PDA

View Full Version : M8 and XP1 side-by-side (it's totally subjective, dude)

noimmunity
03-21-2012, 03:38
Here is a side-by-side comparison (http://www.flickr.com/photos/areality4all/sets/72157629268771810/with/7002497813/) of the XP1 and the M8.

Handheld. Auto WB. Manual Exposure. EXIF info is on the XP1 frames only, but the shutter speeds and aperture for the M8 frames are analogous, if not exactly the same.

M8 frames were PP'ed with RPP in K64 (with sharpening, saturation and local contrast added), then converted via LR3.
XP1 frames were in-camera jpegs "Velvia" mode, with additional sharpening added in post via Nik Raw Sharpener 3.0.

Not an "objective" test by any means, this comparison highlights subjective differences, limited of course by the very different processing to which the different cameras were subjected. Until there is LR support (not to mention RPP) for the XP1, this is the way it is going to be for now, for me.

http://farm8.staticflickr.com/7233/6856378000_6940bb8062_b.jpg (http://www.flickr.com/photos/areality4all/6856378000/)
Inside - M8 (http://www.flickr.com/photos/areality4all/6856378000/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr

http://farm8.staticflickr.com/7048/7002494839_7e1b677041_b.jpg (http://www.flickr.com/photos/areality4all/7002494839/)
Within - XP1 (http://www.flickr.com/photos/areality4all/7002494839/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr

http://farm8.staticflickr.com/7254/6856380538_b1cb41f655_b.jpg (http://www.flickr.com/photos/areality4all/6856380538/)
Brokedown Palace - M8 (http://www.flickr.com/photos/areality4all/6856380538/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr

http://farm8.staticflickr.com/7181/7002497225_bc253cabf1_b.jpg (http://www.flickr.com/photos/areality4all/7002497225/)
Brokedown Palace - XP1 (http://www.flickr.com/photos/areality4all/7002497225/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr

Lastly, at f/1.4:

http://farm8.staticflickr.com/7053/7002497813_8c0ffb4e08_b.jpg (http://www.flickr.com/photos/areality4all/7002497813/)
"Dualing" Ducatis - M8 (http://www.flickr.com/photos/areality4all/7002497813/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr

http://farm8.staticflickr.com/7278/7002498835_86dc2b47b0_b.jpg (http://www.flickr.com/photos/areality4all/7002498835/)
"Dualing" Ducatis - XP1 (http://www.flickr.com/photos/areality4all/7002498835/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr

http://farm8.staticflickr.com/7242/7002482979_2de34c4745_b.jpg (http://www.flickr.com/photos/areality4all/7002482979/) Saône Two - M8 (http://www.flickr.com/photos/areality4all/7002482979/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr

http://farm8.staticflickr.com/7237/6856369098_79183caa55_b.jpg (http://www.flickr.com/photos/areality4all/6856369098/) Saône Two - XP1 (http://www.flickr.com/photos/areality4all/6856369098/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr

Starting with the obvious, we can see the difference in FOV between the 1.33x crop sensor and the 1.5x one.

The next most obvious point is that processing makes all the difference. Unfortunately, it is still not possible for me to subject files from the two cameras to roughly the same process; I did not even try. Even though this side-by-side shootout tells me more about processing than anything, it is perhaps not without merit.

The M8 files look sharper, but the XP1 files have much higher resolution.

Both cameras render color in unique and appealing ways. Neither looks consistently and overwhelmingly "better". I agree that in general I prefer the M8 color (to be fair, the "Velvia" setting accentuates the reds), but the XP1 has a wider "band" or palette in some instances.

http://farm8.staticflickr.com/7202/7002486973_7f86abfa96_b.jpg (http://www.flickr.com/photos/areality4all/7002486973/) Saône - XP1 (http://www.flickr.com/photos/areality4all/7002486973/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr


http://farm8.staticflickr.com/7266/6856370086_a703ac5f5f_b.jpg (http://www.flickr.com/photos/areality4all/6856370086/) Saône - M8 (http://www.flickr.com/photos/areality4all/6856370086/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr

The band or range of the palette on the XP1 photo is impressive (look particularly at the vegetation on the escarpment in the background), wouldn't you agree?

DOF on the CV 35/1.2 is thinner at f/1.4 than it is on the Fuji XF 35/1.4, as expected due to the crop difference.

If this comparison does not meet your expectations or standards, I thank you for your indulgence.

