Fujifilm interview: Secrets of apodization, 5-stop IS, OLED EVFs with 5-millisecond lag (really?) and more!
posted Friday, February 6, 2015 at 11:41 PM EST
All of our core readers will attest to the fact that we at IR love the various technologies lurking inside all those metal and carbon fiber shells that make up the cameras and lenses we use. And since that technology tends to get ever more complex with each passing year, we never grow tired of digging deeper in order to fulfill our (ahem, your) desire for more answers on how they make all this cool gear work.
Fortunately for all of us, the manufacturers of these marvels are generally happy to share a good bit of detail with us (without divulging trade secrets of course) about the nuts and bolts of how the newer and more complex technical aspects work, and we're more than happy to do the asking!
We sat down with Matt Schmitt and five Fujifilm engineers at Photokina 2014 back in September, and they talked us through many of the exciting concepts behind their latest product offerings. We're grateful to them for sharing their knowledge and insights with us! (Apologies to them and you for the delay in getting this out; we'll try to do better with the content of our upcoming interview with them at CP+!)
Imaging Resource interview with Fujifilm - Photokina 2014
William Brawley/Imaging Resource: The X30's a big step up!
Fuji PR: Yeah, it's beautiful, right?
WB: And the viewfinder...
Fuji: The X30 and also the X100T are the third-generation models of our X-series, and this is a very different concept. Especially on the X30, we changed the style itself. The X10 and X20 were very classical designs, but this is quite different, very modern and very stylish.
WB: Yeah, similar aesthetics to the X-T1...
Fuji: One of the biggest points for us with the X30 is the EVF, because we know many fans of the X10 and X20 would like to have an OVF again, but we changed our thinking on that because we developed a very large and clear EVF. If you look through this EVF, perhaps you can feel how large and very clear and high resolution, and also natural color it is.
Dave Etchells/Imaging Resource: Natural color?
Fuji: Yes, natural color. With this EVF, we're using color management technology.
DE: Ah, that's something new in an EVF, I think.
Fuji: Also, this EVF can display in two ways. One way is as the current EVFs do, it reacts to white balance, aperture or color mode, for instance, if you select Velvia color. Currently we display the Velvia color, for instance. But our new idea is to also provide a natural view mode. In this mode, it's more like an OVF. Very clear and no change of color for film simulation or...
DE: Ah, ok; so one mode shows you the film simulation, and the other just is natural color.
Fuji: Yes, that's the first important point I wanted to make about this camera, about the EVF. The second point is about all of the dials, on both the X30 and the X100T as well, we've improved all of dials and operation. For instance, this exposure dial, +/-3EV, instead of +/-2EV on the X20, and we also changed the back-side control to four buttons like the X-T1, instead of the wheel dial on the X20. We've also arranged the controls so the left side is for viewing images, and the right side is to control settings for shooting.
DE: OK, so all the controls for shooting are on the right side of the camera now, instead of having things like single/continuous drive mode and white balance on the left, like the X20.
Fuji: Yes, yes. And also the LCD itself is larger than the previous model, because the second-generation model's LCD was 2.8 inches, but now our third-generation model is 3.0 inches.
Fuji: So you can see how the X30 now, really, is more of a seamless fit with the X-series system that began back with the X100...
DE: Yeah, yeah.
Fuji: Because the X10 and X20 were really still more on the entry-level side. The X30 is now aiming for the real enthusiast photographers, or to be the second or third camera for a professional. And we're really trying to increase the image quality all around. Obviously the image quality, but also the design, and the build quality. Our feedback has been that the X100 and the subsequent X100S and X100T were so nice, we really wanted to apply that to a different price point. So that's why X30 has actually moved up a bit.
DE: I see. So it's really aiming at a different market now than what that part of the line had been.
Fuji: Yes. And the market's becoming really more sophisticated. You know, obviously if someone's shopping for a camera and they have a budget in mind, there's a lot of information out there, and we're aiming to be very competitive.
DE: The resolution spec for the EVF sounds similar to the X-T1, is that right?
Fuji: Yes, 2.36 million dots; also the same as the E2 and E1.
DE: The same as the E2 and E1. But it's different somehow?
