Pages: 1 2 3 [4] 5 6 7   Go Down

Author Topic: Fujifilm GFX-100 actual *user* experience, tips, setup, please?  (Read 17915 times)

Jim Kasson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2370
    • The Last Word
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #60 on: August 08, 2019, 03:42:07 pm »

Jim, I assume that you are using Adobe raw converters to analyze GFX100 images. Have you tried using Capture One instead (I believe there is a 30-day demo)?
Apparently, DPR uses Adobe raw converters for all its camera tests but uses Capture One for GFX100 studio scene instead.
I am not interested in switching to C1 but wonder if Adobe can improve its support of GFX100 files (currently marked as preliminary support).

Yes, I'm using ACR and Lr. At this point, I'm not concerning myself much with the color from the GFX 100. If the final version of the Adobe profiles don't cut the mustard, I'll consider using C1, but I don't like the workflow.

Jim

eronald

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6642
    • My gallery on Instagram
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #61 on: August 08, 2019, 03:45:48 pm »

Lightroom's profiles for the two cameras yield markedly different results. The GFX 100 profile is probably not the one they'll end up with. I believe that's the source of the color differences.

As to the debanding filter, I know of now commercial debanding filter at present. Do you have a link to one? Has RT maped the PDAF rows for the GXF100?

Jim
One would need to take a pic with the same camera, same light, 14 and 16 bit. Otherwise ...

I have no idea what debanding filters are out there for visual spectrum cameras.  My thought is that Sony must have done some research on this when they designed the sensor layout, and would have communicated with their customers.

Edmund
Logged
If you appreciate my blog posts help me by following on https://instagram.com/edmundronald

Jim Kasson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2370
    • The Last Word
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #62 on: August 08, 2019, 03:52:56 pm »

One would need to take a pic with the same camera, same light, 14 and 16 bit. Otherwise ...

I have no idea what debanding filters are out there for visual spectrum cameras. 

But you are telling me that I should have fixed it, right?

My thought is that Sony must have done some research on this when they designed the sensor layout, and would have communicated with their customers.

This issue has been around in one form or another for a long time: a7RIII, Z6, Z7, a9, etc, and I've never seen any communication from the camera manufacturers even acknowledging that there is a problem, much less offering a solution.

By the way, the color differences in 14- and 16-bit modes are probably also influenced by the blackpoint of the sensor, and the precision of its representation. I note that the EXIF blackpoints for 14 and 16 bit mode are not precisely a factor of four apart. The 16-bit black point isn't evenly divisible by two, let alone four.

Jim

eronald

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6642
    • My gallery on Instagram
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #63 on: August 08, 2019, 07:20:44 pm »

But you are telling me that I should have fixed it, right?
Jim


A  hi-fi with a mains hum - is it still hi-fi?

I wasn't saying *you* should fix it, I was saying that when doing any sort of metrics one needs to state how well one can fix it by usually implemented methods, or else the metrics are a bit of a lie.

I can see that a lot of people are very happy with the GFX-100 and it is a well received tool. However saying it has eg. 13 bits of DR is a bit lawyerly if whenever you try to use the last few bits you see a grid or stripes superimposed on your image.

Of course, in practice, the camera's buyers will employ some sort of filtering which will erase this grid, so I guess we should try to get meaningful performance  numbers with the filter applied, and that includes an indication of the degradation effected by the grid *and* the filter.


Edmund

« Last Edit: August 08, 2019, 09:54:10 pm by eronald »
Logged
If you appreciate my blog posts help me by following on https://instagram.com/edmundronald

Jim Kasson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2370
    • The Last Word
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #64 on: August 08, 2019, 11:40:04 pm »


A  hi-fi with a mains hum - is it still hi-fi?

I wasn't saying *you* should fix it, I was saying that when doing any sort of metrics one needs to state how well one can fix it by usually implemented methods, or else the metrics are a bit of a lie.

I can see that a lot of people are very happy with the GFX-100 and it is a well received tool. However saying it has eg. 13 bits of DR is a bit lawyerly if whenever you try to use the last few bits you see a grid or stripes superimposed on your image.

Of course, in practice, the camera's buyers will employ some sort of filtering which will erase this grid, so I guess we should try to get meaningful performance  numbers with the filter applied, and that includes an indication of the degradation effected by the grid *and* the filter.



What are the "usually available methods"? As far as I know, there is no commercial product currently available to fix this.

32BT

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3095
    • Pictures
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #65 on: August 09, 2019, 01:30:12 am »

What are the "usually available methods"? As far as I know, there is no commercial product currently available to fix this.

SOOC jpeg?
Logged
Regards,
~ O ~
If you can stomach it: pictures

Jim Kasson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2370
    • The Last Word
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #66 on: August 09, 2019, 09:45:36 am »

SOOC jpeg?

Is it your claim that there is no PDAF banding in SOOC JPEG's? Haven't tested that, but a 5-stop EV boost and a +100 shadow boost wouldn't work well in any SOOC camera JPEG that I know of.

Doug Peterson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4210
    • http://www.doug-peterson.com
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #67 on: August 09, 2019, 10:27:05 am »

I'll consider using C1, but I don't like the workflow.