What do y'all think?
randomm
03-21-2012, 03:44
Wonderful testing, thank you. Just a quick question: in the first pair, it looks like the cameras were not focused on the same thing (aside from very different DOF). If you look through the window on the door there is stuff still relatively sharp on the wall at the back of the room in the M8 version, when in the X-Pro1 version there isn't (but at the same time the x-pro1 version has the left side of the image very sharp and M8 doesn't). At what f stop were these shot at?
noimmunity
03-21-2012, 03:54
Wonderful testing, thank you. Just a quick question: in the first pair, it looks like the cameras were not focused on the same thing (aside from very different DOF). If you look through the window on the door there is stuff still relatively sharp on the wall at the back of the room in the M8 version, when in the X-Pro1 version there isn't (but at the same time the x-pro1 version has the left side of the image very sharp and M8 doesn't). At what f stop were these shot at?

They were both shot at f/1.4
Sharpening was added to both.
Hope this helps.
Keith
03-21-2012, 04:09
Interesting comparison ... the colours look a little more 'real' from the M8 IMO. Possibly just post processing considering your current limited options with the Xpro.
dexdog
03-21-2012, 04:18
Interesting comparison, thanks for posting. The M8 images look a lot more "film-like" than the XP1 (good K64 emulation), but also somewhat underexposed by half a stop or so. Not sure that I like the velvia mode, but the images are impressive in terms of resolution and tonal contrast. I would be very happy with results from either camera.
sevo
03-21-2012, 04:31
Both cameras render color in unique and appealing ways.

"Render" indeed, as that will be almost entirely a matter of the white balancing and internal post-processing. A raw against raw comparison with the same conversion software would be more revealing than a comparison of a quite undefined stack of softwares (LR3 might be Lightroom 3, but what the heck are RPP and K64?) against JPEGs processed with an internal "film" preset.
Rob-F
03-21-2012, 04:54
I like the warmer color balance of the XP1, but I'd like to see them compared at a more neutral color setting on the XP1 so we could see how its "native" color (if there is such a thing) looks without enhancement.
porktaco
03-21-2012, 05:14
wow. you are making my life a living hell, as i contemplate an x1
Benjamin Marks
03-21-2012, 05:41
It looks like the output of either camera could be tweaked to match the other in PS or LR. Bump up the exposure and saturation a smidge on the M8 files, tweak the saturation. What I am interested in seeing is comparisons between images from the same lens/aperture on both cameras (e.g. Leica lenses w/adapter . . .). But the M8 at base ISO is no slouch, and the X1 results are impressive.

Ben
jsrockit
03-21-2012, 05:43
The bottom line is that both appear to be great cameras.
Steve M.
03-21-2012, 06:22
They look like night and day to me. The film is so much better looking AND sharper (wonder why it looks sharper?). The motorcycle in the foreground of the film shot looks so 3D and sharp it seems to come out of the frame. I don't understand how anyone can see them as similar in any way, yet people seem to. Guess we all see differently. Thanks for posting this. Normally I would copy the images and go into PS to fix them the way I like them, then compare, but no need in this case. It's all in front of me.
dexdog
03-21-2012, 06:27
none of the images are on film, both M8 and XP1 are digital cameras.
David_Manning
03-21-2012, 06:27
Steve...um, there was no film.

Maybe I just didn't understand...
jsrockit
03-21-2012, 06:32
They look like night and day to me. The film is so much better looking AND sharper (wonder why it looks sharper?). .

Where's the film? Placebo effect perhaps? ;)
noimmunity
03-21-2012, 06:44
The bottom line is that both appear to be great cameras.

At base ISO, yes.

In fact, the release of the D800E has essentially validated Leica's no-AA filter approach. The M8 should hold its own (at base ISO) for many more years to come, and may even be recognized some day as an extremely innovative camera.
noimmunity
03-21-2012, 06:49
It looks like the output of either camera could be tweaked to match the other in PS or LR. Bump up the exposure and saturation a smidge on the M8 files, tweak the saturation. What I am interested in seeing is comparisons between images from the same lens/aperture on both cameras (e.g. Leica lenses w/adapter . . .). But the M8 at base ISO is no slouch, and the X1 results are impressive.

Ben

The funny thing is, a high amount of saturation level was added to the M8 files in processing in RPP (Raw Photo Processor (http://www.raw-photo-processor.com/RPP/Overview.html)).

I am breaking out the Silypix RAW converter that was included with the XP1. In the name of 'science'!

Seeing the results (I did not chimp) back home, I do think it would have been more interesting for comparison purposes to set the XP1 film simulation to "ProNegHi".
jsrockit
03-21-2012, 06:51
At base ISO, yes.