Fuji: Yes, because we designed new glass for it, all the glass in this EVF is new.
DE: Oh, all the optics are new.
Fuji: Yes, brand new. And also we choose another Organic EL.
DE: Ah, the actual display element? I think one of the press materials referred to it as Organic EL. Is that different than Organic LED, or is that just another term for Organic LED?
Fuji: Yes, it's Organic LED.
DE: It's not just different optics, but you used a different OLED panel as well?
Fuji:
DE: Oh, so it's a new design for the Hybrid.
Fuji: Yes, a new design [for the X30].
And please see [the X100T]. Look at this EVF; you can see at the bottom of the right side there's a very small...
DE: Yeah, there's a little inset image there.
Fuji: Yes. And optical viewfinder with EVF.
DE: Ah - an actual OVF, but with a live EVF built in as well.
Fuji: Yes, and you can check the focusing and aperture using this small view...
DE: ...using the small EVF. Let me put it in manual focus mode...
Fuji: Yes, like this:
DE: Ah, I see. It's funny, when I first saw that illustration, I was thinking this was just showing an enlarged part of the image as an illustration, not what was actually on the display. I thought it was something that had been overlaid in Photoshop. But yeah, that actually is how it looks through the eyepiece.
Fuji: Yeah. And if the customer doesn't like this mode, it's easy to change with this control, and then you just get the usual OVF bright frame.
[Ed. Note: He's referring to the bright outline around the frame, often seen in rangefinder cameras. The little lever on the left front of the camera, as seen from the front, controls the EVF mode.]
DE: Oh, OK. So that's this lever here?
Fuji: So the photographer can choose from three modes, OVF, EVF, and OVF plus EVF.
DE:
Fuji: But the most important point that we would like to call attention to in the US market is film simulation. As you know, we already have ten film simulations, but Mr. Achida-san is a very, very important engineer for us, and he developed our new Classic Chrome film simulation. The reason we developed this new Classic Chrome mode is that we interviewed many, many photographers around the world and they requested a Classic Chrome mode, like a very old photo...
DE: Oh, so there were a lot of requests for the look of old positive films, slide films...
[Photographed for IR by Jeremy Gray with the X30 - B/W + Y film simulation mode]
Fuji: This color is very suited for documentary or street photography. Our photographer/advisers said that Fujifilm's color is very beautiful and very bright, but in some cases, like shooting documentary or street photos, they wanted a look that was very cool and little bit darker colored.
DE: Cool and a little bit dark...
Fuji:Pro Negative and Pro Negative Hi... these are negative.
DE: Those are negative films, I see.
Fuji: Yes, yes. For professional use.
Fuji: Classic Chrome doesn't simulate a film.
DE: Ah, I see, it's not a specific film.
Fuji: Yes, yes. We express the atmosphere of matte-printed photographs. Deep matte finished prints.
DE: Oh, so it's actually more the finished prints that you're simulating, I see.
[Ed. Note: This was different than my original understanding; I was thinking just in terms of reproducing the look of old slide film. I think they actually are talking about slide (reversal) film, but the results you'd get from it when printing on a matte paper versus viewing the transparency directly.]
Fuji: Yes, film simulation not only of the color, but also the atmosphere or feeling of the base paper.
DE: So it's a little less brighter, maybe, and deeper color.
Fuji: Yes, we studied very old magazines, which didn't have such vivid color, but very dark, cool and saturated color like this, and so we decided to develop this Classic Chrome. After we announced this new Classic Chrome film simulation, it was very well accepted by photographers but then they asked us "how about Neopan black and white color film simulation" or other films. They said "You have many, many film types, so please continue to pursue this idea!"
DE: Continue doing that, yeah.
Fuji: And so this new film simulation will be installed on the X30, X100T, and X-T1.
DE: Ah, so firmware updates for all of them.
Fuji: Yes. And this is the description of the coating of the X-T1 Graphite Silver...
DE: Oh, all the steps it goes through.
[Ed. Note: There was a long discussion here, with many references to the PowerPoint presentation, which I don't have to share with you. As I can recall, and tell from my notes, they start with a bare magnesium surface, then add an anti-oxidation coating, a primer coat, a layer of black paint, a partially-translucent graphite silver coating, and then a final clear-coat, to seal the surface. It turns out that the final surface layer also has very fine black particles in it, which gives an impression of greater depth. The net result is a very rich surface appearance, almost opalescent, but the reflective graphite layer is seen through the clear coat with the fine dark particles suspended within it.
