[TonyW]

OK, I give up, this will be the fourth time I've asked this question, and you keep dodging it.

Why do you think artifacts in 2D line drawings predict performance in real world 3D objects? Do you have any evidence to support this hypothesis.

The only dodging is from you. For the umpteenth time download the raws and see for yourself use any area of the image you want.

Forget the bloody artifacts you are using this to dodge the issues - the comparison of noise without PS and consequently the integrity of the video link. You confessed that you looked at JPEG only, and you are the one bleating about 2D line drawings and artefacts that is a total irrelevance.

Download the raw files examine them and report back

[normhead]

I just clicked on the link you provided and looked around. That's not enough?

Pretty demanding aren't we?

Here's the problem. You think this is worthwhile use of my time. My point is, if it isn't clearly obvious, its not real. When the D800 came out, I spend lots of time comparing K-5 and D800 images. What I found was that I could find areas where the D800 was better, I could find areas where the K-5 was better. Which was better depended on what part of the image you selected to prove your point. So I am really reluctant to start downloading and comparing whole files. Experience has taught me, it doesn't mean much. Variances in focus points creating slightly different alignment of the Depth of Field can make huge differences to how different areas of the photograph appear, especially in 3D set ups.

You may think I'm dodging, but in fact, I'm going on past experience. To conclude there is a real difference the you have to know about factors you can't possibly know looking at the images.

It is a constant source of disappointment to me to see people promoting new tech that is pretty much irrelevant to IQ. Promoting 45 k over 36 K, when 36 K over 24 k gives you very little is irrelevant. It's al amateur of perspective.

The differences between a D850 and a K-1 aren't enough to be worth a straight up trade. If someone offered me one, I'd take it sell the D850 and buy 2 K-1s. Let's keep these differences in perspective, in terms of what they are actually worth.

SO, from all the files I've examined at any size I'm likely to be viewing them, I don't see a discernible difference. I'd pixel peep for you, but past experience has taught me, that's pretty much meaningless.

You confessed that you looked at JPEG only,

There is nothing but jpegs available on Imagine Resources. I'm waiting for them to post more detailed info with raw data and MTF numbers from their tests. Again past experience has taught me that usually differences in detail of less than 10% are not visible in the final output unless you are printing upwards of perhaps 50 inches wide.I assume you might see difference there if you looked close, but it would not affect the artistic value of the image. I suspect that might be true, but i can't prove it. I would expect the D850, to be slightly sharper, but I also expect it will be meaningless to final output.

I can do what you do and see what you see, but that doesn't mean we would agree on anything afterwards. Your pro D850 bias is really apparent.

But just to avoid argument, maybe I should concede that the D850 is best for viewing images at sizes they can't be comfortably viewed at, because you can't see the whole frame, and for creating detail too fine to be observed at normal viewing distances. It's odd. I've noted many times before, on my wife's 4k monitor we can't tell the difference between a K-1 image and aK-5 image, but you want me to buy into there would be a difference between a K-1 and D850. That is going to be one tough sell.

So in essence the D850 is best for viewing images for which composition plays no part in the effectiveness of the image.Of you are close enough to see pixel differences, you are too close to appreciate the composition.

IMHO, you seem to be concerned with largely invisible minute differences that make zero difference to the art being created in the environment it's likely to be viewed. But that's always the way it is between me and you.

To me, a camera being better than another would mean I would enjoy this image more if were take with D850 as opposed to a K-1. I don't think it would. I don't care at all about the pixel peeping.

[TonyW]

I just clicked on the link you provided and looked around. That's not enough?

Pretty demanding aren't we?

Here's the problem. You think this is worthwhile use of my time. My point is, if it isn't clearly obvious, its not real. When the D800 came out, I spend lots of time comparing K-5 and D800 images. What I found was that I could find areas where the D800 was better, I could find areas where the K-5 was better. Which was better depended on what part of the image you selected to prove your point. So I am really reluctant to start downloading and comparing whole files. Experience has taught me, it doesn't mean much. Variances in focus points creating slightly different alignment of the Depth of Field can make huge differences to how different areas of the photograph appear, especially in 3D set ups.

You may think I'm dodging, but in fact, I'm going on past experience. To conclude there is a real difference the you have to know about factors you can't possibly know looking at the images.

