[house]

The only double standard I see is that they used work arounds for the d780 and not the k3iii. They should really have room in their widget for both electronical and mechanical shutter.

The K-3 III has no electronic shutter that's why they didn't use it. It's been mentioned several times in this thread already. If Pentax had enabled electronic shutter the images would be sharp.

It's very difficult to set up tests like this to be accurate. And as always accurate has to do with tolerances perfect doesn't exist. The studios cene widget is actually a bit to effective as it reveals things beyond the tolerances of the test.

[swanlefitte]

The K-3 III has no electronic shutter that's why they didn't use it.

You left that part of my quote off. Yes the work around is problematic. You seem to miss the point that I am saying I understand why they did it the way they did.

[beholder3]

Anyone else get the impression he's got a bloated ego and enjoys the power trip there?

He is a liar, as can be seen when comparing hies lies about the setup used versus what is officially described.

But that is in sync with certain editors there who are prone to outright lying. Not only once.

DPR has become a cesspool.

That is nothing new.

its irresponsible journalism

Calling that hater propaganda "journalism" is unfair to actual journalists.

You can see it in how they can't handle any valid criticism in their (often poor) performances. Not only do they get defensive, they censor any dissent away.

[Rico]

I've only downloaded the ISO 100 test shot thus far. Looking at it in Lightroom 6 (which uses ACR), the only changes I see from the default settings are custom white balance (only slightly different to the "as shot" in-camera values), +0.59 exposure, sharpening and noise reduction set to zero. What specific ACR settings have you found that are fraudulent?

What is there to be confused about BigMackCam? The fraud is that DPR is baking in ACR settings that remove detail from Pentax brand cameras while claiming Pentax cameras have baked in detail removing noise reduction. Get it? It's not what I am saying it is clearly what DPR's own DNG files are showing. The beauty of RAW files you can reset the file back to Camera Raw Default i.e. what the camera originally captured in the file. When you do that reset the file to Camera Raw Defaults with all the Pentax DNG files at DPR turning off the baked in Image Settings applied by DPR you find a whole lot of detail restored to the image. Like I said as far as I can tell it is only Pentax brand they apply special embedded ACR profiles that bakes in the removal of details.

The K-1 ISO 100 Pixel Shift images are a good example. The first is DPR's baked in ACR detail removal. The second image is the file set back to Camera RAW Default:





The K-1 Pixel Shift image of the DPR Comparison Scene is a good reference point to see what it really looks like as to how each manufacturer guesses to what it looks like. The scene of the people and the Fab Four patch demonstrate this well. The cross hatching detail in the scene of the people only really shows up using Pixel shift. The figure on the left edge with the white apron there is fine cross hatching present clearly visible in the K-1 Camera Raw Default image. Without Pixel Shift this fine detail does not show up. But you can see how DPR's baked in K-1 ACR Image Settings lift out all those details visible in the Camera RAW Default image. All I am doing is turning off the baked in settings DPR is applying in ACR which reveals all kinds of detail DPR removed.

DPR has been doing this to Pentax brand camera files for more than 5 years.

[beholder3]

The K-3 III has no electronic shutter that's why they didn't use it. It's been mentioned several times in this thread already. If Pentax had enabled electronic shutter the images would be sharp.

It still is 100% underperformance on dptabloid intern's side. They have admitted they know a way to create sharp images (extracts from a PS photo).
Their job is to deliver a comparable JPG or DNG which shows the qualities of the sensor and image pipeline (that is what the studio toll is there for - NOT any shake reduction capabilities or vibration effects of unknown origins).
They deliberately chose not to deliver what was possible even for their limited skills.

And that is only one out of many more dptabloid underperformances, the biggest one being that tons of other soccer moms obviously are more capable than their interns in producing tack sharp K-3 III images at the same shutter speeds.

I do find this compulsive defending of dptabloid like it was an untouchable personal god a little ouch to read. Influencers are just average Joes on the world wide stage.

[LensBeginner]

I've just today stumbled upon a paper on the same (?) issue on the K-7:
LumoLabs -- Shutter-induced blur for Pentax K-7
The Layman Summary reads as follows:
"It's not about vibration or a „loose“ image sensor and a massive tripod prevents it."

Of course there's no guarantee that the "issue" on the K-3III is the same & has the same solution.

[house]

What is there to be confused about BigMackCam? The fraud is that DPR is baking in ACR settings that remove detail from Pentax brand cameras while claiming Pentax cameras have baked in detail removing noise reduction. Get it? It's not what I am saying it is clearly what DPR's own DNG files are showing. The beauty of RAW files you can reset the file back to Camera Raw Default i.e. what the camera originally captured in the file. When you do that reset the file to Camera Raw Defaults with all the Pentax DNG files at DPR turning off the baked in Image Settings applied by DPR you find a whole lot of detail restored to the image. Like I said as far as I can tell it is only Pentax brand they apply special embedded ACR profiles that bakes in the removal of details.

