Home

Please support mflenses.com if you need any graphic related work order it from us, click on above banner to order!
SearchSearch MemberlistMemberlist RegisterRegister ProfileProfile Log in to check your private messagesLog in to check your private messages Log inLog in

Camera colors: Sony A900/A7 vs Fuji GFX
Post new topic   Reply to topic
Home   Manual Focus Lenses
View previous topic :: View next topic  


PostPosted: Sun Mar 12, 2023 2:52 pm    Post subject: Camera colors: Sony A900/A7 vs Fuji GFX Reply with quote
In another thread (about adapting vintage glass to theFuji GFX series there were quite a few remarks about the differences in color rendering betwenn Sony A7 series and Fuji GFX series of cameras.

Examples are

cbass wrote:
... people and skin tones. There is something very wrong with the Sony.


caspert79 wrote:

It’s a stubborn rumor IMO that certain camera brands have significantly worse colors than other brands.


blotafton wrote:
Sony are historically known for having bad colors. Specifically a green or yellow tint ... Fuji has the best colors in my subjective opinion.


calvin83 wrote:
I feel the SOOC JPEG with auto AWB from GFX100s have much nicer color than the SOOC JPEG from my Sony A7RII when both camera using the default JPEG settings.


I have been shooting quite extensively and in parallel under controlled conditions (flashlight) with the Fuji GFX 50R and the Sony A7RII, using the same lenses (e. g. Mamiya Sekor A 4/120mm Macro): Golden objects, silver objects, historical paintings and clothes. No portraits, no landscape, and very little architecture.

Looking at the results (and manually selecting identical values eg fo white balance), the colors OOC were very similar. Looking ate the results with experts for restauration (who are very critical about colors), there was no clear preference for either Sony or Fuji. A few images were considered to be slightly better with Fuji (usually clothes), and gold came slightly better with the A7RII (according to the experts). But again: no clear preference for one or another.

Certainly both cameras would do the job with very good results. However, since I already owned the A7RII, I refrained from buying into a new additional system (back then no real superwides, let alone shift lenses were available for the GFX).


PostPosted: Sun Mar 12, 2023 3:13 pm    Post subject: Reply with quote
Wow just the other day, I've been reading how good Fuji was doing comparing to Sony bad skin tones , the colors were soooo different, that I couldn't believe that difference would be true for raw files. Besides ,I do know people that have switched back to Fuji ,not being happy with Sony colours (I dunno if it matters the size of the sensor though). So a technical comparison from honest people shooting both of the mentioned cameras(maybe same lens,so no other factors would affect results) , would be nice . On jpeg ,most of the people seem happy with Fuji and no that many on Sony, so....I'm kind of confused (lately I prefer shooting raw) .


PostPosted: Sun Mar 12, 2023 4:06 pm    Post subject: Reply with quote
This paper may be useful to this topic: Use of commercial off-the-shelf digital cameras for scientific data acquisition and scene-specific color calibration https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4028365/ .


PostPosted: Sun Mar 12, 2023 4:25 pm    Post subject: Reply with quote
Much more interesting for me was comparing the A900 with the GFX 50R.

I still am using several A900 DSLRs for landscape and calendar photography. The A900 (A850) camera was largely designed by Minolta engineers. The A900 colors are different from the previous Minolta D7D and D5D cameras, also because of their CMOS sensors (vs CCD in the D5D, D7D and A100), but quite different from the later A7 series colors as well.

The A900 series has extensive controls for adjusting the JPGs not seen in any other digital cameras.

1) There's its specific "DRO" (Dynamic Range Optimization) which runs on a separate chip designed by Michael Tusch / Apical (London). Similar (identical?) chips also were used in the extremely expensive professional Sony video cameras of that time. All other cameras I'm aware of use "sub-software" on the camera image processor (often also by apical).

2) You can adjust "Contrast" and "Saturation" (common), but also "Brightness" and "Zone" (unusual and limited to cameras with the additional Apical chip): Underexposing the RAW and increasing its brightness keeps the highlights, and really pushes the shadows. These two latter values, only found in cameras with the separate Apical chip (ie A900 and A850) give you an unique control about the resulting JPGs (OOC). Usually I cannot replicate these results using either Photoshop, C1, or Sony IDC and RAWs.