It is a constant source of disappointment to me to see people promoting new tech that is pretty much irrelevant to IQ. Promoting 45 k over 36 K, when 36 K over 24 k gives you very little is irrelevant. It's al amateur of perspective.

The differences between a D850 and a K-1 aren't enough to be worth a straight up trade. If someone offered me one, I'd take it sell the D850 and buy 2 K-1s. Let's keep these differences in perspective, in terms of what they are actually worth.

SO, from all the files I've examined at any size I'm likely to be viewing them, I don't see a discernible difference. I'd pixel peep for you, but past experience has taught me, that's pretty much meaningless.



There is nothing but jpegs available on Imagine Resources. I'm waiting for them to post more detailed info with raw data and MTF numbers from their tests. Again past experience has taught me that usually differences in detail of less than 10% are not visible in the final output unless you are printing upwards of perhaps 50 inches wide.I assume you might see difference there if you looked close, but it would not affect the artistic value of the image. I suspect that might be true, but i can't prove it. I would expect the D850, to be slightly sharper, but I also expect it will be meaningless to final output.

I can do what you do and see what you see, but that doesn't mean we would agree on anything afterwards. Your pro D850 bias is really apparent.

But just to avoid argument, maybe I should concede that the D850 is best for viewing images at sizes they can't be comfortably viewed at, because you can't see the whole frame, and for creating detail too fine to be observed at normal viewing distances. It's odd. I've noted many times before, on my wife's 4k monitor we can't tell the difference between a K-1 image and aK-5 image, but you want me to buy into there would be a difference between a K-1 and D850. That is going to be one tough sell.

So in essence the D850 is best for viewing images for which composition plays no part in the effectiveness of the image.Of you are close enough to see pixel differences, you are too close to appreciate the composition.

IMHO, you seem to be concerned with largely invisible minute differences that make zero difference to the art being created in the environment it's likely to be viewed. But that's always the way it is between me and you.

To me, a camera being better than another would mean I would enjoy this image more if were take with D850 as opposed to a K-1. I don't think it would. I don't care at all about the pixel peeping.

I could choose to dissect your latest post here but tbh I am not inclined to at this stage as you are putting forward straw man arguments.

It is therefore IMHO that You have absolutely lost the plot here.
On one hand you say that you see no difference looking on a 4K monitor between k1 and k5 which kinda defeats the arguments you put forward on the other hand of superiority. So which is it difference or no difference?

If you care to actually read the thread you will see that I happily acknowledge the K1 quality with PS but question the validity of comparisons in standard resolution. There is nowhere in this thread where I have said the D850 superior to the K1

No one has stated that the differences worth trade up or down except you but that is just your view which professionals may disagree with for a number of reasons.

You may argue I have a bias towards D850 but that borne out of your ignorance of the facts posted. Your fanboy state clouding your judgement and comprehension which is IMHO

To double resolution requires a quadrupling of pixel count all other factors remaining the same so it is quite obvious that this cannot happen with these two cameras yet you state this in an earlier post comparing detail in the window area. Then you harp on about artefacts, an irrelevance to noise as shown here.

Your opinion about my concerns is unfounded and biased by your own preference and your nonsense about artistic value and art being created is just a red herring to divert from what in this thread is essentially a technical discussion mainly about noise an high ISO.

EDiT

To me, a camera being better than another would mean I would enjoy this image more if were take with D850 as opposed to a K-1. I don't think it would. I don't care at all about the pixel peeping

Why would you enjoy an image more wit a superior camera? And if you do not care about pixel peeping why are you arguing the toss here?

[normhead]

I guess when logic fails, slander.
Just how many ridiculous accusations can you level in one post? Looks like a lot.

Welcome to my ignore list.
Please do not read or respond to any of my future posts.

[TonyW]

Ok someone who seems intimidated by knowledge and turns round to be the hurt party after doing his own mud slinging. Happy to be on your ignore list and I will not reply unless you decide to put out poor information once again

[BigMackCam]

OK, folks, let's calm it down please. There's no reason to be anything other than friendly and respectful in our discussions, right? Otherwise, I have to do my moderating thing, and I'd rather not...