I happily used the M8 to 640 and sometimes beyond. Sure, the Fuji has better high ISO, but that's not nearly as important as many make it out to be. That said, I use my Fuji X100 in low light...so, the X-Pro1 will be used in a similar fashion with the M9 as my sunnyday camera.
noimmunity
03-21-2012, 07:04
I happily used the M8 to 640 and sometimes beyond. Sure, the Fuji has better high ISO, but that's not nearly as important as many make it out to be. That said, I use my Fuji X100 in low light...so, the X-Pro1 will be used in a similar fashion with the M9 as my sunnyday camera.

Yes, the M8 up to ISO640 is completely fine for me, but I was hitting up against that limit all the time after dark.

Worthy of note, too, is the fact that the XP1 with 35/1.4 weighs about 1/3 of the M8 with 35/1.2 .
sojournerphoto
03-21-2012, 07:19
The funny thing is, a high amount of saturation level was added to the M8 files in processing in RPP (Raw Photo Processor (http://www.raw-photo-processor.com/RPP/Overview.html)).

I am breaking out the Silypix RAW converter that was included with the XP1. In the name of 'science'!

Seeing the results (I did not chimp) back home, I do think it would have been more interesting for comparison purposes to set the XP1 film simulation to "ProNegHi".


You could try the Provia setting. I've seen a picture from that that looked very nice.
BobYIL
03-21-2012, 07:21
Somehow the M8 samples look like having less DR than the XP1. I have a rather practical way of comparing DR, under the same lighting conditions the one with lower DR usually looks like being shot "at later hours" of the day or under a more cloudy sky, exhibiting less vividness, less seperation of hues of colors. Actually such details could be better answered by the poster as he only knows what happened during PP.
noimmunity
03-21-2012, 07:23
so, the X-Pro1 will be used in a similar fashion with the M9 as my sunnyday camera.

Awesome!!!

Sounds like a dream-team type set-up. I'll be interested to see how it works for you. In particular, I'm anticipating seamless change between the two bodies.
noimmunity
03-21-2012, 07:25
Somehow the M8 samples look like having less DR than the XP1. I have a rather practical way of comparing DR, under the same lighting conditions the one with lower DR usually looks like being shot "at later hours" of the day or under a more cloudy sky, exhibiting less vividness, less seperation of hues of colors. Actually such details could be better answered by the poster as he only knows what happened during PP.

Well, I confess I have the habit of slightly underexposing on digital, probably because it works better on the M8 that way and the M8 was my only digital until the XP1. Curiously, both cameras were exposed at the same f-stop and shutter speed (as far as possible), but for the M8 that results in underexposure. I'm not really sure why that happened.
jsrockit
03-21-2012, 08:03
Well, I confess I have the habit of slightly underexposing on digital, probably because it works better on the M8 that way .

Gotta preserve those highlights somehow right? ;)
noimmunity
03-21-2012, 08:10
In order to try to approximate the same processing for each image, I cracked out the Silkypix RAW converter that comes with the XP1 and pursued a convoluted series of steps to reproduce a process similar to that with which the M8 files were processed. (RAW processed through Silkypix into TIFF format, then exported into DNG format through LR3 and finally processed through Raw Photo Processor (to parallel the final processing for the M8) into TIFF and exported into JPEG through LR3. Phew!)

First, the M8 photo:

http://farm8.staticflickr.com/7254/6856380538_b1cb41f655_b.jpg (http://www.flickr.com/photos/areality4all/6856380538/)
Brokedown Palace - M8 (http://www.flickr.com/photos/areality4all/6856380538/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr

Now, several versions of the XP1 photo processed from RAW.

First, the one I like the best: it is processed with P160NC (portra) film curve in RPP, no saturation at all.

http://farm8.staticflickr.com/7138/7003209349_3e10ff8e7d_b.jpg (http://www.flickr.com/photos/areality4all/7003209349/)
XP1 RPP P160NC no saturation (http://www.flickr.com/photos/areality4all/7003209349/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr

Next comes the same but with saturation slider set to 20
http://farm7.staticflickr.com/6060/6857078930_63ae59826e_b.jpg (http://www.flickr.com/photos/areality4all/6857078930/)
XP1 RPP P160NC+saturation (http://www.flickr.com/photos/areality4all/6857078930/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr

And finally, processed with the same K64 (Kodachrome 64) film curve as the M8 file:
http://farm7.staticflickr.com/6060/6857078930_63ae59826e_b.jpg (http://www.flickr.com/photos/areality4all/6857078930/)
XP1 RPP P160NC+saturation (http://www.flickr.com/photos/areality4all/6857078930/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr

Just for convenience sake, here again is the OOC jpeg in "Velvia" mode from the X-P1
http://farm8.staticflickr.com/7181/7002497225_bc253cabf1_b.jpg (http://www.flickr.com/photos/areality4all/7002497225/)
Brokedown Palace - XP1 (http://www.flickr.com/photos/areality4all/7002497225/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr

I vastly prefer the RAW processed image to the OOC one.