The final step is to hand-paint the camera name and Fujifilm logo. It's a very long process, and has to be conducted in clean-room conditions, to avoid any dust specks attaching themselves to any of the multiple paint layers. They also said that the body parts being painted are rotated very slowly, while paint is sprayed on them in very thin layers, to produce a very uniform, smooth coating. They also talked some about the "luminescence" of the coating, by which I think they meant what I know of as opalescence - appearing differently, depending on how the light happens to strike it.]
Fuji: For the next point, we installed a new electronic shutter. This shutter is very high-speed, up to 1/32,000 second exposure times.
DE: 1/32,000, wow.
Fuji: Yes. And of course this shutter is an electronic shutter, so we can eliminate the shutter sound. And this is of course a compact system camera, so there's no vibration, no mirror shock.
DE: No mirror vibration, yes, so completely silent. Now is that high-speed shutter only in the new X-T1, or is that also firmware update on earlier versions?
Fuji: Yeah, of course, in December we update the firmware.
[Ed Note.: This new firmware has now been released.]
DE: Oh, OK, so in December there'll be a firmware update for the existing X-T1 model?
Fuji: For the X-T1 Black, yes.
Fuji: Yes. And we also included this new electronic shutter in the X100T.
DE: Ah, yeah - Wow, 1/32,000 second. That's pretty quick.
Fuji: Yes, yes. The reason why we developed this very fast electronic shutter is that our maximum lens apertures are very bright, the f/number is f/1.2 or f/1.4, so in very bright situations, customers couldn't shoot with a fully open aperture, because our shutter speed is limited...
DE: Right.
Fuji: And so we developed this electronic shutter.
DE: So now people can be in broad daylight, sunlight, and still shoot wide-open.
Fuji: Yes, yes. So our X-T1 customers can expand their field of shooting.
The Fuji X30 and its color-managed OLED EVF
DE: Moving to the X30, which we touched on in passing, one of the points I want to make sure to make is that -- even though it's the same resolution as the earlier X20 -- it's a completely new EVF, a new OLED panel. You mentioned that the EVF is now color-managed: I don't think I've seen that before. Have we seen that from any manufacturer previously, or on your previous models?
Fuji: Yes, for us, this natural-color viewing mode is the first time for us.
DE: You said there was color management, and my sense is that usually with an EVF, it's like... Well, we just feed it an RGB signal and whatever the viewfinder puts out, that's what color it is, but it sounds like here there's actually kind of a calibration process involved. I don't know if calibration is the word, but it maybe doesn't just take RGB from the sensor, but processes it so that the color is more accurate, or is it actually RGB from the sensor directly to the EVF?
Fuji: What I said, color management, means we adjust the tone, and the color matrix.
DE: Color matrix, yeah, so it is going through a color matrix to give more accurate color then.
Fuji: Yeah.
DE: So it's not... I said calibration; that wasn't quite what I was meaning, I was meaning more the color matrix. I mean, you're obviously not going in with a sensor and measuring the color of every OLED display panel, but you know what the OLED produces, and then you map it so that it's natural.
Fuji: Yes, that is correct. And not only color management, but also dynamic range.
DE: Dynamic range? Ah.
Fuji: Dynamic range, and especially in this new natural-color mode, contrast is automatically adjusted from 500 candela to 30 candela. 30 candela is like this dark scene...
DE: Very dark, yeah. And the 500 is very bright daylight. How does the maximum brightness compare with your previous EVF, do you know?
Fuji: 200 candela.
DE: Increasing to 500 candela from 200 is significant. That's very bright.
Fuji: And the 500 candela means almost the same level as the LCD of the iPhone.
[Ed. Note: An iPhone5 at maximum brightness is about 550 candela.]
DE: Right, you mean maximum brightness. That's always been an issue for me on some EVF cameras if I'm outside in very bright conditions...
Fuji: Yes.