When I got the GFX 50R in 2020 I decided to take a few landscape shots, comparings its JPGs to the A900 JPGs. The main advantage of the Sony A900 JPGs are their excellent balance between dark shadows and bright (usually sky) areas. Very natural looking, and easy to prepare for printing.

OK - let's go for it.

JPG out of the GFX 50R. AE, manual white balance ("cloudy" = 6500K), and no shadow / highlight adjustment:


Clouds are completely overexposed (quite large areas of "255/255/255") in the JPGs, and shadows are equally underexposed ("0/0/0). Not good.

Now let's adjust the Highlight / Shadow compensation of the GFX50 to its highest values (two separate actions hidden deep within the menu):


Slightly better, but still nothing to write home about. Surprisingly, the clouds still are blown out, and the darkest shadows are ... well "1/1/1" now.

Now the JPG from the A900. Manual white balance set to 6000K (due to warmer colors of the Zeiss 2.8/16-35mm used here). DRO was set to "level 5" (maximum). On the A900 just one value is controling both the shadow and the highlights. And the DRO setting (levels 1 to 5 depending on your intentions) is accessible directly viaan external button (no need for multiple clicks into the menu).
No blown out clouds, and the deepest shadows are at "5/10/5" approximately. The entire image looks pertty "balanced" to my eye.



Now a Photoshop RAW conversion of the A900 RAW using "Auto values" plus max compensation for shadows/highlights:



Then the Photoshop RAW conversion of the GFX50R RAW using "Auto values" plus max compensation for shadows/highlights results in this:


Slightly enhancing the A900 JPG (DRO level 5) quickly adjusting whites/highlights and black/shadows via photohop (twenty seconds maybe):


If necessecary one could even go further (colors both of the dark tree as well as the sunny hills and the white clouds as seen while taking the image):


Compare this to the Fuji GFX50R JPG, treaded similarly as the above A900 JPG, using Photoshop (just whites/highlights and black/shadows adjusted - look how crazy the colors are geting now ...):


That's how I've been evaluating the GFX50R for landscapes, back in 2020. Not suitable for my landscapes purposes, even though the GFX 50R of course has a much better resolution that the A900. These days, I can use the A900 plus Zörk shifting adapter and Mamyia MF lenses if I really need very big "landscape" files (6000x9000 px / 54 MP). Other high res stuff is done with the A7RII plus 17mm Shift, 35mm Shift, and 50mm / 100mm macro primes.

Maybe I mades some mistakes - and maybe others can publish their comparisons as well.

S

Last edited by stevemark on Sun Mar 12, 2023 4:46 pm; edited 4 times in total


PostPosted: Sun Mar 12, 2023 4:32 pm    Post subject: Reply with quote
Since this has migrated to another thread, I think it's best to provide some background. In another thread someone said the Fuji GFX and Sony a7R IV are virtually identical based on the comparison tools on dpreview. I looked at both looking at skin tones and said maybe resolution is similar but the skin tones I am seeing they are not virtually identical.





For your own comparison:

https://www.dpreview.com/reviews/fujifilm-gfx-50r-review/4

Color is a complex topic, but this is not the first time I have seen Sony body's turn people into Simpson's characters.

Was this an error due to lighting or WB on dpreview? Color cast on the lens used? Perhaps.
Is this something that only affects a certain body? Also possible.

Now I say color is complex because it is. Lenses affect colors. Color is not always consistent between camera bodies. Manufacturers apply their own internal color profiles and curves and they do not always match 100% between bodies. Processing from RAW? The software also has its own interpretation of color as well as standard curves it applies. Lightroom is different from Capture One which is different from Luminar and all are different from the in-camera processor. This completely ignores color differences due to WB algorithms and ambient light as well as color casts from artificial lights.


PostPosted: Sun Mar 12, 2023 4:55 pm    Post subject: Reply with quote
@stevemark

It seems like the goal posts shifted? I looked at the cameras in the original thread. 50R and a7R IV. The A900 is a 24 MP DSLR the others are higher resolution mirrorless.

Landscape is difficult to control unless zero clouds or fully overcast. I have shot landscape and even midday the lighting will change within minutes and the effects can be significant from greater saturation to more micro contrast due to more contrast or intense light.

It seems like you are trying to justify the choice you made? No need. If you are happy with your Sony and it works to your satisfaction that is all that matters. Different strokes for different folks.


