[buster2015]

I guess Huawei is an off-brand. My wife last year bought one and that was the best mobile phone camera of 2018. It was just recently surpassed by a few other mobile phones but still sits close to the top. P20 Pro has a resolution of 10MP after pixel binning from a 40MP sensor. I happen to have a K200D with 10MP CCD sensor (from 2008). How well do they compare? Out of curiosity, I took a few photos in my yard. The view of P20 Pro is close to 28mm and thus I used a Tamron 17-50 at 17mm, closed down to f8, to match P20 Pro.

The first two photos show the scene: first is P20 Pro and second is K200D



You can see that the photo from K200D seems to have a better dynamic range. Then, I took crops of the upper left corner: K200D first and P20 Pro second.


To me, photo from P20 Pro is overly saturated and sharpened. But, to be fair it is not far behind a 10-yr old DSLR of the same pixel count.

I then took a photo of the same scene with my *Ist DL, a 6MP DSLR from 2005. For this comparison, I downsized P20 Pro's image to about 6MP level. Here are the crops: *Ist DL first and P20 Pro second.



In this comparison P20 Pro is doing reasonably well against ... a 14-yr old DSLR.

How about another phone? I took a photo of the same scene using my Mi Mix2, a phone from two years go. It is, well, clearly not comparable to either P20 Pro or the DSLRs. Too much noise reduction smeared the image quite a bit.


Message - I will still lug my DSLRs around when I hike.

---

Last edited by buster2015; 07-20-2019 at 04:26 PM.

[biz-engineer]

What were the depth of field (approx focus distance and apertures?) with the K200D and P20 Pro?

[BigMackCam]

Thanks, this is a fun comparison. Clearly, the P20 Pro is giving a very decent account of itself.

Re the *ist DL / P20 Pro comparison, I think with even basic raw development you'd be surprised at how good the *ist DL's files can be. Lens performance and AF focus accuracy play a part in these images, I believe, plus there's very little (if any) local contrast or sharpening adjustments to the DSLR's image. I frequently shoot a Samsung GX-10 (K10D) and GX-1L (*ist DL2), and whilst there is more fine detail - especially at long distances with foliage etc. - I can get remarkably similar results between the two with appropriate raw development adjustments.

Of course, the big wins for the DSLRs are versatility in lens focal lengths, control over subject to background/foreground separation, physical handling and control.

BUT... there's no denying that P20 Pro, for what it is, does a decent job

[biz-engineer]

BUT... there's no denying that P20 Pro, for what it is, does a decent job

I use my P20 for digital copies of documents, a good image is easier to get and transfer with the phone than it is with a K1.
With the K1, any slight tilt of the optical axis to paper plane, or warp, will induce blur of one of the sides of the digital copy.
I don't have to take care about flatness of the paper with the phone and connectivity is easier from the phone.
Now if I want subject separation, the DSLR does it naturally. The phone computed bokeh effect won't look like a natural 3D produced by DSLR.

[BigMackCam]

I use my P20 for digital copies of documents, a good image is easier to get and transfer with the phone than it is with a K1.
With the K1, any slight tilt of the optical axis to paper plane, or warp, will induce blur of one of the sides of the digital copy.
I don't have to take care about flatness of the paper with the phone and connectivity is easier from the phone.
Now if I want subject separation, the DSLR does it naturally. The phone computed bokeh effect won't look like a natural 3D produced by DSLR.

Right tool for the right job

When I want (or I'm happy to have) maximum depth of field, I love shooting my Q7. And I've been shooting a little Canon compact recently with a 1/2.33" sensor that gets pretty much everything in acceptable focus at the wide end. Smaller sensor cameras - including smartphones - definitely have valid applications where they present advantages over larger sensor DSLR / mirrorless cameras...

[buster2015]

What were the depth of field (approx focus distance and apertures?) with the K200D and P20 Pro?

For K200D, it was f8 and distance was about 10 ft. For P20 Pro, it was an auto photo and exif says it was f1.8.

---------- Post added 07-20-19 at 02:57 PM ----------

Thanks, this is a fun comparison. Clearly, the P20 Pro is giving a very decent account of itself.