DE: ...and if I bring the camera up to my eye, it's darker than what I've just been looking at, and my eye has to adapt. You also mentioned that that you've extended the dynamic range on the display by four times, 400%.
Fuji: Yes, yes. If a normal display is 100%, then this EVF is four times that.
DE: That's also extremely important. That's been one of the biggest limitations for me personally with EVFs, namely that, you know, I'm taking a picture and there's dark parts and light parts, and sometimes you there might be a sky with clouds that I want to frame around, but I can't see the clouds in the EVF.
Fuji: Yeah, yeah.
DE: Or I can't see detail in the shadows. I'm really eager to get my hands on this now, and play with it!
Fujinon XF50-140mm WR f/2.8 lens
Fuji: OK, now let's talk about the new lens, OK?
[Ed. Note: Referring to the newly-announced 50-140mm optic]
DE: I definitely want to talk about the lens. There's one particular question I have about viewfinder and viewfinder lag, and how it's measured, so I want to make sure we get that before the end, but OK, this is the new XF50-140mm WR.
Fuji: This one is the new lens: 50-140mm f/2.8 That's incredible, right?
Fuji: And this is a very advanced lens, so there are many new technologies that have been incorporated in it.
DE: Mmm. So there's five ED elements and, actually...
Fuji: Yes, five ED lens and one Super ED lens. The Super ED lens element is comparable to fluorite.
DE: Oh, it has fluorite-like characteristics, yeah.
Fuji: So thanks to these ED lenses, chromatic aberrations have been substantially reduced.
DE: And there are 23 elements in 16 groups...
All: Yes, yes.
Fuji: A very advanced configuration.
Fuji: And weather-resistant.
DE: And how many aspheric elements?
Fuji: No aspheric.
DE: No aspheric. Ah, that's right, I guess it's more for wide-angle that you need aspheric.
Fuji: Yes, yes. And this is the MTF chart. This performance is very good. So the high resolving power is comparable to a a prime lens.
DE: I can't wait for us to test that, then, too.
Fuji: This is very important.
Fuji: There is also our new lens coating technology, Nano GI coating. GI means Gradient Index.
DE: Ah, OK.
Fuji: Our newly-developed Nano GI coating technology reduces ghost and flare effectively in the case of diagonal incident rays, by gradually changing the refractive index from air to glass. Gradually changing by this structure.
DE: So instead of having multiple layers to try to cancel reflections, this way it gradually changes.
All: Yes.
DE: Now, there are different nano-coatings. This shows what looks like a 3D structure, you know, little peaks. Is that 3D, or is it just a continuous smooth layer that changes index?
Fuji: It's a 3D structure.
DE: OK. So it's more like it's a substance with a single refractive index but little, microscopic spikes, so that higher refractive index is gradually introduced into the light path.
Fuji: Yes.
DE: Ah, I see. Very good. [This was very interesting to me; previously I'd only been aware of nano coating approaches that used multiple layers to achieve a graded diffraction index.]
Fuji: The next feature we'd like to talk about is this lens's triple linear motor. There are three motors, at 120 degree intervals, a triple linear motor. This lets us realize very high speed and silent AF.
DE: So previously, have other lenses of yours had linear motors?
Fuji: Two linear motors.
DE: There were two on either side. And now with three, it basically can move more quickly...
Fuji: Yes, yes. The next feature is OIS (optical image stabilization). Five stop OIS is realized.
DE: Really!
Fuji: The key point is the combination of a high-performance gyro sensor with our original algorithm to cancel the drift element. Drift is like a kind of noise.
DE: Yeah, yeah. It's a slow variation, showing the lens continuing to rotate, even though it isn't.
Fuji: Yeah, yeah. So it is very important to extract only the camera shake, and eliminate the drift element.
DE: Yeah, that's the limitation on IS: When get to low shutter speeds, it's a very slow motion you're trying to correct for...
Fuji: Yes, yes.
DE: And so you're basically saying that what has limited that is not electrical noise as we'd normally think of it, you know, rapid noise, but more a slow drift from the gyro. That's interesting.
Fuji: The important thing is that thanks to this OIS system, our camera can shoot with hand. Not with tripod, but easily by hand.
