[EssJayEff]

DPReview TV mentioned the Pentax KF as one of two DSLR cameras (the other being the Canon T8i / 850D) in the budget category (under $1000 USD) . . .

Youtu.be

Youtu.be

.

---

Last edited by EssJayEff; 01-28-2023 at 10:47 AM. Reason: typo

[robgski]

That must have released this early, I am sure it was prepared in advance for publication on April Fools Day

Seriously, it's good to see Pentax get and honest and positive review.

[jgnfld]

Not even a negative or dismissive mention of the K3-iii or K1 series in the over $1K and under $2k department, though.

[biz-engineer]

Surprising he didn't mention the K1 II as landscape camera because it's the lowest priced "field camera" out there. I didn't get why micro 4/3 is even considered as landscape cameras, I would only consider u4/3 for the reach with long lenses, shooting wildlife hand held, but not for shooting landscapes due to the more limited DR of smaller sensors. But I understand that he's got to place Nikon and other popular brands as well.

[Grimmus]

but, but, but ...... there is another thread about how the Kf is the "worst camera of the year". Wish these me-too-bers would make up their minds ;-)

[luftfluss]

Surprising he didn't mention the K1 II as landscape camera because it's the lowest priced "field camera" out there. I didn't get why micro 4/3 is even considered as landscape cameras, I would only consider u4/3 for the reach with long lenses, shooting wildlife hand held, but not for shooting landscapes due to the more limited DR of smaller sensors. But I understand that he's got to place Nikon and other popular brands as well.

The Panasonic Lumix G9's m4/3 sensor has similar low-ISO DR as the Pentax KP and Nikon Z50.

Add to that the G9's superb pixel shift implementation, and it makes sense that it deserves serious consideration in the $1000 USD tier.

[biz-engineer]

Add to that the G9's superb pixel shift implementation

Pixel shift is useless for long exposure photography (as is firmware based ND simulation and pixel shift with any other brand).
The bottom line is, as soon as you get into the multi-exposure photography realm, any camera can produce amazing composite / super-resolution / panoramas, including 20 years old cameras and smartphones. The deal with cameras is really what you can get with one single exposure.
I agree that the G9 dynamic range performance looks impressive, on the test chart. I would consider test charts are a valuable reference, as an engineer, but I still can't explain why 645Z image look better and have move room for post processing, despite BSI full frame sensor test chart showing there should be no meaningful difference. Better would be to shoot the same scene at the same moment with both cameras and see what picture looks best in post processing.

---

Last edited by biz-engineer; 01-28-2023 at 11:36 PM.

[btnapa]

The bottom line is, as soon as you get into the multi-exposure photography realm, any camera can produce amazing composite / super-resolution / panoramas, including 20 years old cameras and smartphones.

Excellent point. A few years ago while attending WPPI in Las Vegas, I remember that Olympus had 30"x40" prints on the walls of their booth touting how good their M4/3 cameras were for landscape work. I never bought into that premise. I have printed my K1 images at 40"x60" size with tons of detail. I doubt any M4/3 sensor can keep up with that. I have also printed a good amount of trade show poster for my clients from my K3 camera. The trade show poster looked fantastic but to sell them as fine art prints the size limit is much smaller, perhaps a 20"x30"?

[luftfluss]

It's not that m4/3 is particularly great for landscape photography, it's that relative to the common 20-26MP APS-C sensors, the 20MP m4/3 sensor doesn't give up much.

If someone were looking to primarily shoot landscapes and spend $1000 USD on a camera, I'd suggest they look at a used full frame model, and then turn the discussion to lenses.

[3by2]

Excellent point. A few years ago while attending WPPI in Las Vegas, I remember that Olympus had 30"x40" prints on the walls of their booth touting how good their M4/3 cameras were for landscape work. I never bought into that premise. I have printed my K1 images at 40"x60" size with tons of detail. I doubt any M4/3 sensor can keep up with that. I have also printed a good amount of trade show poster for my clients from my K3 camera. The trade show poster looked fantastic but to sell them as fine art prints the size limit is much smaller, perhaps a 20"x30"?

Talking about detail is an interesting thing though and it's important to understand why you get more detail from larger sensors. One of the reasons is noise and noise is related to light gathering and sensor amplification. This is where full frame pulls ahead from m43 and apsc. At some point, below a certain ev, sensors have to boost their amplification and you get signal noise. Broadly speaking, noise destroys detail and it destroys tonality, you lose it first in the shadows, so I think below about ev5 a ff camera will always outperform m43 because it needs less amplification because it's gathering more light. Likewise if you need to boost the ISO beyond base, a larger sensor will do better. In other circumstances, where you don't need much amplification from base ISO, it's less certain, so I can quite believe the Olympus claim with a good lens. That's not to say I think m43 is the equal of the K1, I've seen enough K1 images to admire the tonality of that particular sensor and it's dynamic range but not all sensors are equal and a lot of circumstances are equal, so I would think M43 is quite capable of landscapes, limitations excepted and with good technique and post processing you can substantially close the gap.

[northcoastgreg]

The Panasonic Lumix G9's m4/3 sensor has similar low-ISO DR as the Pentax KP and Nikon Z50.

I guess it all depends on what is mean by the word "similar." But in terms of real world results, there's about a two stop difference between the KP and the G9 at base ISO. Now whether that gives a decisive advantage to the KP depends on the individual photographer and how much he/she wants to push shadows, etc. For those photographers who don't do much PP it may not matter. For those that do, it would matter.