PostPosted: Sun Mar 12, 2023 5:06 pm    Post subject: Reply with quote
I this context I would love to know how RAW sensor readout actually works. A lot would depend on whether the amplifiers and D/A converters are on-chip, or whether they are implemented in separate camera circuitry. If they are implemented in separate camera circuitry, then even given an identical sensor, the manufacturer's choice of amplifier and D/A conversion circuitry would result in different S/N ratios and dynamic range.

Stephan seems to have had more involvement with SONY developers here?; are the amplifier circuitry and D/A conversion in the SONY A7 line cameras done on-sensor, or in separate circuitry?

Are there any good books on actual sensor photonics, amplification, readout & D/A circuitry as used in consumer digital cameras? (possibly a closely guarded IP)

EDIT: just noticed I mentioned D/A conversion above which should off course have read A/D conversion... Rolling Eyes (I need more caffeine)

Last edited by RokkorDoctor on Sun Mar 12, 2023 7:45 pm; edited 1 time in total


PostPosted: Sun Mar 12, 2023 5:39 pm    Post subject: Reply with quote
I think it needs to be said~ especially for those using m/f lenses on digital:

For skin tones, get back to the basics first, then work your way up.
My D-810, which wears a Sony made sensor btw Wink , gives little trouble with skin tones, provided exposure is correct.
The metering systems on modern cameras can still be fooled. It still never hurts to run a few checks while shooting.
My particular set-up seems to prefer 1/3 of a stop under exposure in A, and P modes.
I still find myself going to manual in tricky lighting situations, despite all the accuracy that my modern system packs in automated modes.

The one place I have trouble with skin tones is using the camera's somewhat poor on-board non-bouncing flash, which is probably a bit weak to begin with.

-D.S.


PostPosted: Sun Mar 12, 2023 5:42 pm    Post subject: Reply with quote
cbass wrote:
In another thread someone said the Fuji GFX and Sony a7R IV are virtually identical based on the comparison tools on dpreview. I looked at both looking at skin tones and said maybe resolution is similar but the skin tones I am seeing they are not virtually identical.

Color is a complex topic, but this is not the first time I have seen Sony body's turn people into Simpson's characters.

Was this an error due to lighting or WB on dpreview? Color cast on the lens used? Perhaps.


White balance or light source error at DPR, almost for sure. More on tha later.

Here are two 100% crops from the 50 MP Fuji GFX 50R (right) and from the Sony A7RII (left). The crops are from the border, and both images were taken with the very same lens (Mamiya Sekor A 4/120mm Macro) within a few minutes (indirect natural light from the sun):



Both crops above from RAWs - same white balance (manually set - IT'S IMPORTANT), and same exposure / contrast / etc ... values used.


PostPosted: Sun Mar 12, 2023 5:45 pm    Post subject: Reply with quote
Doc Sharptail wrote:

My D-810, which wears a Sony made sensor btw Wink , gives little trouble with skin tones, provided exposure is correct.


I keep hearing that Sony makes all these sensors. They do not. Each manufacturer designs their own sensor, and it is the OEM's design. What they do is use Sony's foundry to produce them. Foundries are extremely expensive. Billions of dollars and it's more economical to outsource to an existing foundry. If Sony made all the sensors, then why wouldn't they keep the best designs for themselves and hold them back from the competition?

If Leica or Zeiss sources their glass from Hoya or Schott does that mean Hoya or Schott made their lenses?


PostPosted: Sun Mar 12, 2023 5:51 pm    Post subject: Reply with quote
stevemark wrote:

Both crops above from RAWs - same white balance (manually set - IT'S IMPORTANT), and same exposure / contrast / etc ... values used.


If you are using Lightroom and selecting the same profile, then Lightroom is equalizing the colors for you from the RAWs. That is the purpose of lightroom profiles. If you are shooting an event and you are using multiple cameras, then you don't want the colors and look all over the place. You want a consistent look. When you process the RAW's in lightroom using the same profile, then lightroom is going to match those colors as best as possible. Of course, this is not a guarantee. Adobe does the best that it can, but it is not always perfect and sometimes wildly off.

Furthermore, white balance is not always enough when using different lenses. I recently did a Summilux vs Topcor test with a manually set white balance and color was not the same. The Topcor was cooler and the Summilux was warmer in the same shooting conditions.