DE: Yeah. Even at long telephoto. I'm interested in that point about drift. Then... It's not a matter of sensitivity, because those are also going to be very small movements at that point, or very slow movements. It's not so much that the sensitivity or the signal from the gyro was too small, it was just because it would shift over time even with no motion, so taking out the... I guess I'm probably saying the same thing, really. The sensor, the gyro, had enough noise margin to detect very small motion, but it would drift, and so by taking the drift out you can extend its useful range. I had thought that the limit was that the gyro wasn't sensitive enough to measure very small movement, but it's not that it's not sensitive, but rather that it would drift. But it's a new gyro sensor also?
Fuji: Yes, yes.
DE: It's a more sensitive gyro as well. Hmm.
Fuji:And this lens is approximately 60% of the weight of a full-size system, so reducing the burden on the photographer.
DE: Oh, and that's an interesting point there, that it's an internal focus, so the lens length is...
Fuji: It's constant, the length is constant.
DE: Yeah, there's not a lot of movement by heavy elements in the lens, so the balance is constant. That's interesting, too. Hmm. Boy, that feels great, doesn't it?
WB: That's a nice, hefty...
DE: Oh,
Fuji: And this lens is weather and dust-resistant and operates to a low temperature of -10°C (14°F).
DE: Colder than I want to be taking pictures!
Fujinon XF56mm f/1.2 APD lens
Fuji: The next topic we want to cover is the 56mm apodization lens. This is a derivative model of the current 56mm f/1.2. You can see the apodization filter...
[Ed. Note: In photography, apodization refers to introducing a soft-edged aperture into the optical path. Rather than the aperture going from transparent (the opening) to completely blocked (the iris leaves), an apodization filter gradually darkens the further you get from its center. This produces very soft-edged bokeh, unobtainable with a conventional aperture. As discussed below, though, the catch is that the soft-edged aperture interferes with phase-detect autofocus systems. Thus far, the only prior apodization lens we're aware of has been the manual-focus Sony 135mm f/2.8 STF SAL-135F28, originally developed by Minolta. It turns out that this lens actually uses two apertures at different positions within the optical path, rather than a gradated ND filter as with this new Fujifilm lens, but the principle is the same. The link above is a good review of this lens, and some explanation of how apodization works.]
DE: Ah, yeah, yeah. Look through the back, that would actually make a good picture. See how it's gradated? Yeah, the effect is like it's dark and comes in slowly.
WB: Yeah, it is. It's almost like a vignette.
DE: Exactly.
Fuji: The apodization filter is like this....
DE: Yeah, and I think there's only been one apodization lens previously. Sony had one -- or rather Minolta originally had one which Sony is continuing to make -- a 135mm, I think.
Fuji: Yes, but the Minolta and Sony lens is a manual-focus lens only.
DE: Only manual-focus, yeah.
Fuji: This is because an apodization filter is not usable for phase-detection AF.
DE: Ahhh! It doesn't work for phase-detection AF: That makes sense, the phase-detection signal would kind of fuzz out for rays coming from the edges of the lens...
Fuji: Yes! This is first autofocus lens with apodization. Contrast AF is more flexible.
DE: Ah. That's interesting, so...
Fuji: Our cameras are using both autofocus systems, both contrast and phase-detection. To focus this lens, we use only contrast AF.
DE: That's interesting.
Fuji: But Minolta or other companies previous cameras only had phase-detection.
DE: ...and so they couldn't autofocus with an apodization filter. That's interesting, so the soft edges means the phase-detection can't find the alignment of the split images, to tell where the subject is.
Fuji: This is a picture of bokeh, with apodization on and off.
DE: Wow. That's a big difference.
Fuji: A very beautiful bokeh effect.
[Images copyright Ivan Loh • Click here for Ivan's 56mm APD review]
DE: Yeah, very. And so there will still be places where there's a very large contrast difference, and you'll still see an edge there because the neutral-density gradient goes from zero to a density of two or four or something, but then if there's more contrast than that, you'll still see a sharp edge. Overall, though, it's much, much smoother, yeah.
Fuji: Yes, the light passing through the surroundings is reduced, so the gradiant bokeh is generated.
DE: Yeah, yeah.
Fuji: To produce this apodization filter is very difficult.
