GFX100S vs SL2/X1D

[SrMi]

Advertisement (gone after registration)

On 3/22/2021 at 1:21 PM, helged said:

Thanks Gordon! 

Are there any image difference between the two GFX50 bodies? Any banding when lifting shadows? I ask of curiosity - haven't tried/tested any of the GFX bodies.

There has been no banding reported on the 50MP sensors. GFX 100 used to have banding caused by OSPDAF (only 100MP sensor) but it was fixed in the meantime.

Edited March 24, 2021 by SrMi

[ZHNL]

I agree with everyone who said SL2 AF is not slower compare to GFX100s though 100s do have visually eye AF. (Small box on eyes) I really can’t pick one over the others.  For the only lens I have (45mm), I think SL2 may be faster  with a few sigma primes I have.

Why SL2? I agree with Jon here about Leica ecosystem integration.(also regarding file quality of SL2 and 100s) Shooting a M glasses on SL2 feels better than any MILC out there other than M with better IQ. For me, it also means I can shoot awesome sigma AF glass such as 14-24, 24-70 and 35/45/65 i seres that cover wider range with smaller size and weight and have faster speed at the same time. 
 

I do feel struggle what to bring with me for future photo trip. GFX is hard to beat here. It almost circle all the box for me. The lens I tested with is every bit as the best glass out there. (45mm) almost zero flaws.

I maybe minority here but a huge plus FUJI over other is adaptability using Leica R and Nikon F glasses I collected over the years. The render is incredible and easily justify the price for by speed up my favorite glass by almost one stop in term of DOF.  Now IBIS make it a even more incredible tool. 
 

as for DR, you have to compare them with the same target print size. In that regard, I bet it will beat all out there other than IQ4 or IQ3100.  Though DR is really not the reason to get this one as only a few very limited cases can take advantage of half stop compare to 50s or 1 stop compare to best FF options out there.
 

The camera is incredible well rounded. Button and build feel so so  but I would say it beat my expectation. Highly recommended. 

Edited March 24, 2021 by ZHNL

[Mr.Q]

I'm not sure which lenses you guys are shooting with, but with the 32-64 the GFX100S is faster focusing in AF-S than the SL2 with 24-90, especially in mixed/dim lighting.

Of course, tracking in AF-C is not even comparable. 

Overall I find the AF system of the GFX100S faster and more reliable in any situation.

[FlashGordonPhotography]

7 hours ago, SrMi said:

I am not following.

Larger pixels generally do not mean lower noise floor unless looking at the pixel level. At the output level (same sized output), it is the larger sensors that have better DR. Pixel size is not significant. Examples: Nikon Z 6 vs. Z 7, Sony A7m3 vs. A7rm3 (same sensor size, different pixel size, same DR).
Multishot is a bit like frame averaging, bringing more "light' to the sensors, thus significantly increasing DR.

I do not understand what you mean by usable DR. Measured PDR is similar between 100MP and 50MP sensors until dual conversion gain kicks in.

https://www.photonstophotos.net/Charts/PDR.htm#FujiFilm%20GFX%2050S,FujiFilm%20GFX%20100

There is no (more) banding caused  by OSPDAF on GFX100 or GFX100s.

With pixel binning you'd be looking at a smaller sized output than the native sensor level. So if you could get the 100MP sensor to deliver a 50MP file you *may* get a small bump in DR. It's the same effect as high res shot without the colour benifits. Use more than one pixel to make one pixel.

The 100MP sensor doesn't have quite the real world DR that the 50MP one has. Same as how the Nikon sensors don't match the X1D sensor except on test charts. I did test them side by side when the GFX100 appeared. I have yet to see any comparison other than a test chart that disagrees.

There are many reports (and samples) of banding on the GFX100 when pushed hard. I have yet to push my GFX100S at all past basic tests.

Gordon

Edited March 24, 2021 by FlashGordonPhotography

[FlashGordonPhotography]

7 hours ago, SrMi said:

The DR will likely be the same as 50S Mark I. The only thing that could lift the DR would be dual conversion gain, which I think is not feasible with the 'old' sensor that is in X1D, GFX50 Mark I and likely in Mark II as well. It is likely that it will still require LENR to eliminate hot pixels.