The K-1 ISO 100 Pixel Shift images are a good example. The first is DPR's baked in ACR detail removal. The second image is the file set back to Camera RAW Default:




The K-1 Pixel Shift image of the DPR Comparison Scene is a good reference point to see what it really looks like as to how each manufacturer guesses to what it looks like. The scene of the people and the Fab Four patch demonstrate this well. The cross hatching detail in the scene of the people only really shows up using Pixel shift. The figure on the left edge with the white apron there is fine cross hatching present clearly visible in the K-1 Camera Raw Default image. Without Pixel Shift this fine detail does not show up. But you can see how DPR's baked in K-1 ACR Image Settings lift out all those details visible in the Camera RAW Default image. All I am doing is turning off the baked in settings DPR is applying in ACR which reveals all kinds of detail DPR removed.

DPR has been doing this to Pentax brand camera files for more than 5 years.

Rico your terminology is very unusual so it's difficult to understand what you are talking about. What does "baked" mean in terms of software settings?

They never state that they run the default camera parameters. I seem to remember they turn off certain corrections and set a standard sharpness (by the number in the acr interface). The default camera profile might have more agressive sharpening. If I'm correctly informed the default is actually different across brands.

Your scoop simply makes no sense to me and more over has nothing to do with the shutter shock detected in the K-3III at moderate shutter speeds.
The focus of this thread is flailing all over the place. People are bringing in all sorts of irrelevant and conspiratorial garbage to cover up the fact that everything points to dpreview doing nothing wrong and the model or camera having an issue with tripod mounted shutter shock under dpreview testing conditions. They even figured out the hack to shoot PS and extract one frame showing that they clearly went beyond what's necessary. I have lots of criticisms about dpreview and all ad or click based media but they've done and demonstrated very clearly what's going on.

Someone over at dprevew did tests and claimed no shutter shock but inspection showed it's clearly there, which was pointed out by a dpreview staff member. The test scene reveals very small issues and many Pentaxians are blind to obvious flaws. No wonder because it's splitting hairs and pictures will look good. When paying large sums of money however it makes sense to take advantage of the full capacity of the gear and it makes sense for dpreview to reveal all and every flaw how minute and irrelevant it may be.

---------- Post added 06-03-21 at 12:08 PM ----------

Their job is to deliver a comparable JPG or DNG which shows the qualities of the sensor and image pipeline (that is what the studio toll is there for - NOT any shake reduction capabilities or vibration effects of unknown origins).
They deliberately chose not to deliver what was possible even for their limited skills.

Why do you think you know what job they have? Personally it don't think it makes sense to alter the testing setup to cater for flaws in individual cameras. If you can't get sharp images using the camera under the same circumstances as other cameras it should be pointed out. If the camera has features that can conceivably be part of a normal workflow I'd say it's fine to use these (electronic shutter) as long as it's clearly pointed out in the review that the camera fails under normal circumstances. Going outside what can form a normal photography workflow would imho be very bad for a review site.

Note I'm not saying this is an issue for lots of people or that all cameras are affected (though there is some evidence now to suggest several are) but stop going crazy about dpreview testing. It's all done well and it reveals the tiniest flaws. There's still scope theyve made mistakes but nothing suggest they have at the moment.

---

Last edited by house; 06-03-2021 at 12:11 PM.

[stevebrot]

The Image Settings applied by DPR staff clearly lift out a whole layer of detail that is persevered in the Camera RAW Defaults. DPR purposefully made the file inferior to the Camera RAW Defaults and not by a little.

That is a pretty significant change.

Other brands DPR only outputs JPG's from each models Camera Raw Defaults.

Do you have proof that other brands were not similarly degraded; meaning proof in the form of XSD from the JPEG metadata or is this conclusion an assumption drawn from lack of data? Sorry to play the skeptic here, but it is my nature.


Steve

[BigMackCam]

What is there to be confused about BigMackCam? The fraud is that DPR is baking in ACR settings that remove detail from Pentax brand cameras while claiming Pentax cameras have baked in detail removing noise reduction. Get it? It's not what I am saying it is clearly what DPR's own DNG files are showing. The beauty of RAW files you can reset the file back to Camera Raw Default i.e. what the camera originally captured in the file. When you do that reset the file to Camera Raw Defaults with all the Pentax DNG files at DPR turning off the baked in Image Settings applied by DPR you find a whole lot of detail restored to the image. Like I said as far as I can tell it is only Pentax brand they apply special embedded ACR profiles that bakes in the removal of details.

The K-1 ISO 100 Pixel Shift images are a good example. The first is DPR's baked in ACR detail removal. The second image is the file set back to Camera RAW Default:

...

