X-Pro 1 reviewed at DPReview

[Keith]

One of the things that really disappointed me about the Xpro, and it's a very shallow attitude on my part, 😱 is it's looks!

The X100 is a very special looking little camera that a lot of people commented on when I had it ... the Xpro is a black lump!

 

[gavinlg]

Humm... where do you see that it matches the detail of a 25MP FF camera? Is there a good comparison review out there? (There may be, I don't spend a huge amount of time looking at reviews and probably wouldn't know about it.)

I do see that it doesn't have as much detail as a lot of other cameras in the DPReview shots, and they do indicate that "the X-Trans CMOS sensor includes a degree of chroma NR".

Thanks

My screen an eyes tell me that it resolves less details than a crop sensor 24MP NEX 7.

From the DPreview resolution testing:

"There are several points to be made here. Firstly, the X-Pro1 shows visibly higher resolution in this chart test than a conventional 16MP Bayer-type camera such as the Nikon D7000, or even the Sony NEX-5N (which has a particularly weak AA filter). In fact, in terms of resolution it's very close indeed to the 24MP Sony NEX-7 or the Sigma SD1, which uses a 15MP Foveon X3 sensor to record full colour information at every pixel, and therefore (like the X-Pro1) uses no AA filter."

 

[gdi]

From the DPreview resolution testing:

"There are several points to be made here. Firstly, the X-Pro1 shows visibly higher resolution in this chart test than a conventional 16MP Bayer-type camera such as the Nikon D7000, or even the Sony NEX-5N (which has a particularly weak AA filter). In fact, in terms of resolution it's very close indeed to the 24MP Sony NEX-7 or the Sigma SD1, which uses a 15MP Foveon X3 sensor to record full colour information at every pixel, and therefore (like the X-Pro1) uses no AA filter."

Thanks I didn't see that, I was looking at the actual test images, where that doesn't seem to hold, from a practical perspective. But, I don't think those comparison cameras are full frame are they?

I did look around some to see if I could find actual tests showing the fuji out-performing a new FF camera. I wasn't able to find such a test, but I did see an interesting article (most here are probably familiar with it) by the guy at ChromaSoft who wrote corner fix and PhotoRaw. He seems very knowledgeable ( not surprising) and has concluded that the camera really doesn't live up to the marketing...

"So my conclusion is, sorry to say, that the Fuji X-Pro1 X-Trans sensor doesn't deliver the Fuji promise of outperforming similarly sized sensors. In fact, it underperforms similar DX sensored cameras - with the official SILKPIX raw developer, the underperformance is too slight to be noticeable under normal circumstances, but is still there if you look closely."

Like I said, I am sure most people interested in the camera is aware of the article. The focus of the article is the difficulty of converting the X-trans files, and he seems to imply that when it is finally done properly the quality and resolution should improve. Of course the article is over a month old, so the situation may have changed and there is actually a good converter out there now. But, it does look like he is right that the results don't yet live up to promises.

I would still like to see a direct comparison with the competition when a top quality conversion is available; it could be interesting. In the meantime, I'll try out the OM-D for a while and see if the Fuji situation changes; it may be one of the situations where it pays to wait for the next generation.

 

[jsrockit]

The X100 is a very special looking little camera that a lot of people commented on when I had it ... the Xpro is a black lump!

And you think the D700 is a looker? 🙂

 

[gavinlg]

Thanks I didn't see that, I was looking at the actual test images, where that doesn't seem to hold, from a practical perspective. But, I don't think those comparison cameras are full frame are they?

I did look around some to see if I could find actual tests showing the fuji out-performing a new FF camera. I wasn't able to find such a test, but I did see an interesting article (most here are probably familiar with it) by the guy at ChromaSoft who wrote corner fix and PhotoRaw. He seems very knowledgeable ( not surprising) and has concluded that the camera really doesn't live up to the marketing...

"So my conclusion is, sorry to say, that the Fuji X-Pro1 X-Trans sensor doesn't deliver the Fuji promise of outperforming similarly sized sensors. In fact, it underperforms similar DX sensored cameras - with the official SILKPIX raw developer, the underperformance is too slight to be noticeable under normal circumstances, but is still there if you look closely."