I've shot landscapes with both m43 and APS-C Pentax. You can capture just as much, probably a little more, detail with an m43 system (assuming you're using top-of-the-line lenses) than with Pentax APS-C, but images taken with Pentax APS-C in my experience just look better. The detail in the Pentax images is rendered in a more natural way, the files are cleaner and have more headroom, and the colors are better.

[luftfluss]

I guess it all depends on what is mean by the word "similar." But in terms of real world results, there's about a two stop difference between the KP and the G9 at base ISO. Now whether that gives a decisive advantage to the KP depends on the individual photographer and how much he/she wants to push shadows, etc. For those photographers who don't do much PP it may not matter. For those that do, it would matter.

I've shot landscapes with both m43 and APS-C Pentax. You can capture just as much, probably a little more, detail with an m43 system (assuming you're using top-of-the-line lenses) than with Pentax APS-C, but images taken with Pentax APS-C in my experience just look better. The detail in the Pentax images is rendered in a more natural way, the files are cleaner and have more headroom, and the colors are better.

Looks to me the KP has about one stop on the G9, using a neutral sample. Here's an example where I gave the G9 a one stop advantage in comparison to the APS-C cameras:




The amount of noise looks about the same, but IMO the "grain" of the G9 looks better and less blotchy than the Pentaxes.

I do agree with you that the files from Pentax gear can have a better "look", particularly when using Pentax glass.

[3by2]

I guess it all depends on what is mean by the word "similar." But in terms of real world results, there's about a two stop difference between the KP and the G9 at base ISO. Now whether that gives a decisive advantage to the KP depends on the individual photographer and how much he/she wants to push shadows, etc. For those photographers who don't do much PP it may not matter. For those that do, it would matter.

I've shot landscapes with both m43 and APS-C Pentax. You can capture just as much, probably a little more, detail with an m43 system (assuming you're using top-of-the-line lenses) than with Pentax APS-C, but images taken with Pentax APS-C in my experience just look better. The detail in the Pentax images is rendered in a more natural way, the files are cleaner and have more headroom, and the colors are better.

On the headroom part, I noticed the same thing after moving from the KP but I also noticed that though the sliders in Lightroom, my preferred editor, seemed to give me more leeway, they also seemed to take me into more unacceptable extremes of compensation whereas my Olympus files allowed me much less leeway but at the same time they didn't allow me to move into the garish. I looked into this a little as I was trying to understand if in fact I'd chosen a system which was more limited in some way.


The more I looked, the less I think I have and that the difference is in the way the raw file is written and how it is mapped out to the tone curve from the original gamma exposure. In fact understanding how the camera exposes is critical and we can't really rely on the histogram information, which is based on a jpeg tone curve to get the exposure right, especially with highlights, there might in fact be a little more beyond what the camera says is blown, which is more important for M43 shooters as bringing up shadows is more likely to increase noise, so it's better to expose closer to the right, as close as you can get in fact.

There's a nice article which goes into some of it here (downloadable pdf)

Of course I haven't yet, fully explored testing the exposure but perhaps I should, in the meantime, I keep my highlights as close as possible to blown without blowing them.

I also looked at the 12 vs 14 bit raw files discussion, which struck me as the photography version of analogue vs digital in audiophile circles, it's all about what we can't see or hear!

Which is moving us away from the KF being a damn fine landscape camera. So to move us back, of course it is, it's a Pentax.

[northcoastgreg]

Looks to me the KP has about one stop on the G9, using a neutral sample.

I would suggest that a neutral sample gives misleading results, and here's why. Some camera companies attempt to game the high ISO scores (and DR scores) on their cameras through lowering saturation. In other words, the control noise they are willing to give up some color. In their DSLR days, Olympus was known for fabulous colors, and that trend continued with their 12 MP Pen cameras. The trouble was that those cameras produced noisy results, so when Olympus moved to the Sony 16 MP sensor, they gave up some of that color to control noise.

When I look at the samples over a dpreview.com, the superiority oif the KP's colors over the G9 is quite striking—so much so that I strongly suspect Panasonic is playing the same game as Olympus. Now curiously enough with OM-D Digital Solutions introduced the OM-1, they claimed the new camera had a two stop high ISO advantage over previous, but in some image comparisons that I've seen, I'm not detecting much difference in noise or grain, but I am seeing a notable difference in color and the amount of detail captured. So when judging performance of a sensor, you can't just look at the noise on a "neutral" sample—color saturation is critical as well.

[luftfluss]

I would suggest that a neutral sample gives misleading results, and here's why. Some camera companies attempt to game the high ISO scores (and DR scores) on their cameras through lowering saturation. In other words, the control noise they are willing to give up some color. In their DSLR days, Olympus was known for fabulous colors, and that trend continued with their 12 MP Pen cameras. The trouble was that those cameras produced noisy results, so when Olympus moved to the Sony 16 MP sensor, they gave up some of that color to control noise.

When I look at the samples over a dpreview.com, the superiority oif the KP's colors over the G9 is quite striking—so much so that I strongly suspect Panasonic is playing the same game as Olympus. Now curiously enough with OM-D Digital Solutions introduced the OM-1, they claimed the new camera had a two stop high ISO advantage over previous, but in some image comparisons that I've seen, I'm not detecting much difference in noise or grain, but I am seeing a notable difference in color and the amount of detail captured. So when judging performance of a sensor, you can't just look at the noise on a "neutral" sample—color saturation is critical as well.

But if we give the KP a 2 stop advantage over the G9, the KP image - while retaining its color better - is noisier and has less detail:

Image comparison: Digital Photography Review






As far as the OM-1's "two stop high ISO advantage" claim, I think it turned out that was due to their new software's so-called AI noise reduction. A spurious claim by OMDS, IMO.