Re the *ist DL / P20 Pro comparison, I think with even basic raw development you'd be surprised at how good the *ist DL's files can be. Lens performance and AF focus accuracy play a part in these images, I believe, plus there's very little (if any) local contrast or sharpening adjustments to the DSLR's image. I frequently shoot a Samsung GX-10 (K10D) and GX-1L (*ist DL2), and whilst there is more fine detail - especially at long distances with foliage etc. - I can get remarkably similar results between the two with appropriate raw development adjustments.

Of course, the big wins for the DSLRs are versatility in lens focal lengths, control over subject to background/foreground separation, physical handling and control.

BUT... there's no denying that P20 Pro, for what it is, does a decent job

Both DSLR photos were straight jpgs at default natural settings. Since K200D and *ist DL used the same lens, I was suspecting that *Ist DL was slightly out of focus. But, after some manual focusing it was still like that. I think (without proof of course) that it was due to the lower MP of the camera. Low-pass filter could play a role there, too.

[BigMackCam]

Both DSLR photos were straight jpgs at default natural settings. Since K200D and *ist DL used the same lens, I was suspecting that *Ist DL was slightly out of focus. But, after some manual focusing it was still like that. I think (without proof of course) that it was due to the lower MP of the camera. Low-pass filter could play a role there, too.

Understood However, without live view, manual focusing is unlikely to give a precise result, and certainly no more precise than AF, even unadjusted. Plus, I'd guess there are JPEG processing engine differences between the *ist DL and K200D, since the processing advanced as the model lines progressed.

Assuming two identical, well-exposed, well-focused images from both the K200D and *ist DL are down-sampled in resizing for display on a device that's considerably lower in resolution than the native photos - which would be the case if using a FHD monitor - then there should be little to distinguish between the two. I've determined AF adjustment settings for each of the lenses I use on both my GX-10 and GX-1L, and with those settings in place, my findings confirm this.

I'm not trying to pour water on your tests - far from it, I enjoyed them, and I think they show how useful something like the P20 Pro can be when it suits the use case. But my experience with the GX-1L (and my less-used *ist DL) is that with accurately adjusted AF, and at reasonable reproduction sizes, it's capable of very similar results to higher resolution cameras...

[Wheatfield]

Message - I will still lug my DSLRs around when I hike.

The real message is that this is a business where real estate matters.
Nothing else.

[Rondec]

The phone photo really shows what I haven't liked about them in the past. Greens in particular just don't have a natural hue, whereas the CCD sensor does a very nice job of rendering them very naturally. In addition, as you mentioned, dynamic range seems better on the K200 -- at least the shadow areas are more visible.

Clearly phone cameras have come a long way over time, though (not surprisingly) they haven't caught up with larger sensored cameras.

[BigMackCam]

A couple of responses to responses here that are a little "off piste" for me, since I'm generally highly critical of smartphone cameras ...

(Neither response is meant argumentatively, but in the spirit of constructive challenge and enquiry)

The real message is that this is a business where real estate matters.
Nothing else.

Digging into that, Bill, with respect... Are you saying that more sensor real-estate is always better in every use case, or that different sizes - larger and smaller - have different applications where either (with the associated lenses) can be better?

You know I shoot both full frame and APS-C, as well as smaller sensor cameras like the Q7. And I'm as big a fan of full frame as any other photographer... in the right circumstances. In a test chart scenario under various lighting levels, I'd agree that larger sensor cameras of sufficient resolution and of the same generation will out-perform smaller sensor cameras of sufficient resolution, especially as ISO rises. But in real shooting scenarios I'd posit that smaller sensor cameras, with lenses of equivalent field of view, can sometimes be better - depending, of course, on the very specific use case, target reproduction size and viewing distance.

Again, I'm not trying to be argumentative here... I just feel that your "real estate matters" statement, whilst undoubtedly accurate in many situations, deserves some qualification... especially for other members (since I'm pretty sure I know what you meant )...