Like I said, I am sure most people interested in the camera is aware of the article. The focus of the article is the difficulty of converting the X-trans files, and he seems to imply that when it is finally done properly the quality and resolution should improve. Of course the article is over a month old, so the situation may have changed and there is actually a good converter out there now. But, it does look like he is right that the results don't yet live up to promises.

I would still like to see a direct comparison with the competition when a top quality conversion is available; it could be interesting. In the meantime, I'll try out the OM-D for a while and see if the Fuji situation changes; it may be one of the situations where it pays to wait for the next generation.

Yeah it's the same with any kind of sensor - the algorithms get better over time. For instance my 5ds files are significantly better than they were 3-4 years ago because adobe has significantly improved their algorithms. When you consider the JPEGS are currently slightly better than/similar to the raw files converted by adobe raw in outright detail, you realize theres definitely a lot of room for improvement, as the camera is just converting the raw data itself - in camera.

Anyway, I don't find any of the review sites samples to be that great - maybe check out some of the out of cam JPEGS from the fujifilm website? (links to some below)

sample 1

sample 2

To me, there is nothing left wanting in the amount of detail in these files. It's completely natural, there's no m4/3s style halo sharpening (that I'm really sensitive to - other people don't seem to see it), and it's smooth and subtle. I would place it similar to a 5d2 - which is similar to the 5d3. That's just me personally.

 

[gdi]

Yeah it's the same with any kind of sensor - the algorithms get better over time. For instance my 5ds files are significantly better than they were 3-4 years ago because adobe has significantly improved their algorithms. When you consider the JPEGS are currently slightly better than/similar to the raw files converted by adobe raw in outright detail, you realize theres definitely a lot of room for improvement, as the camera is just converting the raw data itself - in camera.

I agree, that old files can be improved with improved software. In fact my GH2 shots' noise can be much better with LR 4 than before. I think my problem is that I continue to fall for the marketing and internet hype when an innovative camera comes out; I remember being excited to get a 1Ds (first version!) because it was better than 6x7. I should have learned by now, that improvements have proven to be evolutionary, not revolutionary! I agree that the XPro produces good files, but I can't help but think that all the buzz about it beating new FF cameras is over the top - at least until there are good comparisons to view, or I just buy one and see for myself (not beyond the realm of possibility).

To me, there is nothing left wanting in the amount of detail in these files. It's completely natural, there's no m4/3s style halo sharpening (that I'm really sensitive to - other people don't seem to see it), and it's smooth and subtle. I would place it similar to a 5d2 - which is similar to the 5d3. That's just me personally.

I know what you mean everyone, including me, has to decide for themselves and rationalize their decisions. The thing that I am most sensitive to about certain cameras, like the fuji, is when noise is reduced and takes detail along with it. I guess being a long time film shooter (and an M9 owner), I would rather have a little noise and more detail, than noiseless, too smooth images. But I can see how others would place a very high value on low noise, High ISO images. Those samples do look good.

Regarding the halos, I never shoot jpgs, so I control sharpening on my M4/3s camera, so if there are halos, it's my fault.

 

[semilog]

I...think that all the buzz about it beating new FF cameras is over the top - at least until there are good comparisons to view, or I just buy one and see for myself (not beyond the realm of possibility).

Of course that's over the top, and I wouldn't take anyone seriously who's making such a claim. What the X-Pro1 does do is more or less equal the FF cameras (D700, 5DI/II/III) in resolution, DR, and in low light performance, and it beats all but a few of the current APS-C sensors (the one in the NEX-7 being the obvious standard of comparison).

It does all this while costing a bit more than half of what a 5DIII does, in a smaller and lighter body with an outstanding and genuinely innovative viewfinder. That's pretty cool, I think. Bottom line is that for practical photography the 5D's, the NEX-7, the D7000, the X-Pro are all on par and have different strengths and weaknesses in operational terms, different quirks, take different lenses, etc.