Thanks in advance

[audiobomber]

The video is grossly flawed, there's no way to conclude anything from it, except that he doesn't know what he's talking about. His initial claim that equalizing file sizes is cheating is absurd. Printing an image or displaying it at monitor resolution are the only two ways to correctly view a photo, and both resize the image. Also, the Pentax is clearly underexposed vs. the Nikon. A fair comparison would start with equal brightness. Finally, I would like to tie his hands to the chair. Flapping his arms around like that in front of the video camera is really annoying.

DXOMark rates the D850 sensor about 1/3 stop better overall than the K-1. The lower base ISO is a factor but DR is higher for the D850 throughout the higher ISO range. OTOH, the K-1 has slightly better noise performance. They scored the D850 higher in Color Depth, again due to the low base ISO on the Nikon. Tonal range slightly favours the K-1 above base ISO.
https://www.dxomark.com/Cameras/Compare/Side-by-side/Nikon-D850-versus-Pentax-K-1___1177_1075

The DPR comparator seems to shows a bit more chroma noise in the K-1 at ISO 25,600. Note that I am using the Compare viewing mode, which equalizes file size.

There's not enough in any of this to make me choose one over the other. Camera features, price and lenses dwarf any of the minor performance differences. Of course I would choose the K-1 if I were interested in FF, but hey, look at how well that KP does. Come on K-3 III.
Image comparison: Digital Photography Review

---

Last edited by audiobomber; 12-07-2017 at 08:42 PM.

[shardulm]

I actually re-watched the video as I find some comments appalling.
He is showing raw file comparisons, isn't he?. (I don't care about the post processed images at this point). RAW images is the starting point for post processing. Up until 8 mins he shows standard RAW image comparisons and K-1 shows better noise handling in those RAWs. Also the underexposed comparison shows better noise handling on K-1, standard RAW again. Isn't a good starter image has better post processing mileage or potential?.
Are we arguing that his presentation of these images is flawed and or inaccurate?
Don't mean to rekindle the thread but I thought standard RAWs were better on K-1 unless I am blind or my monitor is selectively adding noise to the D850 part of the images within the video.

[normhead]

Note that I am using the Compare viewing mode, which equalizes file size. There's not enough in any of this to make me choose one over the other.

Not the price?
I'm going to go with the video is appalling...it's just an enthusiastic guy losing it. He focussed on detail in the shadows when in many cases the K-1 was worse in 96% of the image. But, the pixel shift image was worth the price of admission.

In the IR jpegs at 3200 ISO the D850 image was better, but zoom out to look at the whole image there was no difference. Pixel level differences have to be really extreme to make a difference in prints or web images. The differences were not extreme.

Plus I had no idea there were folks around that enthusiastic about Pentax, (although anyone who bought a K-1 instead of a D850 should be.) That part I found amusing. I also found it amusing he trashed "the camera store" guys. IMHO those guys give Pentax more breaks than anyone else, well at least before this guy came along. If they said it's better that's fine with me. But that doesn't change that the K-1 is the best for me personally because of size constraints, economic constraints etc. etc. but if I had to characterize the guy who made the video, my word would be "goofball".

But I mean that in the nicest way.

[TonyW]

.....
Are we arguing that his presentation of these images is flawed and or inaccurate? ....

That is certainly my argument in light of those first few minutes and respect of noise in standard raw. I cannot repeat the increased noise shown in the video fir the Nikon files at the ISO he does. This is enough to cause concern about the value of the rest of his findings.

Rather than view a rendered sRGB jpeg which is the comparison site you may want to download real raw files at whatever ISO you choose. I am fairly convinced that you will not see what the video shows and more likely to see that there is nothing to choose between. As stated PS is a different case

[UncleVanya]

I actually re-watched the video as I find some comments appalling.
He is showing raw file comparisons, isn't he?. (I don't care about the post processed images at this point). RAW images is the starting point for post processing. Up until 8 mins he shows standard RAW image comparisons and K-1 shows better noise handling in those RAWs. Also the underexposed comparison shows better noise handling on K-1, standard RAW again. Isn't a good starter image has better post processing mileage or potential?.
Are we arguing that his presentation of these images is flawed and or inaccurate?
Don't mean to rekindle the thread but I thought standard RAWs were better on K-1 unless I am blind or my monitor is selectively adding noise to the D850 part of the images within the video.

I think if you go back and look at the K-3 vs. K5iis conversations resampling becomes a big mudpuddle that is hard to avoid getting yourself mired in. The question is do you simply compare RAW vs. RAW - or do you first equalize by either upsampling the lower resolution image to the same as the higher resolution or downsampling the higher resolution image to the lower resolution. I'm not going to advocate either approach.

