Fujifilm X-T3 high ISO image quality and noise reduction

X-T3 + 56 1.2 test shot from this evening. 1/640 f/1.4 ISO 1600 slightly underexposed. jpg straight out of camera. Looks good to me.

1. Full

25600iso SOOC X-T3

Two Hot Shoes wrote in post #18718913
25600iso SOOC X-T3

Hosted photo: posted by Two Hot Shoes in
./showthread.php?p=187​18913&i=i54749864
forum: Fuji Digital Cameras

Zoomed in on phone, screen shot

Hosted photo: posted by Two Hot Shoes in
./showthread.php?p=187​18913&i=i24933749
forum: Fuji Digital Cameras

First of all Congrats on the Xt3 Kim!!

OK to be fair.....wouldn't this be more accurate??

Fuji's 25600 iso (SOS standard) x .71 = Canon/Nikon/Sony 18,176 ISO REI which is commonly used for comparison tests we see online.

So for comparison sakes if you set Nikon/Sony/canons ISO 25600 (sos standard). The Fuji should be set to 36056 ISO (SOS standard). This would be apples to apples comparison in iso standards in photo comparison tests.

Even slight differences in ## by increasing by a small number still is more of a true comparison scientific test. So this is where I find comparison from Nikon D500 vs Fuji X-t2 and other comparisons online is probably a bit inaccurate.

What would be interesting is to shoot wide open in that setting but in extremely low light. That would show the difference in how stressed the sensor would be. This would indicate a more realistic pushed light sensitivity in extremely low light.

Many tests I've seen I do not feel so bad owning an X-t2 s far as having same or even better low light high iso performance than the new X-t3.

Don't get me wrong I'd love to have the quad processor improvements in the XT3!!!!!

All in all I don't think many can complain of the iso performance of the aps-c sensor.

I think perhaps your numbers are not accurately representing how camera manufactures choose the ISO numbers for any given exposure. The SOS and the REI are guidelines that suggest a range of factors when determining the 'ISO' value of any setting. It's left up to the manufacturer to decide on that value. They are not a set in stone value but the SOS is much more stringent. So I really don't know where you got that multiplier from but if it makes you happy, well...

Cheers, it's a big step up in AF for sure I've yet to use it on a shoot just late last night messing about a little. There are a load of cool features, the quality of the video output is really up there.

Two Hot Shoes wrote in post #18719058
I think perhaps your numbers are not accurately representing how camera manufactures choose the ISO numbers for any given exposure. The SOS and the REI are guidelines that suggest a range of factors when determining the 'ISO' value of any setting. It's left up to the manufacturer to decide on that value. They are not a set in stone value but the SOS is much more stringent. So I really don't know where you got that multiplier from but if it makes you happy, well...

Cheers, it's a big step up in AF for sure I've yet to use it on a shoot just late last night messing about a little. There are a load of cool features, the quality of the video output is really up there.

Some non Fuji user will read this thread maybe looking for some visual comparison regarding high iso images.

https://www.youtube.co​m …ontinue=980&v=i​BAMOvlZ8qg

If you start looking at 13:00 into this older video regarding Fuji X-t1 this is just a visual difference from consumer grade Nikon aps-c body and Fuji X-t1. It still appears Fuji is still using the SOS (ISO) standard. Since Nikon/Canon/Sony uses REI this is never a real good indication of apples to apples comparison.

No argument here......just pointing out how we can take visual comparisons as facts when in a full story must be disclosed. In the higher range of high ISO settings every little bit counts if you start seeing 1/3 to 1/2 stop differences between images.

So far even with my personal gear my Canon system can have a "good" exposure but taking the identical photo with same setting with Fuji the image is darker.

This means the 3 components of the "EXPOSURE Triangle" is not truly using identical number standards in the ISO "component" if you compare the REI ISO standard that Nikon/Canon/Sony vs Fuji's SOS standard.


IT IS HOW IT IS....... The X-T3 still is one of the biggest and bold improvements Fuji has ever pushed out of it's R&D. So pleased and can see how people can go crazy about this camera specs and real life use!

Yea... so the title of this thread is ‘Fujifilm X-T3 high ISO image quality and noise reduction’

If someone posts an image at quite high ISO taken on a Fuji X-T3 than BOOM right on point.

Yes Fuji are still using the more recent and more stringent set of guidelines to choose the ISO markings. The SOS ones.

I’ve always said that 12800 ISO on any of the 24mp cameras was really nice and I’ve posted up loads of images taken with that ISO in good and bad lighting. To be it looks like the 26mp camera is about 2/3 better or at least a little cleaner at 25600ISO. Anyway these days 1/3 of a stop won’t make or break a good photo so who cares if another brand has chosen to pick a different value for the ISO reading not going to stop anyone making good images given how good modern cameras are. That vid is from 2015 very old by today’s standards. He still shoots Fuji BTW

ISO51200 F/2.8 1/200 shutter, out of camera JPEGs unedited Eterna simulation X-T3

IMAGE LINK: https://flic.kr/p/2avJ​gQs

IMAGE LINK: https://flic.kr/p/2avJ​gKs

AlanU wrote in post #18719212
Some non Fuji user will read this thread maybe looking for some visual comparison regarding high iso images.

https://www.youtube.co​m …ontinue=980&v=i​BAMOvlZ8qg

If you start looking at 13:00 into this older video regarding Fuji X-t1 this is just a visual difference from consumer grade Nikon aps-c body and Fuji X-t1. It still appears Fuji is still using the SOS (ISO) standard. Since Nikon/Canon/Sony uses REI this is never a real good indication of apples to apples comparison. ...

Because there is really no magical adjustment factor to use that would make the Fujifilm and Canon ISOs comparable in any meaningful way, the simplest comparison would be to set up each camera so that they are shooting JPEG, with a neutral picture style or film sim, custom white balanced in the scene and setting the aperture and shutter to identical settings. Then the ISO of each camera would be adjusted to get a similar JPEG rendering in terms of overall image brightness. For example, set up the first camera with fixed aperture and custom WB. Set the camera on ISO 200 and adjust shutter speed until you get a gray target exposed "properly." Then set up the second camera with the same aperture, custom WB and the same shutter speed as the "properly" exposed image from the first camera and then adjust ISO on the second camera to get a JPEG image that is as close as possible to the "properly" exposed image from the first camera. Now you have a direct comparison between ISOs.

You could do this across the range of ISO and you could do it in both directions - that is, use the Fujifilm camera as the "first" camera and sweep through its ISO range, replicating each Fujifilm ISO with the corresponding Canon ISO; then switch and use the Canon as the reference camera and replicate each ISO image form the Canon with the corresponding FUjifilm ISO setting that gives identical results to the Canon.

Although this would be a more empirical approach to comparing ISO, it would be a direct comparison and you could plot the relationship between Fujifilm ISO and Canon ISO directly instead of assuming some scale factor that, even if it is meaningful, may not be constant across the ISO range.

I no longer have my Canon equipment, otherwise I would do the comparison and publish it here. I can try it with my Ricoh GRII and my X-H1 for a proof of concept though.

kirk