Now, all that said, if you're interested in this stuff, Falk Lumo's latest white paper on what FF really means provides a great deal of food for thought. Lumo is a physicist who was first to report (correctly) that the Pentax K-x -- the first camera to use the EXMOR sensor later used in the D7000 -- had essentially equalled the D700 for image quality. He thinks that with the Nikon D600 we're about to see a major re-shuffling of the DSLR market. I bet he's right.

 

[semilog]

I agree, that old files can be improved with improved software. In fact my GH2 shots' noise can be much better with LR 4 than before.

Right, and consider that the GH2 uses a Beyer array -- at this point a highly mature technology -- while the X-Pro1 is using a brand new array. It's going to take time to really optimize the de-mosaicing.

A real concern that I have with the Fuji tech is that since everyone else uses Beyer, the software R&D effort is focused there. The Fuji array is a niche market for software developers and simply won't get as much attention. My hope is that development of good de-mosaicing software will be done as a consequence of it simply being an interesting technical problem, and thus a market-independent nerd-attractor. To some extent that seems to be happening.

Fuji could really speed this process if they would just release an SDK. That they have not indicates that they are in at least one crucial respect really missing the boat.

 

[gdi]

Of course that's over the top, and I wouldn't take anyone seriously who's making such a claim.

Well, at the time it was a very well respected, and of course hated, internet blogger, guru, expert. And then there was the claim that the M8 could produce large prints (20 X something) that matched 4x5. (I knew better in this one since I had both systems).

What the X-Pro1 does do is more or less equal the FF cameras (D700, 5DI/II/III) in resolution, DR, and in low light performance, and it beats all but a few of the current APS-C sensors (the one in the NEX-7 being the obvious standard of comparison).

That us just what I am looking for evidence of, rather than just the claims! If that is true then it should be easy to show a clear comparison. I may have missed it, but the ones I see indicate that is not the case. And at least one very reliable source claims that the opposite is true in practical terms. I am very open to seeing and acknowledging the real world advantages of the camera.

 

[semilog]

Put it this way. The X-Pro1 and NEX-7 sensors appear to have rather well-matched performance overall -- I think it safe to say that these are the current top-of-the heap in APS-C. And the quantitative tests indicate that the NEX-7 sensor is on par with (say) the D700 for DR, sensitivity, and resolution.

As Falk Lumo points out, now we have to deal with the implementation: the lenses used, the focus accuracy, etc. In other words, we have to deal with the throughput of the whole imaging chain and not just the sensor. At that level, smaller sensors thend to be harder to implement. For example, the position and alignment of sensor vs. lens is more critical.

In addition, really objective testing is, as Ctein has pointed out, really, really hard to do well. Even with a standard scene as DPR uses, it's very very very hard to equalize FOV, keep the lighting constant, deal with the varying curvature of field of the different lenses that must be used for different cameras (i.e., where do you put the focus point), etc.

What the results from several places -- DPR, Imaging Resource, etc. -- say is that in practical terms the best APS-C cameras (including but not limited to the X-Pro), when fitted with good lenses, can equal the current generation of FF cameras in most respects. And it is therefore foolish to choose among all of these cameras based (solely) on the sensor. Other factors are going to be more definitive.

The D800 may (for resolution) be half a generation ahead of the rest, just as the OM-D EM5 appears to be half a generation ahead of the other µ4/3 cameras at present.

My rule of thumb is that the sensor area and/or number of pixels has to change by a factor of two for there to be an interesting/useful difference in image quality (square root of 2 = 1.4x difference in linear print size; 1.2x differences in linear print size are too minor for me to care about).

Thus the difference between APS-C and m4/3, or the difference between APS-C and FF, is not interesting. But the difference between FF and µ4/3 (2x) may be apparent and useful, assuming rough parity in sensor technology.

 

[gdi]

Right, and consider that the GH2 uses a Beyer array -- at this point a highly mature technology -- while the X-Pro1 is using a brand new array. It's going to take time to really optimize the de-mosaicing.

A real concern that I have with the Fuji tech is that since everyone else uses Beyer, the software R&D effort is focused there. The Fuji array is a niche market for software developers and simply won't get as much attention. My hope is that development of good de-mosaicing software will be done as a consequence of it simply being an interesting technical problem, and thus a market-independent nerd-attractor. To some extent that seems to be happening.