The K-1 Pixel Shift image of the DPR Comparison Scene is a good reference point to see what it really looks like as to how each manufacturer guesses to what it looks like. The scene of the people and the Fab Four patch demonstrate this well. The cross hatching detail in the scene of the people only really shows up using Pixel shift. The figure on the left edge with the white apron there is fine cross hatching present clearly visible in the K-1 Camera Raw Default image. Without Pixel Shift this fine detail does not show up. But you can see how DPR's baked in K-1 ACR Image Settings lift out all those details visible in the Camera RAW Default image. All I am doing is turning off the baked in settings DPR is applying in ACR which reveals all kinds of detail DPR removed.

DPR has been doing this to Pentax brand camera files for more than 5 years.

I don't think I'm confused (though it's certainly possible).

Just to humour me... Can you tell me precisely which settings are the ones you claim DPR has changed and "baked in", and how their edited values differ from the ACR defaults? The only changes I can detect are those I specified in my previous reply; white balance (fair enough), about half a stop EV boost, then sharpening and noise reduction set to zero. Aside from EV, all of these give the most honest representation of the raw files... sharpening and noise reduction would undoubtedly make them look better, and yes - these are usually applied automatically as ACR / Lightroom defaults; but they're not what comes straight from the camera.

The camera profile is correct ("Embedded"), the tone curve is flat... there are no other adjustments I can find. Only these I've mentioned.

I'm not trying to be difficult, truly I'm not - I'm prepared to accept what you're saying; I'd just like to know which settings were fraudulently baked in.

[robbiec]

What would we talk about if we didn't have DPR?

[Rico]

I don't think I'm confused (though it's certainly possible).

Just to humour me... Can you tell me precisely which settings are the ones you claim DPR has "baked in", and how their edited values differ from the ACR defaults? The only changes I can detect are those I specified in my previous reply; white balance (fair enough), about half a stop EV boost, then sharpening and noise reduction set to zero. Aside from EV, all of these give the most honest representation of the raw files (even if sharpening and noise reduction would undoubtedly make them look better).

I'm not trying to be difficult - I'm prepared to accept what you're saying; I'd just like to know which settings were fraudulently baked in.

The settings fraudulently baked in are those DPR is using to remove a layer of detail in ACR. I can not make it any clearer than that. I am not doing any tricks here all I am doing is opening DPR's files then turning off the baked in ACR settings they applied. Here are all the ISO 800 files from the D500 with Camera Raw Defaults, the K-3III baked ACR JPG they have for download, the Fuji X-T4 Camera Raw default then finally the K-3III DNG set back to Camera Raw Defaults which turns off the baked in DPR ACR settings:









DPR is putting JPG files for Pentax into their comparison tool they bake with ACR to remove detail.

[house]

Your comparison is lacking the D500 and XT-4 "Camera raw default"

Everyone knows the camera raw defaults will have more sharpening applied that the settings used for the studio scene. It's not fraudulent removal of a layer of detail but a disabling of sharpening, one done across cameras (as far as acr allows it to be done).

In fact ACR or any Adobe software should never be used for comparative reviews because you cannot ensure equal settings. Settings are applied that aren't visible in the ui.

[BigMackCam]

The settings fraudulently baked in are those DPR is using to remove a layer of detail in ACR. I can not make it any clearer than that. I am not doing any tricks here all I am doing is opening DPR's files then turning off the baked in ACR settings they applied. Here are all the ISO 800 files from the D500 with Camera Raw Defaults, the K-3III baked ACR JPG they have for download, the Fuji X-T4 Camera Raw default then finally the K-3III DNG set back to Camera Raw Defaults which turns off the baked in DPR ACR settings:
...
DPR is putting JPG files for Pentax into their comparison tool they bake with ACR to remove detail.

OK, I think I see the reason for our disconnect, and it may well be my fault. You're talking about JPEG files, whereas I'm not even looking at their JPEG files... I never bother looking at those in DPR's studio shot comparisons, as I'm simply not interested in JPEGs. I look only at the raw files to assess camera performance - and, in those, the only things that have been altered (appropriately so) are the settings I mentioned. The raw shots are an honest representation of what the camera produced. They look much better when I load them into RawTherapee, but that's because (a) the profiling is somewhat different, and (b) RT applies its "capture sharpening" tool by default. Switch the latter off, and the results are very, very similar to what I see in Lightroom.

I differ strongly with you on the claim of fraud, but I've said my piece and you're entitled to your opinion, of course.

I promised myself I wouldn't get involved in threads like these when I started posting again after my break, so I'll step out at this point...

[stevebrot]

I can not make it any clearer than that.

Are you aware that all of those files have intact XMP from the ACR processing and that the Fuji X-T4 was treated exactly the same as the K-3iii? The D500 version is a bit different in that the default sharpening was left in, however.

Added: I processed the DNG to TIFF using dcraw (no sharpening and only basic curves) and the results have the same softness as the images published by DPR.

I am with BigMacCam and am out of here...


Steve

[swanlefitte]

A comparison tool of unsharpened images of one camer to sharpened images of another is only good for misunderstanding both image quality and sharpening.