Jim, I humbly submit that you might change your mind on C1 workflow if presented with a wholistic soup-to-nuts deep dive on how to get the most out of C1 workflow wise. Some software is pretty easy to self-teach and end up the same place as a formal walk through, but I don't find C1 to fit in that category. We teach a C1 Masters Class and I'd be very glad to comp a seat for you in recognition of your contributions to the community over the years.

I suspect you'd come away from the class enjoying working with C1 a great bit more. But also, if you still don't like C1 after that class, you can be very confident it's not worth further effort :). Win:Win.

Jim Kasson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2370
    • The Last Word
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #68 on: August 09, 2019, 10:32:29 am »

Jim, I humbly submit that you might change your mind on C1 workflow if presented with a wholistic soup-to-nuts deep dive on how to get the most out of C1 workflow wise. Some software is pretty easy to self-teach and end up the same place as a formal walk through, but I don't find C1 to fit in that category. We teach a C1 Masters Class and I'd be very glad to comp a seat for you in recognition of your contributions to the community over the years.

I suspect you'd come away from the class enjoying working with C1 a great bit more. But also, if you still don't like C1 after that class, you can be very confident it's not worth further effort :). Win:Win.

Thanks for the generous offer. I'll think about it. Does C1 have code to mitigate the GFX 100 PDAF banding? Can you brush in the fix?

Jim

Doug Peterson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4210
    • http://www.doug-peterson.com
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #69 on: August 09, 2019, 10:41:06 am »

Thanks for the generous offer. I'll think about it. Does C1 have code to mitigate the GFX 100 PDAF banding?

It's an entirely different math pipeline, and I found the pixel-level technical quality from my Fuji X Pro 1 and Fuji XH1 to be significantly better in C1 than LR last I tested. But I could not speak to whether any of the pipeline is specific to addressing the GFX 100 PDAF banding.

Can you brush in the fix?

There is definitely not a specific-to-this-problem tool that could be brushed in. But almost any adjustment in C1 can be applied as a local adjustment with a wide range of tools for masking and refining that mask; including tools commonly used to present noise and poor pixel quality in the best visual way, such as noise reduction, sharpening, color editor, and curves.

Jim Kasson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2370
    • The Last Word
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #70 on: August 09, 2019, 11:06:09 am »

It's an entirely different math pipeline, and I found the pixel-level technical quality from my Fuji X Pro 1 and Fuji XH1 to be significantly better in C1 than LR last I tested. But I could not speak to whether any of the pipeline is specific to addressing the GFX 100 PDAF banding.

There is definitely not a specific-to-this-problem tool that could be brushed in. But almost any adjustment in C1 can be applied as a local adjustment with a wide range of tools for masking and refining that mask; including tools commonly used to present noise and poor pixel quality in the best visual way, such as noise reduction, sharpening, color editor, and curves.

Thanks. The issue exists in the raw files, so it needs a fix in the raw developer. The software that fixes it has to know the PDAF pixel row spacing, which in the GFX 100 is 18 pixels.

On another front, there is interest in the Raw Therapee community in coding up a fix, just as they did for the Sony and Nikon MiLCs with the problem.

Jim

kers

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4388
    • Pieter Kers
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #71 on: August 09, 2019, 11:30:45 am »

Do i understand that you have a 100mp sensor with after every 18 rows of pixels a blank row.
In effect a 95mp pixel sensor with horizontal stripes that have to be interpolated.
This and the bayer pattern need a certain form of interpolation.
Obviously this concept is a type outside the concept of DNG and may differ with each sensor type.
Enhance detail will not work and may even enhance striping.
Logged
Pieter Kers
www.beeld.nu/la

Jim Kasson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2370
    • The Last Word
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #72 on: August 09, 2019, 11:42:42 am »

Do i understand that you have a 100mp sensor with after every 18 rows of pixels a blank row.

That's not how it works, but the PDAF rows have fewer image-forming pixels.

Jim

32BT

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3095
    • Pictures
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #73 on: August 09, 2019, 01:37:34 pm »

Is it your claim that there is no PDAF banding in SOOC JPEG's? Haven't tested that, but a 5-stop EV boost and a +100 shadow boost wouldn't work well in any SOOC camera JPEG that I know of.

Not a claim. It had a question mark behind it. Perhaps Fuji internals does compensate for the pixels. Perhaps not. If it does, these pixels effectively become like the ones on the ancient SuperCCD, therefore my remark about increased DR, not decreased DR.

They are btw not blank pixels as far as I can tell. I can not see any problems on one of dpreview's raw files for example. What do you mean therefore with "fewer image-forming pixels"?

Regarding 16bit mode, the following still seems interesting and relevant: considering the "reduced sensitivity" of the Red and Blue channels, how does 16bit mode affect the bit precision of these channels?
Logged
Regards,
~ O ~
If you can stomach it: pictures

Jim Kasson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2370
    • The Last Word
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #74 on: August 09, 2019, 02:30:47 pm »

Not a claim. It had a question mark behind it. Perhaps Fuji internals does compensate for the pixels. Perhaps not. If it does, these pixels effectively become like the ones on the ancient SuperCCD, therefore my remark about increased DR, not decreased DR.

