[QCdude]

Hi fellows,


I'm a kind impressed by this little camera performance. It is new on the market.
Here is a bunch of test shots I googled.

But the most interesting part of it is the sensor: "The 16.3-megapixel X-E2 is equipped with the X-Trans CMOS II sensor that Fujifilm unveiled earlier this year. Like the previous X-Trans CMOS, the new sensor doesn’t utilize an optical low pass (anti aliasing) filter (so that light can reach directly and effectively to the surface of sensor, which would maximize resolution and improve noise reduction) to enhance the picture quality." Ok, we've already heard this (K-5 IIs, D800E). BUT: "To compensate, a proprietary color filter array controls the moire and false color generation through highly randomized pixel arrangement".

"X-Trans CMOS II sensor incorporates an original color filter array with a highly random pattern, eliminating the need for an optical low-pass filter (OLPF). These filters are used in conventional systems to inhibit moiré at the expense of resolution. The X-Trans CMOS II sensor array lets the sensor capture unfiltered light from the lens, achieving an unprecedented level of resolution."

Nice technology, isn't it?

It is a kind of pricey model, and the lenses are $500+ . Well, the quality goes at the price.

I hope that such a sensor will become eventually available for Pentax too.

[jeffshaddix]

I have yet to find any moire in my K-5 iis photos.

I remember thinking the same thing when the first X-trans sensor came out, but I don't think it's as relevant as they make it out to be compared to the K-3 for instance.

A K-3f (foveon) however would give a much different IQ level (at it's own costs).

[QCdude]

I have yet to find any moire in my K-5 iis photos.

I remember thinking the same thing when the first X-trans sensor came out, but I don't think it's as relevant as they make it out to be compared to the K-3 for instance.

A K-3f (foveon) however would give a much different IQ level (at it's own costs).

To be honest I'm not very optimistic about 24Mp APS-C sensors yet, looks like they are exhibiting higher levels of noise, etc. Well, this is pure physics: smaller and denser pixels thus generating more heat, etc. The latest K-3 image samples seem to proof it. It will take time while this technology get mature enough. Another problem that I would need to upgrade my PC to be able to process 24Mp RAWs - yet another $500-800 ... That why I pulled the trigger - the K-5 IIs last Black Friday at Amazon was at $799.

The 16Mp sensors are at their apogee today and Pentax has mastered to get the maximum out of it. Fujifilm has improved these sensors even more.

p.s. You can easily print a 5x3 ft poster out of 16Mp image. Do I need more? No, or at least I don't see any other reason yet.

[pinholecam]

Personally, I don't find that all the marketing speak on the X-trans translates to a better o/p at low iso levels.
Just look up comparators on DPR and IR to determine if its any better for yourself.
To me, the x-trans does have a bit of high iso advantage though.

The X series cameras so far have also failed to deliver on fast handling and AF (and I only mean decently fast).
I've handled the X-pro and XE1 a couple of times.
I also know a few folks who got the XE1 or X-pro and sold them off due to the slower AF and response.

What makes them attractive to me are the lenses.
Designed from new and ground up for aps-c and the camera flange distance.
Fast, not excessively big, rather cheap for what the speed, pretty sharp wide open and focal lengths for aps-c.
They are somewhat light too, but I'm not too sure that is a bad thing or good thing.

[kenafein]

Fuji has continuously improved the firmware on these cameras. The AF is much better, and the X-E2 is supposed to be a huge improvement with its hybrid focus. I know one review site re-did their Xpro1 review over a year later and said that the camera was transformed, and they bumped it up to their favorite mirrorless camera. Most of the lenses are quite good. It's a tempting system, for sure; but I want everything.

[Rondec]

I don't think the difference in the color array actually makes that much difference, except that it makes it a lot harder for folks like Adobe to support it in their software.

[QCdude]

I was playing with X-e2 a bit in a store and liked it. That why I decided to do more research after. Btw, it is 100% 'made in Japan' also.
I felt like the focus was fast enough, silent and precise, I did it in low light also. And btw the video must be great within this technology too, there was no SD card in it, so I couldn't test the video. Now this is my #1 candidate if I decide to get a secondary camera. The zoom lenses exhibit quite a bit of distortion at the widest extremity (check PZ for X-e1 lenses), but if you shoot jpeg this being corrected by firmware on the fly. Raw image distortions can be corrected by converter - not a problem either.

[pinholecam]

Fuji has continuously improved the firmware on these cameras. The AF is much better, and the X-E2 is supposed to be a huge improvement with its hybrid focus. I know one review site re-did their Xpro1 review over a year later and said that the camera was transformed, and they bumped it up to their favorite mirrorless camera. Most of the lenses are quite good. It's a tempting system, for sure; but I want everything.

Not too sure about that.
I know a friend who bought one, sold it off because of the AF, then believed the firmware claims, bought one again,.... disappointment and sold if off again.