DE: Yeah, I guess thinking about it, trying to get that... it has to be a very precise, very uniform neutral density gradient. Because if it's uneven, you'll see it in the bokeh.
Fuji: Here is another comparison.
F4: This is a normal lens, and this is the apodization lens.
DE: Yeah, yeah. That's pretty significant!
WB: Wow.
DE: Look at all that little fine structure in there. Wow, that's really a significant difference.
WB: Yeah. And so this is being sold alongside the regular 56mm f/1.2, it's not replacing it?
Fuji: No, it doesn't replace that.
WB: OK, that's what I thought.
Fuji: They will coexist. And the effect of apodization filter is maximum at the wide-open aperture. Then it changes depending on aperture value.
DE: Oh, yeah. And actually, that was a question I asked you about, what the T-stop was. Because I knew the apodization and the T-stop, the transmission was going to be very different. So it's... there's actually a separate scale, it shows you what the actual is, and what the effective is. I hadn't realized that. Yeah, that's very good.
Fuji: So the inlaid letters show the effective number, and beyond f/5.6, apodization disappeared.
DE: Then it has no effect any more, yeah. Boy, really great-looking lenses, I can't wait to get our hands on 'em.
Fuji: Yeah, I mean it's really great that the weather-resistant family is growing, because that really brings to life what the X-T1 is all about.
DE: Yeah. So one question we had is something that we see as being kind of a problem in the industry, it's a lack of standards. So you showed the picture of the zoom lens with all the weather-sealing, and when Fujifilm says a camera is weather-sealed, we can pretty-well count on that it actually is weather-sealed, you know. It's gonna be splash resistant.
Fuji: Yeah, yeah.
DE: But some other manufacturers say "Oh, it's weather-resistant", but there's actually no seals. It must be frustrating. Is there any standard in the industry? Does CIPA have anything to say "If you want to say it's weather-sealed or weather-resistant, then it has to be able to stand up to at least this."? For machine tools and indicators, there's the IP [Ingress Protection] Code where there's a number and it means it can stand, you know, this much splash, or you know, have the, you know, a jet of water on it. But there's no specific standards for the camera industry.
[Ed. Note: The IP Code is sometimes applied to cameras, but typically only those claiming to be completely waterproof and submersible, rather than simply weather-resistant or splashproof.]
Fuji: Dave, I think part of the problem there is every industry has their own standard. Like the marine guys have their own sort of standards, because theirs is more about submerging and you know, so a nautical flashlight or something will have a rating as to in sea water, how long -- and sea water is different than fresh water.
DE: But in the camera industry, there's no standard.
Fuji: There's no standard.
DE: And so I won't name specific names, but there's, you know, one popular manufacturer that has some very high-quality cameras, and they say that they're, you know, weather-resistant or water-resistant, but they've had light leaks. And so I figure, well, if light can bounce around and get in there, then, you know, splash or water could certainly. And you know, someone opened them up and there's actually no seals any place, but they can still say that they're weather-resistant. It's kind of hard for us, because we have feature comparisons, and so one useful feature would be to be weather-resistant, but, you know, we have no idea what that means when someone says weather-resistant. I guess it's not really... This is, it's probably more of a comment than a question, but just, you know, somebody tell CIPA "Please make some kind of standard."
Fuji: Make some standard. But this lens will for the first time can stand -10 degree.
DE: -10C, yeah.
Fuji: Another brand, another maker, never shout like this.
DE: They never make a claim about...
Fuji: They claim only in zero...
DE: Oh, zero degrees C. Yeah, -10 is pretty cold.
Fuji: Because like our system, many photographer shoot at scene in very cold place like Canada or northern lights. And so we should...
DE: So you make a point of that, yeah.
Fuji: Yeah, yeah.
DE: I think the pictures that show the weather-seals on the bodies and things are very important, because manufacturers that say weather-resistant quote-unquote, but really aren't sealed per se, they don't have those pictures of sealing.
Fuji: Yeah. We go so far as to give basically an approximate number of weather seals, so we go just as far as we can just to really reinforce the fact that it is truly weather-sealed, and it's not just a marketing concept.
DE: Maybe that's one way we can differentiate, is if we've seen pictures of weather-sealing, that's when we say it's weather-sealed, in our feature comparison. And if we haven't seen pictures, we say, you know, not known.