Pretty useless if it does, since you'd be limited my the DR of the gamma-compressed, DCT-compressed JPEG files.

They are btw not blank pixels as far as I can tell. I can not see any problems on one of dpreview's raw files for example. What do you mean therefore with "fewer image-forming pixels"?

How can you tell? The pixels in the raw file are interpolated over.

The PDAF pixels are not used to form the image in the raw file. Their locations are filled by the camera using interpolation.

There is something in Fuji's fixing up of the PDAF pixels in camera that gets the mean wrong. This is not the case with Sony OSPDAF MILCs. On those cameras, the sigma changes in the PDAF rows, but, absent flare, the mean is correct.

Regarding 16bit mode, the following still seems interesting and relevant: considering the "reduced sensitivity" of the Red and Blue channels, how does 16bit mode affect the bit precision of these channels?

In 16 bit mode, those channels have 16 bit precision, just like the green channels. Read noise is independent of raw plane assignment, to a first order.

Jim

32BT

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3095
    • Pictures
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #75 on: August 09, 2019, 03:33:38 pm »

Pretty useless if it does, since you'd be limited my the DR of the gamma-compressed, DCT-compressed JPEG files.

Considering where the problem resides, I'm fairly sure 3 gamma-encoded 8bit channels trump 1 linear 16bit channel any time of the day, even with all of the heavy JPEG compression. Still, what you might try is how the camera reacts to severe shadow punishment as much as it allows before it spits out that jpeg. Another issue of course is blackframe subtraction which I believe Fuji allows with a high iso noise reduction setting some where IIRC.

How can you tell? The pixels in the raw file are interpolated over.

Because I'm looking at the CFA data, see attached. It represents the bottom of the DPR test file, generally containing the problematic pixels. Pushed the levels up quite high, but would be hard pressed to say I see significant banding of any relevant kind. Admittedly, I need to write some code to get a graybalanced representation, which can sometimes reveal more problems. I'll do that tomorrow. If you have a raw file you want to specifically look at for those pixels, I might be able to run it here.


The PDAF pixels are not used to form the image in the raw file. Their locations are filled by the camera using interpolation.

There is something in Fuji's fixing up of the PDAF pixels in camera that gets the mean wrong. This is not the case with Sony OSPDAF MILCs. On those cameras, the sigma changes in the PDAF rows, but, absent flare, the mean is correct.

Okay, I might give that a check tomorrow as well if I get around to it.

In 16 bit mode, those channels have 16 bit precision, just like the green channels. Read noise is independent of raw plane assignment, to a first order.

Yes, I understand, but apparently the 16bits of the green channel do not provide increased DR, at least as I understand from the discussion. The question though is what happens for the lower values of the R and B channels? If those values are encoded with actual more precision (not just rescaled from 14bit) then there is a significant advantage for raw converters to compute crosschannel mixes. This can reduce the noise produced by the converter (not by the camera).

Logged
Regards,
~ O ~
If you can stomach it: pictures

Jim Kasson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2370
    • The Last Word
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #76 on: August 10, 2019, 05:13:00 pm »

Another issue of course is blackframe subtraction which I believe Fuji allows with a high iso noise reduction setting some where IIRC.

The GFX 100 won't do blackframe subtraction at these shutter speeds.

Jim

32BT

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3095
    • Pictures
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #77 on: August 11, 2019, 01:40:18 am »

The GFX 100 won't do blackframe subtraction at these shutter speeds.

Jim

I understand, but what I recall from the E3 was that it was recommended to turn long-exp nr off since it did affect the end result regardless of exposure time, but that may be sooc jpeg. I unfortunately don't recall the source.

The other thing why I mention this: the original Sony A7 seems to make an exposure of some kind after first use. That is: when it has not been used for a long time, and then turning it on and shooting a couple of frames, and then turning it off, it then will make a relatively long exposure. It may be some kind of sensor cleaning method, but it could be a blackframe exposure which it keeps in long sleep mode.

Obviously, we are far away from that generation of Sony chip, but it gave food for thought: if it is indeed some kind of ad-hoc type of session blackframe, and we base all kinds of testing results on one particular session...
Logged
Regards,
~ O ~
If you can stomach it: pictures

32BT

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3095
    • Pictures
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #78 on: August 11, 2019, 02:10:14 am »

I separated the channels, that way any patterns will not be obscured by the bayer grid itself.

I have to admit that there seems to be banding in both of the green channels that is not visible in the red and blue. Not sure if this is related to pdaf pixels, since the banding seems to modulate over several rows. The banding is visible on even as low as +2 stops.

I'll attach 4 images on the next 4 posts: R, G1, G2, B @ +2 stops, +4, +6, and +8 stops

Logged
Regards,
~ O ~
If you can stomach it: pictures

32BT

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3095
    • Pictures
Re: Fujifilm GFX-100 actual *user* experience, tips, setup, please?
« Reply #79 on: August 11, 2019, 02:11:22 am »

+2 stops
Logged
Regards,
~ O ~
If you can stomach it: pictures
Pages: 1 2 3 [4] 5 6 7   Go Up