There is something more complex that also happens within the body that I don't understand yet. During that test both the Summilux and Topcor although one is warmer and the other is cooler and the white balance was manually set shifted colors at f/4 and that shift went away by f/8. Went away is poorly worded. The color shift changed from f/4 to f/8 but both lenses looked more similar in how the color shifted at f/4 and the same at f/8 with a different color shift. Different lenses, but pretty much a similar focus shift at the same aperture. My assumption is some processing within the body or the sensor.

Again, it seems like you are trying to justify the choice you made to others. I don't think you need to because people shoot different subjects and cameras behave differently. There are dozens if not hundreds of design optimizations and choices to make from the lenses to the sensor to the processing and probably others I am not considering. For me it would double the work to get anything useable out of the Sony, but your experience may vary.

Last edited by cbass on Sun Mar 12, 2023 6:32 pm; edited 4 times in total


PostPosted: Sun Mar 12, 2023 5:59 pm    Post subject: Reply with quote
cbass wrote:
Doc Sharptail wrote:

My D-810, which wears a Sony made sensor btw Wink , gives little trouble with skin tones, provided exposure is correct.


I keep hearing that Sony makes all these sensors. They do not. Each manufacturer designs their own sensor, and it is the OEM's design. What they do is use Sony's foundry to produce them. Foundries are extremely expensive. Billions of dollars and it's more economical to outsource to an existing foundry. If Sony made all the sensors, then why wouldn't they keep the best designs for themselves and hold them back from the competition?

If Leica or Zeiss sources their glass from Hoya or Schott does that mean Hoya or Schott made their lenses?


Actually, no.

There is plenty of photographic evidence of Sony markings on nikon camera sensors.
They are made by Sony.

-D.S.


PostPosted: Sun Mar 12, 2023 6:07 pm    Post subject: Reply with quote
Doc Sharptail wrote:

Actually, no.

There is plenty of photographic evidence of Sony markings on nikon camera sensors.
They are made by Sony.

-D.S.


You can always buy a ready-made sensor from Sony, but because you use Sony's foundry does not mean it is a Sony designed sensor. Same with Leica back in the day. Sometimes they bought a Minolta design and glass from Minolta. Other times they bought glass from Minolta. Very different things.

https://petapixel.com/2018/06/15/the-nikon-d850s-sensor-is-made-by-sony-report/

Quote:
The Nikon D850 contains one of the most highly regarded sensors on the market today, and Nikon has touted the fact that it was designed entirely by Nikon and not simply an off-the-shelf sensor. But in case you were wondering who its manufactured by, a new report has now concluded that it’s a Sony-made sensor.


Quote:
ChipMod also tells NikonRumors that the characteristics of the sensor are typical of Sony-made sensors, but that the sensor is clearly different from the sensors found in Sony’s a7 II and a7 III mirrorless cameras.

This seems to confirm the fact that Nikon designed the D850’s sensor and had Sony manufacture it to Nikon’s specifications.


In your case the D810 is an off-the-shelf Sony sensor. The D850 is made in a Sony foundry but designed by Nikon. I just see people paint this stuff with a broad brush.

Last edited by cbass on Sun Mar 12, 2023 6:19 pm; edited 2 times in total


PostPosted: Sun Mar 12, 2023 6:08 pm    Post subject: Reply with quote
RokkorDoctor wrote:

Stephan seems to have had more involvement with SONY developers here?;

My contacts were limited to the 2007-2011 time frame. The two guys managing photo stuff (M. Gessler) and his senior manager (O. Emmerich) were very easy going, and so were the guys from Japan (including the chief engineer guiding the DSLR team, and the chief engineer for camera lenses). Around 2012 Sony Germany took over the Swiss branch, and an Austrian (working in Vienna) was made head of Sony Switzerland, too. Not a good choice. The misconfigured and therefore unlucky A77 / A99 "mirrorless DLSRs" (remember tem??) were introduced, and I decided not to work with Sony any more. So did the Gessler / Emmerich duo, and others. At some point Japan intervened, and Emmerich now is back, and head of the entire Sony Switzerland.

RokkorDoctor wrote:

are the amplifier circuitry and D/A conversion in the SONY A7 line cameras done on-sensor, or in separate circuitry?


Probably. The 24 MP A900 sensor was the first to have the A/D conversion integrated on-sensor (6000 A/D converters, one for each row). In addition there's a 24dB amplifier on-sensor (pre A/D?? digital? I don't know), and before A/D conversion there's some analogue noise reduction (correlated double sampling - sensor is read twice, and noise is subtracted).

