comparison.
The A7 III starts with a native sensitivity between ISO 100 and 51200. The sensitivity can be expanded to 204800 for stills and 102400 for video. The X-H1 has a smaller range between ISO 200 and 12800 ISO, with an extended range as high as 51200 for stills and 25600 for video. What Yuryev concluded (and you can clearly see) is that not only did the Fuji under-perform at similar ISO values, which is typical of APS-C vs Full-frame, but the X-H1 was at least 2/3 stop underexposed. It’s almost as if their ISO rating is actually false.
The X-H1 is really being touted as a video shooter, and the spec sheet says as much, but one has to wonder what this evaluation from Max means for Fuji’s primary video shooter, and what can be done to rectify the problems.
[REWIND: What’s The Video Platform Of The Future? Facebook, YouTube, or You?]
9 Comments
Leave a Reply
You must be logged in to post a comment.
I’m not really concerned with comparing iso #s. What matters is if i can get a good end result and proper exposure. I was just shooting some low light 4k video last night with the XH1 (and a 24mm f2.8 lens) and noticed anything beyond 3200 iso became so grainy it’s unusable. Also noticed I have 2 hot (white) pixels showing up in all of my 4K footage which is not good :/ I never noticed this with daylight shots. Now that it’s out of warranty I’m screwed until the XH2 is released. Note to self Check your sensors early!
Does it have anything to do with the fact that Fujifilm uses a different ISO standard? I believe it is called the SOS standard, a supposedly more objective one.
yes it does. The standard has several standards.
The ISO 12232:2006 standard has a variety of techniques for calibrating digital ISO. However, by Japanese law, only the most recent two, SOS (Standard Output Sensitivity) or REI (Recommended Exposure Index) may still be used. SOS is less subjective, but it only applied to a specific automatic calibration, single-zone automatic metering system in sRGB space. REI must be used for raw or alternate color spaces.
Hi. I do not get the big fuzz about this. Image quality, noise, depend almost uniquely on the quantity of light a system can absorbe. Front of the same aperture lens their are many way to increase quantity of light : exposure time, flash, image stacking, more efficient detector (BSI), bigger detector …. The iso is just a tool to brighten the picture and it does it slightly better than the exposure slider on you photo software. The fuji’s one is just underrated in term of grey level it produce compare to other brands (for the same flux of light hitting the detector). But this does not change the image quality which depend on the quantity of light absorbed.
My guess : the idea behind Fuji mind is to surf on the quick camera review fashion. For people who make quick uncontrolled test, the DR appear higher for a given iso compare to other brand (because the gain is lower). Also for people that does not know where the noise come from, they take a 1600 iso with fuji, normally exposed to the desired grey level and a 1600 iso with an e.g. Nikon APS-C exposed to the same grey level and say “the fuji 1600 iso looks better” however it does look better because more light has been absorbed to give the same grey level at same iso.
Moreover at and after iso 800 the XH-1 is iso invariant (constant detector read noise), brightening the picture in post or with higher iso will not make noticeable difference. It matters however a bit more for jpeg shooter than raw shooters.
In the digital area less and less people rely to iso because they have a direct feedback of the exposure (histogram or live view, or reviewing photo). We set the exposure and aperture to let the maximum amount of light in, taking care of freezing motion, put everything needed in focus and avoid saturation in necessary bright details. This is what will settle the image quality. The iso comes after to put the histogram to the right, whatever the value it does not matter.
Cheers.
Does it have anything to do with the fact that Fujifilm uses a different ISO standard? I believe it is called the SOS standard, a supposedly more objective one.
yes it does. The standard has several standards.
The ISO 12232:2006 standard has a variety of techniques for calibrating digital ISO. However, by Japanese law, only the most recent two, SOS (Standard Output Sensitivity) or REI (Recommended Exposure Index) may still be used. SOS is less subjective, but it only applied to a specific automatic calibration, single-zone automatic metering system in sRGB space. REI must be used for raw or alternate color spaces.
Hi. I do not get the big fuzz about this. Image quality, noise, depend almost uniquely on the quantity of light a system can absorbe. Front of the same aperture lens their are many way to increase quantity of light : exposure time, flash, image stacking, more efficient detector (BSI), bigger detector …. The iso is just a tool to brighten the picture and it does it slightly better than the exposure slider on you photo software. The fuji’s one is just underrated in term of grey level it produce compare to other brands (for the same flux of light hitting the detector). But this does not change the image quality which depend on the quantity of light absorbed.
My guess : the idea behind Fuji mind is to surf on the quick camera review fashion. For people who make quick uncontrolled test, the DR appear higher for a given iso compare to other brand (because the gain is lower). Also for people that does not know where the noise come from, they take a 1600 iso with fuji, normally exposed to the desired grey level and a 1600 iso with an e.g. Nikon APS-C exposed to the same grey level and say “the fuji 1600 iso looks better” however it does look better because more light has been absorbed to give the same grey level at same iso.
Moreover at and after iso 800 the XH-1 is iso invariant (constant detector read noise), brightening the picture in post or with higher iso will not make noticeable difference. It matters however a bit more for jpeg shooter than raw shooters.
In the digital area less and less people rely to iso because they have a direct feedback of the exposure (histogram or live view, or reviewing photo). We set the exposure and aperture to let the maximum amount of light in, taking care of freezing motion, put everything needed in focus and avoid saturation in necessary bright details. This is what will settle the image quality. The iso comes after to put the histogram to the right, whatever the value it does not matter.
Cheers.