The phone photo really shows what I haven't liked about them in the past. Greens in particular just don't have a natural hue, whereas the CCD sensor does a very nice job of rendering them very naturally. In addition, as you mentioned, dynamic range seems better on the K200 -- at least the shadow areas are more visible.

Clearly phone cameras have come a long way over time, though (not surprisingly) they haven't caught up with larger sensored cameras.

You probably know from various posts of mine, Vincent, that I'm a big fan of older CCD sensor cameras. But I do think it's possible to process modern CMOS sensor camera images so that they're very close (if not quite identical) to their CCD forebears. I can process my K-3 photos to get very similar results to those from my GX-10 / K10D, though I do think that the CCD sensor captures light differently (no proof there - just a sense). But where colour is concerned, a lot of modern cameras - smart-phones especially, I think - tend to give a rather cold, blue-biased rendering (much like modern computer monitors). When profiled with suitable tools, though, they're usually capable of much more accurate and/or pleasing rendering...

[Rondec]

A couple of responses to responses here that are a little "off piste" for me, since I'm generally highly critical of smartphone cameras ...

(Neither response is meant argumentatively, but in the spirit of constructive challenge and enquiry)



Digging into that, Bill, with respect... Are you saying that more sensor real-estate is always better in every use case, or that different sizes - larger and smaller - have different applications where either (with the associated lenses) can be better?

You know I shoot both full frame and APS-C, as well as smaller sensor cameras like the Q7. And I'm as big a fan of full frame as any other photographer... in the right circumstances. In a test chart scenario under various lighting levels, I'd agree that larger sensor cameras of sufficient resolution and of the same generation will out-perform smaller sensor cameras of sufficient resolution, especially as ISO rises. But in real shooting scenarios I'd posit that smaller sensor cameras, with lenses of equivalent field of view, can sometimes be better - depending, of course, on the target reproduction size and viewing distance.

Again, I'm not trying to be argumentative here... I just feel that your "real estate matters" statement, whilst undoubtedly accurate in many situations, deserves some qualification... especially for other members (since I'm pretty sure I know what you meant )...



You probably know from various posts of mine, Vincent, that I'm a big fan of older CCD sensor cameras. But I do think it's possible to process modern CMOS sensor camera images so that they're very close (if not quite identical) to their CCD forebears. I can process my K-3 photos to get very similar results to those from my GX-10 / K10D, though I do think that the CCD sensor captures light differently (no proof there - just a sense). But where colour is concerned, a lot of modern cameras - smart-phones especially, I think - tend to give a rather cold, blue-biased rendering (much like modern computer monitors). When profiled with suitable tools, though, they're usually capable of much more accurate and/or pleasing rendering...

Sure. I take decent photos with my cameras (all of which have CMOS sensors), but what do you think of the color of the grass in the phone photo?

[BigMackCam]

Sure. I take decent photos with my cameras (all of which have CMOS sensors), but what do you think of the color of the grass in the phone photo?

I could come up with a variety of descriptions The kindest would be "optimistic", but a more realistic one might be "garish" But it seems - to me, at least - to be aimed towards maximum visual impact, and that kind of ties in with the target audience for smart-phone cameras. With the P20 Pro, though, I believe it's possible to shoot raw files, in which case it's perfectly possible to create a better profile with a colour chart and the necessary software... and then, I'm not sure the colour differences alone would be much, if anything, of an issue...

[Kaisas]

P20 pro has an amazing quality of pictures, when you get the right spot and choose the right moment you can take nice pictures.

[BigMackCam]

P20 pro has an amazing quality of pictures, when you get the right spot and choose the right moment you can take nice pictures.

For sure, and that's a beautiful photograph. What it shows is the observational and compositional skills of the photographer. Excellent... really!

BUT... how does it stack up at larger reproduction sizes? What do 50% and 100% crops look like? They don't define whether this is a great photo or not, because it clearly is... but the limitations of the camera show through at larger reproductions - don't they?

[Kaisas]

I agree with you, if you %50 or more crop the quality will be very low, i mean from little sensor wht you expect? Mostly I print pictures from my p20 pro to instax mini printer, I'm so happy for the quality but if you print larger format I'm sure you can't be happy of the resolution.