Read at page 20 here in my A900 book:
http://artaphot.ch/images/PDFs/AlphaSystemBuch_I/Sony_Alpha_Vollformat-System_00-23_72dpi.pdf

That said, I have no further knowledge on the sensors; by far most of discussions and testing back then was related to lenses (including pre-prod new stuff).

S


PostPosted: Sun Mar 12, 2023 6:24 pm    Post subject: Reply with quote
cbass wrote:
Doc Sharptail wrote:

My D-810, which wears a Sony made sensor btw Wink , gives little trouble with skin tones, provided exposure is correct.


I keep hearing that Sony makes all these sensors. They do not. Each manufacturer designs their own sensor, and it is the OEM's design. What they do is use Sony's foundry to produce them. Foundries are extremely expensive. Billions of dollars and it's more economical to outsource to an existing foundry. If Sony made all the sensors, then why wouldn't they keep the best designs for themselves and hold them back from the competition?


Here I have some firsthand background information from early 2011. Japan had been devastated by the great earthquake. Sony DSLR production was devastated. Nikon Sendai (professional stuff) was destroyed. The chip production was halted completely. A huge mess, not to speek about radiation and related fears. I had been invited to go to Japan by the Sony chief of lens design, butof course the visit was cancelled.

Now the sensors ... Sony mainly was using APS-C back then (the A900 / A850 was three years old), and after re-establlishing its APS-C sensor production the Sony sensor division was very clear that they first would fulfill their external contracts (including Nikon as the main client), and then and only then give "surplus" sensors to Sony camera division (or whatever it was called then). It was a matter of honor to fulfill the external contracts first. Japanese ethos.

Of course another factor was that Nikon was designing (& producing?) those incredible far-UV lenses necessary for the Sony sensor production. No Nikon far UV lenses => no Sony sensors at all!!

At least for the D3x (24 MP FF) I know that the Sensor was the same as the one in the A900, even though Nikon time and again claimed the contrary (direct info from Sony). Sony kept quiet. Nikon had their own image processors in the D3x, of course, and images from A900 vs D3x were different, but the sensor was the same.

No information from my side on later developments, though!!

S


PostPosted: Sun Mar 12, 2023 6:31 pm    Post subject: Reply with quote
Sony produces the majority of camera sensors in the market, including for medium format cameras from Pentax, Fuji and Hasselblad. Nikon uses a mix of sensors, including Sony. I think only Canon exclusively produces their own sensors, at least for dslr/ mirrorless. Another fairly big manufacturer of sensors is Toshiba.

Of course it doesn't say too much when it comes to color accuracy.


PostPosted: Sun Mar 12, 2023 6:38 pm    Post subject: Reply with quote
caspert79 wrote:
Sony produces the majority of camera sensors in the market, including for medium format cameras from Pentax, Fuji and Hasselblad. Nikon uses a mix of sensors, including Sony. I think only Canon exclusively produces their own sensors, at least for dslr/ mirrorless. Another fairly big manufacturer of sensors is Toshiba.

Of course it doesn't say too much when it comes to color accuracy.


Indeed. Even if sensors are the same the color filter array can vary. You have Fuji with X-Trans. You have your traditional Bayer. Sony has some filter arrays that do not stick to the normal RGB but the greens are emerald and perhaps other variations. However, I don't think they use the emerald filter array in any of their mirrorless cameras.

https://en.wikipedia.org/wiki/Color_filter_array


PostPosted: Sun Mar 12, 2023 6:48 pm    Post subject: Reply with quote
Doc Sharptail wrote:

There is plenty of photographic evidence of Sony markings on nikon camera sensors.
They are made by Sony.

-D.S.


Not only that - but at least in the not-too-far past there was very close cooperation between Nikon and Sony related to the development of the machines producing the sensors (including far-UV special lenses with extremely high resolution, made by Nikon).

Around 2010 Nikon Japan and Zeiss Germany were the only two companies developing and producing these state-of-art lenses.

cbass wrote:
caspert79 wrote:
Sony produces the majority of camera sensors in the market, including for medium format cameras from Pentax, Fuji and Hasselblad. Nikon uses a mix of sensors, including Sony. I think only Canon exclusively produces their own sensors, at least for dslr/ mirrorless. Another fairly big manufacturer of sensors is Toshiba.

