[normhead]

if you move the frame to the bottom right corner (with tube paints and brushes) and look at the object near the brushes (looks like a sponge? or some plant?) - d850 shows more detail there. I guess that's where the extra 9mp are going, or a sharper lens. K-1 looks better DR-wise, absolutely. With PS it's not even a contest. But it seems d850 (or the lens they've used) resolves more detail. It's more noticeable if you use the "Comparison" mode which downsamples the 45mp to 36mp.

I've quite often noticed differences in focus points before.I even found a part of those photos that showed a K-5 was sharper than a D800. Same settings but more DOF on the APS-c images. You really have to examine all of both images. And even then they are different copies of the same lens. There's no telling whether you have a good sample or a bad sample of any given lens. If the images are close. I think yo have to look at the best part of both images.

They are pretty much the same. You can only draw so much information from these images. That being said. You can tell when the systems are close, and if they are, there's no reason to favour one over the other, besides say ergonomics, low light performance, etc. etc. etc. These days you look at the IR images, then you decide based on other criteria. As mentioned, they are all good.

I certainly saw no noticeable advantage to a D850 image. 24 to 36 MP is a 50% increase and you should be able to see difference in some images, not in all. 36-42 is a .17 increase. That's hardly worth paying for. The 50% increase from a K-3 or D750 only prodiuces a 33% increase in resolution. That's just the law of diminishing returns at work. The point .17 increase in a D850 is probably less than a 10 percent increase in LW'PH. It's a very rare images here the K-1 won't resolve it, but the D850 will. It's a very rare image where a K-5 image shows any difference from a K-1 image when not shooting test subjects.

Much of what people believe about paying for more resolution or even better lenses simply isn't true. The first 4k is the low hanging fruit. After that you need a huge MP increase to show relatively minor improvement.

---

Last edited by normhead; 09-14-2017 at 04:58 PM.

[Rondec]

if you move the frame to the bottom right corner (with tube paints and brushes) and look at the object near the brushes (looks like a sponge? or some plant?) - d850 shows more detail there. I guess that's where the extra 9mp are going, or a sharper lens. K-1 looks better DR-wise, absolutely. With PS it's not even a contest. But it seems d850 (or the lens they've used) resolves more detail. It's more noticeable if you use the "Comparison" mode which downsamples the 45mp to 36mp.

I think these have more to do with the lens used. Corners feel more soft on the K-1 and there is less contrast in the K-1 image than in the D850 image. But I don't know that that is indicative of sensor performance because I think in the center I don't really see much difference.

[Jeff Lopez]

Im wondering if these tests comply with the EMVA1288 standard..

[awscreo]

I've quite often noticed differences in focus points before.I even found a part of those photos that showed a K-5 was sharper than a D800. Same settings but more DOF on the APS-c images. You really have to examine all of both images. And even then they are different copies of the same lens. There's no telling whether you have a good sample or a bad sample of any given lens. If the images are close. I think yo have to look at the best part of both images.

They are pretty much the same. You can only draw so much information from these images. That being said. You can tell when the systems are close, and if they are, there's no reason to favour one over the other, besides say ergonomics, low light performance, etc. etc. etc. These days you look at the IR images, then you decide based on other criteria. As mentioned, they are all good.

I certainly saw no noticeable advantage to a D850 image. 24 to 36 MP is a 50% increase and you should be able to see difference in some images, not in all. 36-42 is a .17 increase. That's hardly worth paying for. The 50% increase from a K-3 or D750 only prodiuces a 33% increase in resolution. That's just the law of diminishing returns at work. The point .17 increase in a D850 is probably less than a 10 percent increase in LW'PH. It's a very rare images here the K-1 won't resolve it, but the D850 will. It's a very rare image where a K-5 image shows any difference from a K-1 image when not shooting test subjects.

Much of what people believe about paying for more resolution or even better lenses simply isn't true. The first 4k is the low hanging fruit. After that you need a huge MP increase to show relatively minor improvement.

Isn't d850 45mp? 45.7 I think

Image looks sharper in other parts too, the hand drawn portrait, some other parts. But I agree it might be due to the lens. I'm not arguing that d850, just saying the title is a bit clock bait-ish)

[normhead]

Im wondering if these tests comply with the EMVA1288 standard..

Who cares, whatever you have, it's better than nothing. If that's your best data, then you have to go with it. What else do you have>

[BrianR]

"Pentax K-1 wiping the floor with Nikon D810, D850"

Seems like a strange choice. I mean they are weather sealed bodies, but I wouldn't expect them to be very absorbent. Maybe their straps are outsourced to Bounty?

[biz-engineer]

What does mopping the floor mean? Did you not want to say something like K1 knock the socks off of D850? Any other sound expression coming from across the Atlantic ocean?