Fuji: Or maybe use the number of seals, I mean...
Fuji: Yeah, you know, I think in the first weather-resistant lines we said more than 70, yeah. So you know, that becomes a way to articulate it.
DE: Yeah, maybe there's ways of grading. That's a good thought.
Fuji: In the case of this lens, we are sealing over 13 points.
DE: 13 points on the lens, just on the lens, wow.
Fuji: But then when you add the body, you know...
WB: This is a whole weather-sealed combo.
Fuji: Yeah. And add the vertical grip to it, that's weather-sealed also, which people don't realize that one.
DE: Oh, yeah.
Fuji: That the grip seals the body, and then it's sealed, so...
DE: Oh, so it's, yeah, you've got this complete sealed envelope when the grip is on. There's another area of specification that I think there's a lot of confusion -- at least I'm confused about what it really means -- and that's the viewfinder lag spec. So Fujifilm has had very, very fast viewfinder lag. In fact, the X-T1 was really kind of remarkable.
Fuji: 0.005 seconds. Yeah.
DE: Yeah. I'm trying to figure out what that means, though, because the simplest assumption would be that the viewfinder's refreshing 200 times a second, and so every five milliseconds it's repainted the screen, but I don't think the viewfinder paints that fast. Is the viewfinder somehow synchronized with the sensor so that at each point it's five milliseconds, or are you refreshing the sensor at one speed, and the EVF is different? Does it vary? Especially in low-light, the sensor's gonna need more time to accumulate light to get the image, but that means that you know, that the sensor maybe takes 1/15th of a second...
Fuji: So for the X-T1 it refreshes the image 54 times per second, right.
DE: That's like the viewfinder refresh rate.
Fuji: Yes, the refresh rate.
Fuji: In hertz.
DE: Yeah.
Fuji: And each image when it's appeared that you see, that is .005 seconds behind real time, you know.
Fuji: The real image.
Fuji: So every time it refreshes, you're actually seeing the image which is .005 seconds delay.
DE: So the sensor accumulates an image and then it transfers it. The refresh rate is 54 frames a second and so that means it starts up at the top, and then 1/54th of a second later it's at the bottom... So at the top, the image is acquired and now five milliseconds later the top line is painted. But then when it gets down to the bottom it's a 50th of a second later, so that's... I don't even know, whatever that is in milliseconds, that's going to be 20 milliseconds or something.
Fuji: So it's like scanning, it's always changing.
DE: Oh, OK, so the sensor basically is...
Fuji: ...going down.
DE: ...scanning at the same time...
Fuji: Yes. It's not like you're waiting...
DE: ...so it's synchronous.
Fuji: Yeah, it's not taking a single frame, and then it's rewritten back out.
DE: Yeah, and so as the line of the sensor's read out, then it's read onto... So basically the sensor and the EVF are scanning at the same speed.
Fuji: Yes, that's correct.
DE: Oh OK. So it goes together. OK. And at each point it's always only five milliseconds behind real time.
Fuji: And another important point is even though it is low-light scene, we somehow keep high rate, or...
DE: Oh, the 54 stays the same even though it's low-light.
Fuji: Yeah. But prior to the X-T1, it couldn't keep high frame rates in low light. For instance, the frame rate would change to 30 or 24... Yeah.
DE: Yeah, the refresh rate would decrease.
Fuji: And so not draw as smoothly in the viewfinder. So our EVF's point is very smooth and very... even though it is dark scene, it is very smooth. And also, natural and high-resolution.
Fuji: The X-E2, the firmware that came out in mid-summer makes the X-E2 viewfinder work the same way as the X-T1.
DE: Yeah, it works the same as the X-T1 does. Oh, OK, good, good.
Fuji: And the X100's EVF mode and X30, also the same.
DE: Same mode, yeah. Yeah, that was what was confusing to me, because it sounded like it would have to be 200 frames per second, but I knew that it can't refresh that fast. So basically it's synchronous, that it goes together. Very good, OK, that's good.
The new OLED, that's actually something that another company develops OLEDs. Fujifilm doesn't make OLEDs yourself, right?