The X1D does not require LENR. It doesn't even offer LENR as a function. I use it regularly for exposures over 5-10 minutes. Longest I've shot was 38 minutes. The Pentax 645Z is much the same.

Gordon

Edited March 24, 2021 by FlashGordonPhotography
had to check camera menus

[FlashGordonPhotography]

2 hours ago, Mr.Q said:

I'm not sure which lenses you guys are shooting with, but with the 32-64 the GFX100S is faster focusing in AF-S than the SL2 with 24-90, especially in mixed/dim lighting.

Of course, tracking in AF-C is not even comparable. 

Overall I find the AF system of the GFX100S faster and more reliable in any situation.

My 32-64 arrives tomorrow or Friday. I will certainly be interested in the results. Also have a 100-200 and 23 coming.

I must say I have zero issues in AFS with the SL2, even in low light.

Gordon

[SrMi]

Advertisement (gone after registration)

3 hours ago, FlashGordonPhotography said:

The X1D does not require LENR. It doesn't even offer LENR as a function. I use it regularly for exposures over 5-10 minutes. Longest I've shot was 38 minutes. The Pentax 645Z is much the same.

Gordon

Yes. I was talking about GFX50s Mark I and Mark II (rumored) (question asked by helged).

[SrMi]

3 hours ago, FlashGordonPhotography said:

With pixel binning you'd be looking at a smaller sized output than the native sensor level. So if you could get the 100MP sensor to deliver a 50MP file you *may* get a small bump in DR. It's the same effect as high res shot without the colour benifits. Use more than one pixel to make one pixel.

The 100MP sensor doesn't have quite the real world DR that the 50MP one has. Same as how the Nikon sensors don't match the X1D sensor except on test charts. I did test them side by side when the GFX100 appeared. I have yet to see any comparison other than a test chart that disagrees.

There are many reports (and samples) of banding on the GFX100 when pushed hard. I have yet to push my GFX100S at all past basic tests.

Gordon

As output, I meant a print or online share, not the generated raw file. Pixel binned, 50MP, and 100MP files should be all compared at the same output size. IMO, that is the more relevant criteria than looking at the pixel level.
I agree that test charts do not tell the whole story.

[setuporg]

It's encouraging to read about fast AF on the 100S, and amusing to see the demands for SL2 to speed up its AF or roll over.

My current experience with the SL2, X1Dii and S 007 shows that each AF implementation is intimately coupled with the overall ergonomics of each camera system and puts me in the photo mood appropriate for it.  SL2 is fast enough for all situations where I find myself.  The fastest AF I need it to catch kids or dogs or cats or pigeons, and it is sufficient.  The main problem with AF and such reasonably slow subjects is misfocusing.  Then the spot AF is best.  The S is perfect for it.  I've never needed multiple focus points really, so S is fine here.  The problem for S is the S100 (don't confuse with 100S:), called "the S Noctilux", which has such a narrow DOF that it focuses on eyelashes instead of irises or the nose or such visibly enough to frustrate one after the fact.  

The slow AF of the X1D i/ii is actually a feature.  When shooting with X1D or 907x, I feel that it takes its time to lock on the target.  The delay is comparable to the manual focusing with M.  You train the AF on the area you need in focus, taking the time to see and ascertain it is what you want.  Instead of superimposing two rectangles, you ensure you got the right AF point confirmed.  The delay guides your mind in the same way the M focusing does -- you prefocus mentally, choosing the object in focus.  If it is moving you prepare for that.  I don't experience X1D slow AF as a deal breaker, rather as a specialty.  Given everything shared so far, I'd rather hold on to all the XCD lenses and wait for X2D than chase the new GFX body and its own menagerie of lenses.  Yes, Fujifilm lenses have real aperture rings which is preferable, but the XCD lenses shine.

[FlashGordonPhotography]

6 hours ago, SrMi said:

Yes. I was talking about GFX50s Mark I and Mark II (rumored) (question asked by helged).