Fuji could really speed this process if they would just release an SDK. That they have not indicates that they are in at least one crucial respect really missing the boat.

You make a good point about the maturity of mainstream sensors ( I think they are Bayer sensors, FWIW) and the focus of developers. It is something to consider. I am just surprised that Fuji hadn't figured it out prior to releasing the system. The work done by Sandy McGuffog was along the lines you anticipate - trying to rise to challenge. Maybe Fuji will prove more supportive as you say, an SDK with the right (efficient and accurate) methods could be released and we would see a lot of competion to get out products.

Put it this way. The X-Pro1 and NEX-7 sensors appear to have rather well-matched performance overall -- I think it safe to say that these are the current top-of-the heap in APS-C. And the quantitative tests indicate that the NEX-7 sensor is on par with (say) the D700 for DR, sensitivity, and resolution.

As Falk Lumo points out, now we have to deal with the implementation: the lenses used, the focus accuracy, etc. In other words, we have to deal with the throughput of the whole imaging chain and not just the sensor. At that level, smaller sensors thend to be harder to implement. For example, the position and alignment of sensor vs. lens is more critical.

In addition, really objective testing is, as Ctein has pointed out, really, really hard to do well. Even with a standard scene as DPR uses, it's very very very hard to equalize FOV, keep the lighting constant, deal with the varying curvature of field of the different lenses that must be used for different cameras (i.e., where do you put the focus point), etc.

What the results from several places -- DPR, Imaging Resource, etc. -- say is that in practical terms the best APS-C cameras (including but not limited to the X-Pro), when fitted with good lenses, can equal the current generation of FF cameras in most respects. And it is therefore foolish to choose among all of these cameras based (solely) on the sensor. Other factors are going to be more definitive.

The D800 may (for resolution) be half a generation ahead of the rest, just as the OM-D EM5 appears to be half a generation ahead of the other µ4/3 cameras at present.

My rule of thumb is that the sensor area and/or number of pixels has to change by a factor of two for there to be an interesting/useful difference in image quality (square root of 2 = 1.4x difference in linear print size; 1.2x differences in linear print size are too minor for me to care about).

Thus the difference between APS-C and m4/3, or the difference between APS-C and FF, is not interesting. But the difference between FF and µ4/3 (2x) may be apparent and useful, assuming rough parity in sensor technology.

I understand the claims involve the theoretical potential of the camera, and the transitive logic of comparison with similar cameras with proper converters. I don't discount that logic, but for taking photos today, it might prove frustrating.

The OM-D may be half a generation ahead of the leaders in M4/3s, but the more I compare, the less real advantages I see in IQ. I mentioned that I ordered one and it is on its way; I am now second guessing whether it is worthwhile as an upgrade to my GH2. And comparing the 2 to the Nex or Fuji, I see little advantage - so I agree that selecting the next size sensor is probably not worth the trouble and expense.

It will be interesting to check in in 6 months to see what the situation is with Fuji....

 

[semilog]

The OM-D may be half a generation ahead of the leaders in M4/3s, but the more I compare, the less real advantages I see in IQ.

If as reported it's a Sony sensor the differences will be most evident in terms of DR and especially DR at high ISO. At native ISO the 8 megapixel Kodak CCD in my old Oly E-500 still looks shockingly, gobsmackingly good for 4-generation-old tech... but push past ISO 400 and it just completely falls apart. It's at the edges of the performance envelope that the newer sensors really strut their stuff.

I like the X-Pro sensor, but I will admit that I am a Sony sensor partisan. This is based on over a decade of extremely positive experiences with Sony interline CCDs in scientific applications. CCDs that Sony started shipping over 12 years ago have only in the last year been eclipsed for price-performance by the very newest CMOS devices.

 

[Audii-Dudii]

The OM-D may be half a generation ahead of the leaders in M4/3s, but the more I compare, the less real advantages I see in IQ. I mentioned that I ordered one and it is on its way; I am now second guessing whether it is worthwhile as an upgrade to my GH2. And comparing the 2 to the Nex or Fuji, I see little advantage - so I agree that selecting the next size sensor is probably not worth the trouble and expense.