Both the M8 and the XP1 were taken at 1/125 at f/1.4.

Now I have to look at these myself...
noimmunity
03-21-2012, 08:17
The X-Pro1 file processed in RPP according to the film curve of Kodak Portra 160 NC with no saturation is to my mind the best image.

When viewing the M8 image, the thinner DOF is much more obvious.

The M8 file was processed, remember, with a fairly high amount of color saturation.

The colors from the M8 are "true" but lack "depth", but the X-Pro1 files processed as such appear to have greater "clarity" and "depth" to my eyes.

Interesting.

I will process one more image in this way, then call it quits (gotta do some work, after all).
f16sunshine
03-21-2012, 08:30
Nice!

It's a fun comparison. I agree it's hard to say this is an objective comparison but the spirit in which you conduct it is subjective.
Both cameras/lenses do a very nice job.

Would it be too much to ask you to re-process the very last landscape through the RAW Processor?
Tim Gray
03-21-2012, 08:40
Can I ask how you are processing the XP1 images in RPP? I didn't think it supported that camera yet...
noimmunity
03-21-2012, 08:58
Andy, I hope I understood correctly to which landscape you were referring (pls tell me if it is not this frame).

First, the XP1 processed through RPP as Portra 160NC
http://farm8.staticflickr.com/7094/6857246064_658ca229e6_b.jpg (http://www.flickr.com/photos/areality4all/6857246064/)
XP1 RPP Portra 160 NC (http://www.flickr.com/photos/areality4all/6857246064/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr

Here is the M8 file processed through RPP with K64 and +30 saturation.
http://farm8.staticflickr.com/7266/6856370086_a703ac5f5f_b.jpg (http://www.flickr.com/photos/areality4all/6856370086/)
Saône - M8 (http://www.flickr.com/photos/areality4all/6856370086/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr

Finally, the OOC XP1 jpeg in "Velvia" mode
http://farm8.staticflickr.com/7202/7002486973_7f86abfa96_b.jpg (http://www.flickr.com/photos/areality4all/7002486973/)
Saône - XP1 (http://www.flickr.com/photos/areality4all/7002486973/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr

Both were shot at 1/1000 at f/2.8
noimmunity
03-21-2012, 09:01
Can I ask how you are processing the XP1 images in RPP? I didn't think it supported that camera yet...

Hey, Tim, it's a pretty tortuous process that involves conversion from Silkypix into TIFF format, then export from LR3 into DNG format, at which point the files become readable for RPP. My layman's understanding is that both TIFF and DNG being lossless formats, there shouldn't be any loss--other than in terms of time for the user-operator :p
Tim Gray
03-21-2012, 09:10
Makes sense. Just so you know, which I assume you do, the raw conversion has already taken place in that scenario. But it's an interesting way to get at the RPP tone curves. Thanks again for the comparisons.
noimmunity
03-21-2012, 09:10
It looks to me like there is some over-accentuation of the red channel happening on the XP1 TIFF files when they are imported out of LR3 into RPP.

It looks much better (color-wise) when processed directly out of LR3.

I will post in a sec...
noimmunity
03-21-2012, 09:11
Makes sense. Just so you know, which I assume you do, the raw conversion has already taken place in that scenario. But it's an interesting way to get at the RPP tone curves. Thanks again for the comparisons.

Yup, I realize that, Tim. The only reason I was going to all that trouble was for precisely the reason you cite--and to achieve some kind of parallel process vis-a-vis the M8 files.
f16sunshine
03-21-2012, 09:15
Yes that is the one. Thank you for taking the time.
noimmunity
03-21-2012, 09:19
You're welcome!

Here it is again, this time processed through LR3, much flatter response:
http://farm8.staticflickr.com/7112/7003422973_77c8854cae_b.jpg (http://www.flickr.com/photos/areality4all/7003422973/)
Processed through LR3 (http://www.flickr.com/photos/areality4all/7003422973/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr
f16sunshine
03-21-2012, 09:32
Comparing this image the M8 looks sharper by a tiny bit.
The Fuji image has a lot more life. Sort of glows. Mostly probably from the PP but still the impression is that it is very lively :D
I'm excited to see more from this camera!
bobbyrab
03-21-2012, 09:40
Sorry if I'm being an idiot here, but all your M8 files look very flat, I just opened the shot you ran through RPP K64 +30 saturation, and I gave it the simplest tweak in levels and it looks much closer to the fuji now.
tbarker13
03-21-2012, 09:42
They look like night and day to me. The film is so much better looking AND sharper (wonder why it looks sharper?). The motorcycle in the foreground of the film shot looks so 3D and sharp it seems to come out of the frame. I don't understand how anyone can see them as similar in any way, yet people seem to. Guess we all see differently.