There is another friend who claims he was told by the Fuji pro at a road show that the focus was actually very fast.
'Just need to snap the shutter button and don't bother with the LCD and such', he was told, and according to him, he tried that and it worked.
Though he's a bit of a newbie and I'd take it with a dose of salt.
The first friend I mentioned is an experienced shooter though.

[lister6520]

I wonder whether when we'll see a stills camera with a trichroic beam splitter and 3 sensors as used in cinematic cameras. You get the advantages of the Foveon sensor and the Leica M Monochrom but without the former's loss of sensitivity and the latter's loss of colour.

While it would be expensive for sure it still wouldn't need to be nearly as expensive as the Leica.

[richmondthefish]

I hear the X-E1's with the firmware updates are pretty decent. You can get them easily for $500....i was thinking about one. I really want a RX1 but thats $$$$$$$$$$$$

[Byrd-2020]

To be honest I'm not very optimistic about 24Mp APS-C sensors yet, looks like they are exhibiting higher levels of noise, etc. Well, this is pure physics: smaller and denser pixels thus generating more heat, etc. The latest K-3 image samples seem to proof it. It will take time while this technology get mature enough. Another problem that I would need to upgrade my PC to be able to process 24Mp RAWs - yet another $500-800 ... That why I pulled the trigger - the K-5 IIs last Black Friday at Amazon was at $799.

The 16Mp sensors are at their apogee today and Pentax has mastered to get the maximum out of it. Fujifilm has improved these sensors even more.

p.s. You can easily print a 5x3 ft poster out of 16Mp image. Do I need more? No, or at least I don't see any other reason yet.

QCdude: First, congratulations on scoring a K-5iis at $799. (An incredible bargain!) Second, I think you are correct about the inherent noise problem that accompanies the pixel density of the 24mp APSC sensor (at least with the current state of sensor technology). Advances are being made in image processors that mitigates some of this. But I too have been less than impressed with the sample images coming out of the K-3. (My choice would still be for the K-5iis over the K-3.)

I

[vonBaloney]

To be honest I'm not very optimistic about 24Mp APS-C sensors yet, looks like they are exhibiting higher levels of noise, etc. Well, this is pure physics: smaller and denser pixels thus generating more heat, etc. The latest K-3 image samples seem to proof it.

From what I understand, and I read a technical article about this somewhere, the greater resolution MORE THAN compensates for the increased noise -- you won't see it (more noise vs the 16mp images) in your prints except at very large sizes -- sizes that you probably wouldn't want to make from a 16mp image. In other words, yes it has more noise that is visible if you pixel peep, but in practice you won't see it unless you are printing very big, so you don't pay the noise penalty until then. But you get increased detail, so that's the trade-off. Big prints w/ 24mp = more detail, more noise, big prints w/ 16mp = less detail, less noise. I'll take the former because we sometimes sell large prints, but if you never make big prints the difference in IQ between K-3 and K5iis is likely largely invisible.

[RobA_Oz]

While it seems to be true that there is an increase in noise from the K-5 to the K-3 sensor, having both cameras I'm not convinced it's significant for all but a minority of work, though that may be important in some situations.

The point about the absence of Moiré in K-3 photos (as a result of the higher resolution) is well made, but it's also worth noting that a number of members here now have both the K-5iis and the K-3, so it would be more instructive to have them comment about the noise and Moiré differences between those cameras (I have an original K-5). Nonetheless, many K-5iis user have reported few, if any, problems with Moiré here, so the comparison between this camera and the Fujifilm X-e2 is more valid than comparing the X-e2 with the K-3.

It's important to bear in mind the the X-trans sensor pattern doesn't eliminate Moiré, even though it does reduce the probability of it occurring, due to the reduced regularity of the colour cycling.

[Byrd-2020]

I don't think the difference in the color array actually makes that much difference, except that it makes it a lot harder for folks like Adobe to support it in their software.

As an owner of an Xpro-1 and X100s (I also owned the original X100), I would disagree. I have shot thousands of images with these cameras. I have also done comparisons to the images made by my K-7 and K-5 (both now gone). There is, I believe, a noticeable difference--especially at higer ISOs.

This difference, to my eye, manifests itself most noticeably in the noise pattern. The color array of the Xtrans sensor provides, so some have claimed, a pattern that more closely approximates that of film grain. I have also heard it said (though I don't know if it's true) that the engineers at Fujifilm actually made this one of their goals in the original design of the sensor.

[Rondec]

QCdude: First, congratulations on scoring a K-5iis at $799. (An incredible bargain!) Second, I think you are correct about the inherent noise problem that accompanies the pixel density of the 24mp APSC sensor (at least with the current state of sensor technology). Advances are being made in image processors that mitigates some of this. But I too have been less than impressed with the sample images coming out of the K-3. (My choice would still be for the K-5iis over the K-3.)

I

Noise on the K3 is equivalent to the K5/K5 II, not better or worse. DXO Mark gave the K3 a sports iso score of 1216 and the K5 II one of 1235. That's the same score. And if you look at the SNR curves, they lie right on top of each other. I own both cameras and it feels like that is right. I will shoot both up to iso 3200 standardly and 6400 in a pinch.