Of course it doesn't say too much when it comes to color accuracy.


Indeed. Even if sensors are the same the color filter array can vary. You have Fuji with X-Trans. You have your traditional Bayer. Sony has some filter arrays that do not stick to the normal RGB but the greens are emerald and perhaps other variations.


Back in 2008 when the A900 / D3x 24 MP FF Sony sensor was introduced, manufacturing it was a big problem. Originally only about 10% did meet the specifications (which were pretty loose themselves). Sensors from each wafer (about 30 cm diameter) had to measured, a correction profile for the respective wafer was developped. THen all sensors from that single wafer were calibrated using the correction profile. Quite some work. My first A900 was in the range of CHF 4500.-- ... and it had better colors than the second one I got a few months later.

S


PostPosted: Sun Mar 12, 2023 7:00 pm    Post subject: Reply with quote
stevemark wrote:


At least for the D3x (24 MP FF) I know that the Sensor was the same as the one in the A900, even though Nikon time and again claimed the contrary (direct info from Sony). Sony kept quiet. Nikon had their own image processors in the D3x, of course, and images from A900 vs D3x were different, but the sensor was the same.

No information from my side on later developments, though!!

S


Good to know from an inside source. I do believe that some of these companies would deny someone else made their sensor for reputation purposes.


PostPosted: Sun Mar 12, 2023 8:18 pm    Post subject: Reply with quote
cbass wrote:
stevemark wrote:


At least for the D3x (24 MP FF) I know that the Sensor was the same as the one in the A900, even though Nikon time and again claimed the contrary (direct info from Sony). Sony kept quiet. Nikon had their own image processors in the D3x, of course, and images from A900 vs D3x were different, but the sensor was the same.

No information from my side on later developments, though!!

S


Good to know from an inside source. I do believe that some of these companies would deny someone else made their sensor for reputation purposes.


Sometimes the wording is very carefully chosen in such statements and designed to be easily misinterpreted.

It is e.g. very easy to get confused when you read that back in the 50's/60's Minolta was one of only a couple of camera makers to manufacture their own glass (Minolta & Pentax (= Asahi)). This usually gets Nikon and Canon users a bit on the defence saying that Nikon and Canon did absolutely make their lenses in-house. But producing the lenses from that glass (pressing, grinding, polishing, coating) is not the same as actually making the raw glass (mixing the powdered constituents into a crucible, melting, mixing the melt, first annealing, fracturing & selecting or pressing blanks, and then further annealing). The way Minolta advertised this at the time made it seem at first glance that the likes of Nikon & Canon did not make their own lenses, but careful reading of what they really said makes it obvious they were referring to the manufacture of the raw glass only.


PostPosted: Sun Mar 12, 2023 8:22 pm    Post subject: Reply with quote
stevemark wrote:
RokkorDoctor wrote:

Stephan seems to have had more involvement with SONY developers here?;

My contacts were limited to the 2007-2011 time frame. The two guys managing photo stuff (M. Gessler) and his senior manager (O. Emmerich) were very easy going, and so were the guys from Japan (including the chief engineer guiding the DSLR team, and the chief engineer for camera lenses). Around 2012 Sony Germany took over the Swiss branch, and an Austrian (working in Vienna) was made head of Sony Switzerland, too. Not a good choice. The misconfigured and therefore unlucky A77 / A99 "mirrorless DLSRs" (remember tem??) were introduced, and I decided not to work with Sony any more. So did the Gessler / Emmerich duo, and others. At some point Japan intervened, and Emmerich now is back, and head of the entire Sony Switzerland.

RokkorDoctor wrote:

are the amplifier circuitry and D/A conversion in the SONY A7 line cameras done on-sensor, or in separate circuitry?


Probably. The 24 MP A900 sensor was the first to have the A/D conversion integrated on-sensor (6000 A/D converters, one for each row). In addition there's a 24dB amplifier on-sensor (pre A/D?? digital? I don't know), and before A/D conversion there's some analogue noise reduction (correlated double sampling - sensor is read twice, and noise is subtracted).

Read at page 20 here in my A900 book:
http://artaphot.ch/images/PDFs/AlphaSystemBuch_I/Sony_Alpha_Vollformat-System_00-23_72dpi.pdf

That said, I have no further knowledge on the sensors; by far most of discussions and testing back then was related to lenses (including pre-prod new stuff).