[derekkite]

If that was the case, I'd be a egocentric incompetent looser.

Really good images are not even 0,1% caused by equipment. Either I can or I can't. No lame excuses there.

Depends what you shoot. I can't capture a bird flying towards me unless I do a pre focus point. Not complaining though, I get shots my friend with a D500 can't get due to low light conditions. The quick focus and frame rate of his camera gets shots that I couldn't get.

If Pentax released something like this I think I would have to buy it.

[Ian Stuart Forsyth]

Reading this article below, it seems that the current FF BSI sensor uses silicon with a lesser well depth than the 36mp sensor. Perhaps this explains the BSI sensor's struggle with brightness in high contrast scenes

I don't know with the new D850 sensor, just looking at the raw file histogram and the D810 raw histogram at iso 64 we don't see a shift from the normal headroom that Nikon uses (1-1.5 stops) from clipping. If we were to see less of FWC I would think we would see a shift with the increased size of the exposure that you normally get from iso 64 and lose that 1-1.5 stop headroom, this would show as a shift to the right in the raw histogram





And as I am writing this up I just noticed that the clipped data to the right is the same between both the D810 and the D850( Green 0.025%) further telling me that they should have the same FWC at iso 64

[biz-engineer]

Anyway, I don't get the image rendering I'd hope for from a camera such as D850. It renders more "digital" look than ever (same with D810 and K1 to a lesser extent), super optimized in a lab environment etc... and not so great for producing images that inspire emotions. It has super duper AF, FPS and ugly colors as no one seems to care about that except Fuji. For the price, they could at least make some efforts to get something else than photos that match an RGB color calibration profile.

[texandrews]

Anyway, I don't get the image rendering I'd hope for from a camera such as D850. It renders more "digital" look than ever (same with D810 and K1 to a lesser extent), super optimized in a lab environment etc... and not so great for producing images that inspire emotions.

Hmmm....I don't deny that many of these cameras have a clinical look, but I'm not having much difficulty with that in my artwork....

It has super duper AF, FPS and ugly colors as no one seems to care about that except Fuji. For the price, they could at least make some efforts to get something else than photos that match an RGB color calibration profile.

Are you talking about OOC jpg's? My experience with raw files is that they are so easy to manipulate in all ways today that I can get any look I want. Certainly we are not having problems with color at my museum....

[biz-engineer]

My experience with raw files is that they are so easy to manipulate in all ways today that I can get any look I want. Certainly we are not having problems with color at my museum....

Thanks for the comment. JPEG rendering is one thing that can be adjusted to taste, that requires additional effort for non Fuji x users, but eventually users can create their own presets, that's Ok. That said, sensor tech (CMOS vs CCD) and CFA seems to have an impact in how it is possible to render an image. Recent article about Phase One new IQ3 camera back shows just that. Camera/sensors are often measured against parameters such as dynamic range and snr, but how about overall image rendering? I don't see either K1 or D850 having made much progress in that regards (that was my point).

[Rondec]

Thanks for the comment. JPEG rendering is one thing that can be adjusted to taste, that requires additional effort for non Fuji x users, but eventually users can create their own presets, that's Ok. That said, sensor tech (CMOS vs CCD) and CFA seems to have an impact in how it is possible to render an image. Recent article about Phase One new IQ3 camera back shows just that. Camera/sensors are often measured against parameters such as dynamic range and snr, but how about overall image rendering? I don't see either K1 or D850 having made much progress in that regards (that was my point).

I'm not sure I follow you. Are you saying the K-1 didn't make progress over existing cameras like the D810? Because you are probably correct. I think they have a very similar sensor. As far as the D850 goes, I think its claim to fame has to do with its read out speed and the presence of 4K video. Image quality is pretty similar to previous cameras. If you don't need 9 fps or 4K video, then probably the D810 and K-1 are better/cheaper options.

I will say that when I can use it, pixel shift really does seem to bump image quality to another level, but obviously it is only usable in some situations.

[surfar]

As far as the D850 goes, I think its claim to fame has to do with its read out speed and the presence of 4K video. Image quality is pretty similar to previous cameras

If you want to see poor video performance from this D850,watch TheCameraStores YouTube review.....The AF of this Nikon is terrible!

[bossa]

In Australia the D850 will be about 2x the price of a K-1, which is utterly ridiculous. I am trying to sell a couple of D800E's (I'll keep my D810) but I can't budge them at A$2,000. In the DPR dynamic range images I felt that at the same ISO and EV vales the Pentax killed the D850, especially in pixelshift mode.

I'm looking to simplify my gear and have been thinking about getting a K-1, but I can't help thinking they'll release another model within the next year.