I'm happy to have participated in this thread but it's mostly academic since I am not expecting to buy either anytime soon.

[Rondec]

I can't see a difference in images, either on imaging resource or DP Review, on standard D850 or K-1 files. DXO Mark put four points difference between the two cameras, basically saying that K-1 is 0.2 EVs worse at base iso with regard to dynamic range and that the sports iso score is a little better for the K-1 than the D850. Regardless, it is rough to stare at these blotchy shadowed areas and try to imagine that one is better than the other.

Clearly the pixel shift images are better from the K-1. I use pixel shift a lot, but certainly is it more limited in its usefulness.

[GMounk]

Not the price?

I also found it amusing he trashed "the camera store" guys. IMHO those guys give Pentax more breaks than anyone else, well at least before this guy came along. If they said it's better that's fine with me. .

I agree, the review by "The Camera Store" on the K-1 was good and IMO honest. I watched and read several reviews on the K-1 but it was " The Camera Store" review that convinced me it was the camera to buy.

I have noticed reviewers will sometimes isolate things in such a way so as to make a one camera appear to be much better than another. There were several factors regarding my decision to purchase a K-1.

I also think if people want to compare cameras then maybe they should compare cameras that sell for the same price. For me personally, I don't see the sense in comparing one camera with another camera that is worth significantly more. I feel the K-1 is the best full frame camera for the dollar. I don't think any other camera maker gives the same quality/value for the dollar. What other camera has five user settings?

[normhead]

I can't see a difference in images, either on imaging resource or DP Review, on standard D850 or K-1 files. DXO Mark put four points difference between the two cameras, basically saying that K-1 is 0.2 EVs worse at base iso with regard to dynamic range and that the sports iso score is a little better for the K-1 than the D850. Regardless, it is rough to stare at these blotchy shadowed areas and try to imagine that one is better than the other.

Clearly the pixel shift images are better from the K-1. I use pixel shift a lot, but certainly is it more limited in its usefulness.

There was discussion long ago, where we petty much came to the consensus that less than 5 points in a DxO rating was indistinguishable to the eye. More than 5 points you could see a difference, so your comment would fall in line with that.

The question is do you simply compare RAW vs. RAW - or do you first equalize by either upsampling the lower resolution image to the same as the higher resolution or downsampling the higher resolution image to the lower resolution. I'm not going to advocate either approach.

I'm not going to buy a camera that's 24 MP, so it's the same size and noise as the 16 MP version. When I use the K-3, I reduce the ISO so its as good as the K-5 at 800 ISO. I want the full 24 MP to be good. The thing about a k3 is you don't need noise reduction from 100-400 ISO and you often don't need it where I shoot at 640 ISO. That's what you pay for. If i could shoot 800 ISO and reduce the image to K-5 size to get the same noise... that would defeat the purpose of having a K-3. Just shoot with a K-5 or the K-1. You have to use a camera for it's strengths. If you don't get what you want at 800 ISO, shoot at a lower ISO. Don't shrink your image. If you print, that's important.

Pick up the camera that's good at what you want it for and go with it. Talking about what you have to do to make it like some other camera is insane.

[UncleVanya]

I'm not going to buy a camera that's 24 MP, so it's the same size and noise as the 16 MP version. When I use the K-3, I reduce the ISO so its as good as the K-5 at 800 ISO. I want the full 24 MP to be good. The thing about a k3 is you don't need noise reduction from 100-400 ISO and you often don't need it where I shoot at 640 ISO. That's what you pay for. If i could shoot 800 ISO and reduce the image to K-5 size to get the same noise... that would defeat the purpose of having a K-3. Just shoot with a K-5 or the K-1. You have to use a camera for it's strengths. If you don't get what you want at 800 ISO, shoot at a lower ISO. Don't shrink your image. If you print, that's important.

Pick up the camera that's good at what you want it for and go with it. Talking about what you have to do to make it like some other camera is insane.

I didn't advocate doing so. But what if I have only the one camera and want lower noise and I'm happy with 16mp? Then I can benefit from 24mp when I do have the light and downsample for better noise when I don't. There are use cases but for me I'm happy with the K-3 up to 3200 often depending on what I'm doing. We all have our limits...