S


Like 1 small thanks Stephan, interesting read.

I have to assume that once they figured out how to embed the amplifiers & A/D converters on the sensor chip, that would be done in all future generations, if only for readout-speed considerations.


PostPosted: Sun Mar 12, 2023 10:16 pm    Post subject: Reply with quote
RokkorDoctor wrote:

Sometimes the wording is very carefully chosen in such statements and designed to be easily misinterpreted.

It is e.g. very easy to get confused when you read that back in the 50's/60's Minolta was one of only a couple of camera makers to manufacture their own glass (Minolta & Pentax (= Asahi)). This usually gets Nikon and Canon users a bit on the defence saying that Nikon and Canon did absolutely make their lenses in-house. But producing the lenses from that glass (pressing, grinding, polishing, coating) is not the same as actually making the raw glass (mixing the powdered constituents into a crucible, melting, mixing the melt, first annealing, fracturing & selecting or pressing blanks, and then further annealing). The way Minolta advertised this at the time made it seem at first glance that the likes of Nikon & Canon did not make their own lenses, but careful reading of what they really said makes it obvious they were referring to the manufacture of the raw glass only.


The (German language) text used in Swiss Minolta advertising wasn't ambiguous - it always was clear that they were talking about Minolta manufacturing their own glass ("Glas" in German, the raw material for producing lenses), and not about "glass" ("Objektive" German).

Usually the English language is more precise than the German, but in this case in German we have two clearly different words for

1) "lens" (german "Linse"); image from wikipedia


2) ... and "lens" (German "Objektiv")


Strange enough, the imprecise English terms now are being incorporated into the German language as well ... ("Altglas" for "vintage lenses" and "Linsen" instead of "Objektive").


PostPosted: Sun Mar 12, 2023 10:26 pm    Post subject: Reply with quote
cbass wrote:
Since this has migrated to another thread, I think it's best to provide some background. In another thread someone said the Fuji GFX and Sony a7R IV are virtually identical based on the comparison tools on dpreview. I looked at both looking at skin tones and said maybe resolution is similar but the skin tones I am seeing they are not virtually identical.





For your own comparison:

https://www.dpreview.com/reviews/fujifilm-gfx-50r-review/4

Color is a complex topic, but this is not the first time I have seen Sony body's turn people into Simpson's characters.

Was this an error due to lighting or WB on dpreview? Color cast on the lens used? Perhaps.
Is this something that only affects a certain body? Also possible.

Now I say color is complex because it is. Lenses affect colors. Color is not always consistent between camera bodies. Manufacturers apply their own internal color profiles and curves and they do not always match 100% between bodies. Processing from RAW? The software also has its own interpretation of color as well as standard curves it applies. Lightroom is different from Capture One which is different from Luminar and all are different from the in-camera processor. This completely ignores color differences due to WB algorithms and ambient light as well as color casts from artificial lights.


The portraits of the test scene at dpreview have deteriorated. Those are probably printed by a inkjet photo printer. It doesn't take very long to fade under the influence of light. The reds have gone. Red is always the first color to fade.


PostPosted: Sun Mar 12, 2023 10:54 pm    Post subject: Reply with quote
D1N0 wrote:

The portraits of the test scene at dpreview have deteriorated. Those are probably printed by a inkjet photo printer. It doesn't take very long to fade under the influence of light. The reds have gone. Red is always the first color to fade.


Would be a terrible mistake if true (not impossible though).

Can someone download and re-process the RAWs? My current software isn't suited for A7RIV Raws, I think ...

We really should figure that out.

S


PostPosted: Sun Mar 12, 2023 11:26 pm    Post subject: Reply with quote
D1N0 wrote:


The portraits of the test scene at dpreview have deteriorated. Those are probably printed by a inkjet photo printer. It doesn't take very long to fade under the influence of light. The reds have gone. Red is always the first color to fade.


Sadly, I don't think this is the case. This is not the first time I have seen a Sony go Simpsons. This guy's channel has nothing to do with cameras. Here is a test he did long ago to try to decide which camera to use for his channel.

https://youtu.be/C21w0Ogxh8U

Look at his skin tones with the Sony. Unless Josh faded also, but he was a live human being at the time of filming and not printed by an inkjet.
Post new topic   Reply to topic
Home   Manual Focus Lenses
Goto page 1, 2  Next
Page 1 of 2