Does the GFX50S have some hot pixels at long exposures? Honest question, as I would have assumed it behaved much like the X1D or 645Z. Possibly a processor issue rather than the sensor? Interesting....

Gordon

[bags27]

Apologies if this was already posted. I just saw it elsewhere:

https://www.photonstophotos.net/Charts/PDR.htm#FujiFilm GFX 100,Hasselblad X1D-50c,Leica M10 MONOCHROM,Leica M10-R,Leica SL2-S

Edited March 24, 2021 by bags27

[SrMi]

2 hours ago, FlashGordonPhotography said:

Does the GFX50S have some hot pixels at long exposures? Honest question, as I would have assumed it behaved much like the X1D or 645Z. Possibly a processor issue rather than the sensor? Interesting....

Gordon

IIRC, GFX50S had hot pixels at long exposures unless turning LENR on. Pentax 645z had a mandatory LENR (slow shutter NR is either on auto) until a firmware update allowed turning it off. I assume that any camera offering LENR has hot-pixels at long exposures.

[aksclix]

1 hour ago, SrMi said:

IIRC, GFX50S had hot pixels at long exposures unless turning LENR on. Pentax 645z had a mandatory LENR (slow shutter NR is either on auto) until a firmware update allowed turning it off. I assume that any camera offering LENR has hot-pixels at long exposures.

Why only cameras offering LENR? longer exposures produce hot pixels no matter what 

[LD_50]

11 minutes ago, aksclix said:

Why only cameras offering LENR? longer exposures produce hot pixels no matter what 

Because if they offer it, it’s needed. If they don’t offer it, they may have solved for hot pixels another way. 

[FlashGordonPhotography]

4 hours ago, bags27 said:

Apologies if this was already posted. I just saw it elsewhere:

https://www.photonstophotos.net/Charts/PDR.htm#FujiFilm GFX 100,Hasselblad X1D-50c,Leica M10 MONOCHROM,Leica M10-R,Leica SL2-S

What's VERY interesting is that according to PTP the GFX100S has almost a stop better DR than the GFX100 although the sensors are identical. Probably some firmware improvements going on as hardware is identical.

I have all those cameras except the SL2S. It'll be interesting to see how they compare in the real world.

Gordon

[aksclix]

1 hour ago, LD_50 said:

Because if they offer it, it’s needed. If they don’t offer it, they may have solved for hot pixels another way. 

Hmm.. for example? Any cameras that didn’t have LENR and solved hot pixels another way? I probably never used one such camera so asking 

[bags27]

1 hour ago, FlashGordonPhotography said:

What's VERY interesting is that according to PTP the GFX100S has almost a stop better DR than the GFX100 although the sensors are identical. Probably some firmware improvements going on as hardware is identical.

I have all those cameras except the SL2S. It'll be interesting to see how they compare in the real world.

Gordon

There's some comparative "talk" here; not photos but related experiences. Of course, it's on a medium format board, so the Fuji may have an advantage.

https://www.dpreview.com/forums/thread/4563110

[LD_50]

1 hour ago, aksclix said:

Hmm.. for example? Any cameras that didn’t have LENR and solved hot pixels another way? I probably never used one such camera so asking 

The X1D was mentioned multiple times on this thread as not having or needing LENR. I don’t own that camera so I don’t know it’s capabilities.  

I think some MF cameras offered dark frame subtraction done whenever the photographer chose, rather than at the time of exposure. Phase One or Hasselblad if I remember correctly. 

[SrMi]

1 hour ago, aksclix said:

Hmm.. for example? Any cameras that didn’t have LENR and solved hot pixels another way? I probably never used one such camera so asking 

Of the cameras I own, Hasselblad H6D and X1D have solved hot-pixels another way. AFAIK, they are the only ones who do not need LENR.

[Advertisement]

Hallo,

Look here: Leica L-Mount

[aksclix]

13 minutes ago, SrMi said:

Of the cameras I own, Hasselblad H6D and X1D have solved hot-pixels another way. AFAIK, they are the only ones who do not need LENR.

Thanks both of you.. I’ve had the x1d for a month and a half but haven’t quite used it much yet.. it’s great to know this about the camera..