I don't mean to pick on you Steve, but this is sort of priceless. Perhaps the differences between film and digital aren't as pronounced as some would like us to believe they are.:) But as you say, we all see things differently.
I'd add that sometimes we just see what we want to see.
bobbyrab
03-21-2012, 09:45
Sorry, I can't remember how to add the image without using flickr. Hope you don't mind

http://farm7.staticflickr.com/6118/7003492419_2b487ee330_b.jpg (http://www.flickr.com/photos/56793717@N07/7003492419/)
M8level (http://www.flickr.com/photos/56793717@N07/7003492419/) by fatbobbyrab (http://www.flickr.com/people/56793717@N07/), on Flickr
noimmunity
03-21-2012, 09:47
Sorry, I can't remember how to add the image without using flickr. Hope you don't mind


Hi, I was just going to suggest that you post it, thanks!

Nice!

BTW, I could supply the RAW files if anybody wants to play.

My thinking originally was to do for this comp as little PP as possible, but in the end, I think this shows above all that you just can't get away from PP.

I'm definitely gonna spend some time doing these over without RPP ;)
bobbyrab
03-21-2012, 09:54
Actually my effort's blacks are a little heavy, but that can be tweaked.
I think all these new sensors have more pop built in, I'm finding the same with my 5d mark III over the mark II, the files have better DR which results in a better contrast.
BobYIL
03-21-2012, 09:54
Sorry, I can't remember how to add the image without using flickr. Hope you don't mind

http://farm7.staticflickr.com/6118/7003492419_2b487ee330_b.jpg (http://www.flickr.com/photos/56793717@N07/7003492419/)
M8level (http://www.flickr.com/photos/56793717@N07/7003492419/) by fatbobbyrab (http://www.flickr.com/people/56793717@N07/), on Flickr

Hah! Now it looked like Kodachrome 64! Thank you...
noimmunity
03-21-2012, 10:58
Playing with the files a bit more after dinner...I notice that in patches of blue sky even at base ISO the XP1 files have a certain graininess. Not chroma noise, just graininess. Anyone know why that might be? On screen it looks kinda cool, but I have no idea how it would print.
Spyro
03-21-2012, 14:16
First, the XP1 processed through RPP as Portra 160NC
http://farm8.staticflickr.com/7094/6857246064_658ca229e6_b.jpg (http://www.flickr.com/photos/areality4all/6857246064/)
XP1 RPP Portra 160 NC (http://www.flickr.com/photos/areality4all/6857246064/) by areality4all (http://www.flickr.com/people/areality4all/), on Flickr


Oh wow, look what happened to the reds :eek:
noimmunity
03-21-2012, 22:58
Oh wow, look what happened to the reds :eek:

see post #31. it has to be a by-product of the tortuous (and wholly unnecessary) development protocol.

still hoping someone will pipe in with explanations for the attractive grain seen at base ISO on the XP1.
lynnb
03-21-2012, 23:14
Thanks for this thread. Even at web size the files look impressive. Now I'm getting GAS. Must... stop.... looking...
kdemas
03-22-2012, 01:48
I don't mean to pick on you Steve, but this is sort of priceless. Perhaps the differences between film and digital aren't as pronounced as some would like us to believe they are.:) But as you say, we all see things differently.
I'd add that sometimes we just see what we want to see.

OK, absolutely classic. In a lighthearted way of course, no piling on intended.
Jubb Jubb
03-22-2012, 02:48
Thanks for posting the comparison. I know it is completely subjective, but IMHO, I still prefer the shots from the M8 over the fuji. I don't know if the film profiles do the photos justice, however it is a clever idea from fuji to do that.

Thanks for sharing.
j j
03-22-2012, 02:51
still hoping someone will pipe in with explanations for the attractive grain seen at base ISO on the XP1.

Guess: is it the film simulation?
jsrockit
03-22-2012, 04:51
The fact that the Fuji holds its own (and even surpasses it in some regard) against a Leica (be it even an outdated one) speaks volumes about this APS-C CMOS camera. Should be fun to use.
paulfish4570
03-22-2012, 04:52
i'm thinking the fuji images all took a snappier edge to edge. colors? i have no idea which ones are more true. i have to wonder if the fuji lens is better edge to edge, front to back, than the cv ...