I can't speak to the GH2's IQ, but I have a GX1, and based upon the handful of side-by-side comparisons I've made between it and the X-Pro1, there does appear to be a slight, but distinct and noticeable, difference in IQ between them, with the advantage to Fuji (IMO, anyway.)

That said, my interest in the X100 and X-Pro1 has more to do with their traditonial-style form-factors than it does their IQ. For the same reason I made my initial foray into digital capture via the Panasonic DMC-LC1, Leica Digilux 2, and Panasonic DMC-L1, and more recently, chose to build my medium-format digital outfit around the Contax 645 body instead of any of the more modern, still in production alternatives, I find a significant degree of satisfaction in working with cameras that rely upon traditional external controls (shutter speed dial on the top plate, aperture ring on the lens, etc.) and to an extent, their IQ is slightly less important to me.

I like to think I'm not a complete Luddite, as I find that autofocus (implemented well, both in procedure and practice) has its place, and I actually prefer to focus and compose with an LCD versus a viewfinder (optical and electronic), but for the same reason I pointed out above, I skipped several generations of 35mm SLRs from the mid-'80s on because I simply didn't like how they felt in my hands or the non-intuitive (for me!) process of working with them.

Which brings me to my point: The slight IQ improvement that I see when comparing the files from my X-Pro1 to my GX1 is merely a lagniappe, as the reason I made the switch from one to the other is due, first and formost, to the X-Pro1's form-factor, not its performance. And in that respect, it hasn't disappointed me in the slightest! 🙂

 

[semilog]

^---- Lucid and sensible. Right on.

 

[GaryLH]

I can't speak to the GH2's IQ, but I have a GX1, and based upon the handful of side-by-side comparisons I've made between it and the X-Pro1, there does appear to be a slight, but distinct and noticeable, difference in IQ between them, with the advantage to Fuji (IMO, anyway.)

That said, my interest in the X100 and X-Pro1 has more to do with their traditonial-style form-factors than it does their IQ. For the same reason I made my initial foray into digital capture via the Panasonic DMC-LC1, Leica Digilux 2, and Panasonic DMC-L1, and more recently, chose to build my medium-format digital outfit around the Contax 645 body instead of any of the more modern, still in production alternatives, I find a significant degree of satisfaction in working with cameras that rely upon traditional external controls (shutter speed dial on the top plate, aperture ring on the lens, etc.) and to an extent, their IQ is slightly less important to me.

I like to think I'm not a complete Luddite, as I find that autofocus (implemented well, both in procedure and practice) has its place, and I actually prefer to focus and compose with an LCD versus a viewfinder (optical and electronic), but for the same reason I pointed out above, I skipped several generations of 35mm SLRs from the mid-'80s on because I simply didn't like how they felt in my hands or the non-intuitive (for me!) process of working with them.

Which brings me to my point: The slight IQ improvement that I see when comparing the files from my X-Pro1 to my GX1 is merely a lagniappe, as the reason I made the switch from one to the other is due, first and formost, to the X-Pro1's form-factor, not its performance. And in that respect, it hasn't disappointed me in the slightest! 🙂

I am basically in the same camp. Which sensor is better does not matter to me, I wanted a camera that was back to basics. The Fuji x 100 and xp1 cameras provided it.. On the plus side was the high iso performance.

I use the Ricoh gxr w/ m module for the same reason, back to basics. I still use it for all legacy lenses... I have an m adapter for the Fuji xp1 but overall the gxr still handles better. If Fuji brings in some variation of focus peaking, then my answer may change.

Gary

 

[gdi]

Going back to basics or a particular form factor, now those are reasons no one can argue with - personal choice!

 

[CK Dexter Haven]

I don't want to argue the point, but how is an XPro1 or X100 "back to basics?" Don't these cameras offer the same features as a D800 or 5DMkIII? But, more, since they have hybrid viewfinders?

I get that they look like 50 year old rangefinders in some sense, but isn't that just costuming? They're every bit as full-featured and complex as the top-spec dSLRs. Is it merely that they have aperture rings around the lenses, and shutter dials on the top deck? Is it really that simple?

 

[SausalitoDog]

CK -

You are more or less right. But the "costuming" is a large factor in the use. Size, shape, feel... All contribute to how one is used. These cameras won't replace DSLRs for many things (sports, wildlife, any fast moving subjects etc) but they are fantastic "walkaround" cameras for street photos and yet they still have some ability beyond the odd grab-shot.

Maybe the "back to basics" has more to do with HOW people are using them than what they are capable of.

Tom

 

[GaryLH]

I don't want to argue the point, but how is an XPro1 or X100 "back to basics?" Don't these cameras offer the same features as a D800 or 5DMkIII? But, more, since they have hybrid viewfinders?

I get that they look like 50 year old rangefinders in some sense, but isn't that just costuming? They're every bit as full-featured and complex as the top-spec dSLRs. Is it merely that they have aperture rings around the lenses, and shutter dials on the top deck? Is it really that simple?

We each have our own ideas of what basic is all about. I take pictures with folders as old as the 1930's as well.. And u can't get anymore basic then that🙂

But when it comes to digital cameras, at the end of the day, I don't want to remember which button is for what or which menu to use to get to your basic functions like shutter speed or aperture.

Weight, ovf, and form factor make up the other reasons.

The rd1 is as close the back to basics as u can get. But my eyesight is not as good as it used to be..

Today's dslr's are every bit as good as the Fuji cameras at a price points that are less.. Canon and Nikon have the advantage In terms of economy of scale that Fuji does not. Over time for reason most likely associated w/ cost and reliability the controls such as aperture and shutter speed have disappeared.

If u feel it is costuming, I think u maybe missing the point. Gone are the days of glancing down to double check shutter speed, aperture and exposure compensation before bringing the camera up to take the picture. And yes u can tip the camera over to check the lcd info in a lot of cameras. Even the digital ps originally had a ovf of some sort such as Canon G series.

My favorite 35mm film camera was really the Contax g2 not the Leica M and not the Nikon SLR, the xp1 so far comes the closest to the Contax of any digital camera....

Gary

PS. I forgot to mention that for me part of back to basics is making these controls that I use all the time available right at my finger tip. So yes as sausalito dog said it is really more about the way I use it and my expected handling characteristics...

 

[Audii-Dudii]

Is it merely that they have aperture rings around the lenses, and shutter dials on the top deck? Is it really that simple?

Yeah, believe it or not, but it's really that simple.

For me, the aperture and focus rings on the lens and the shutter-speed and exposure compensation dials on the top plate account for perhaps 90% of the rationale why I bought mine. (Mind you, this is because the high IQ was a given; if the sensor and processing pipeline were a complete turd, though, then I probably would have passed.)

I don't deny this is a completely irrational desire on my part, but the bottom line is that I want a camera that I can operate entirely by feel alone. I want a camera that instantly communicates to me how it's setup without my having to actually look at it. And I want to be able to adjust the aperture or shutter speed without ever looking at the camera, either. (I would also like to be able to focus it without ever looking at the LCD or through the viewfinder, but I guess that will have to wait for another day...)

I want it to be light enough so that I can carry it in my hand all day (I prefer wrist straps to neck straps) and small enough so that I can easily travel with it. I also don't need (and would prefer not to pay for) multiple focus points, fancy metering schemes, or the ability to capture video.

Lastly, because I'm a wide-angle kinda guy, I want a camera that offers (or will offer in short order) a selection of several prime lenses that are shorter than normal for the format.

So why didn't I buy an M9, you ask? Easy ... because I don't like the 3:2 format, I like to compose and focus using the LCD, and because I already have too much money invested in my Contax 645/P30+ outfit. For the money I saved buying an X-Pro1 instead of an M9, I'll live with the 3:2 format and the features I don't need (so long as they remain invisible to me). And if I ever drop it and break it or have it stolen from me while I'm photographing in a seedy part of town ... well, it will hurt a little bit, to be sure, but not enough to make me cry.

For all of the above reasons and more, this is why I bought my X-Pro1 and not a D800 or 5DIII. (And if you're curious to see what I've been doing with it, check out the last month or two worth of posts at my photoblog: http://audiidudii.aminus3.com/)