Fuji: It's designed by Fujifilm, but not manufactured by us. And the optics are ours.
DE: Right. Ah, that's interesting. I figured those were just products, company X makes OLEDs, but you actually designed it and then had someone...
Fuji: Manufacture it to our specification.
DE: Like a chip fab or something.
The triple linear motor, how much faster is that? Can you quantify? Of course, it's a different lens design, but... Do you have any sense, I mean does it give you 30% more, or 50% more power?
Fuji: OK, so, about in telephoto side, autofocus speed in CIPA, it's probably 200 millisecond, 0.2 second.
DE: That's the speed, OK, but I was thinking with three linear motor, how does that compare if it was two linear motor. How much advantage or improvement does that give you?
Fuji: The focusing element in this lens is very heavy, so two linear motors cannot move it.
DE: It would just be very slow.
Fuji: Yes, yes.
DE: Yeah, I see. So it's really, it's not so much that it's 50% better, it's that it just couldn't work with two linear motors. I see. With the apodization filter, are there any drawbacks or reasons you wouldn't use that lens other than just the reduced light transmission? Are there any trade-offs with the apodization filter other than light loss?
WB: And the contrast detect focus, I guess. It's contrast detect only.
Fuji: Some of that the speed of focus. It's going to be a little bit slower, depending on the subject matter of course, but it's not gonna be as fast.
DE: Yeah, so really not anything in terms of image quality, but it is gonna be slower...
Fuji: And the transmission, the T-value, basically.
DE: Yeah, I was wondering if there were any other tradeoffs.
WB: There's no other, like, chromatic aberration tradeoffs, or you know...
DE: Yeah, optical, so no other... the apodization filter doesn't introduce any other optical problems like chromatic aberration or...
Fuji: No, no, no. The MTF is a little bit higher than normal, but...
DE: Oh, the MTF is higher?
Fuji: A little bit higher.
Fuji: It's stopping down a little bit. It's actually... Well, not stopping down, it's getting rid of the marginal rays. So you're using the sweet spot of the lens more with it.
DE: Oh, I see, I see! That's really interesting.
Fuji: So the MTF would be, it would be a hair sharper... Yeah.
DE: And it's... Yeah, and it's not a, it's probably not a linear thing, it's not like I mean, so, at f/1.2 this is actually f/1.7, so it's not like the regular lens necessarily stopped to f/1.7 would be...
Fuji: Right.
DE: So what you lose in brightness, you kind of gain in sharpness.
So the X-Trans sensor design has been very, very successful in terms of image quality. We get great sharpness and no artifacts or aliasing...
Fuji: Moiré.
DE: No moiré, yeah. But it also puts a lot of demands on processing, and our impression has been that the X-Trans works very well for still shots, but not so good for video. I'm assuming that's because there's so much processing to figure out the pixel pattern. Is that something that can improve as processors get faster? You know, if you have more processing power, is it then possible to get a better video image?
Fuji: In the future, our LSI will become faster and capable of more calculations, and maybe our movie image quality will also be able to improve. But it depends on not only this, but also on CMOS sensor reading speed.
DE: So you need to also have faster readout on the sensors maybe, yeah.
Fuji: But there is potential.
DE: Potential, yeah. Well of course, there's a standard reader question. You know, everyone asks "What about full-frame?", you know, will there be a full-frame X-Trans or... I know you guys can't comment about future products anyway.
Fuji: But also we don't need to go to full-frame size, because as you see our APS X-mount system is very small and very lightweight, but image quality itself is as you see on our booth...
DE: Yeah, the huge prints.
Fuji: Huge prints are no problem for resolution and dynamic range or sensitivity, and at this moment we have no plans to go to full-frame. But of course we investigate the ability of full-frame or medium-format. Of course, we know. But at this moment, Fujifilm is concentrating on enhancing our X-mount APS-C lineup.
DE: Yeah, so your strong sense is that, you know, any improvements in image quality or low-light, that it wouldn't be a good tradeoff relative to the size and the compactness. And maybe there's a market for that, but it might be a smaller market, and it's not one that you're going to be addressing. I think that's actually all of our questions. Oh, actually, we ran a little bit over, so... I think we have everything we need. I appreciate it very, very much.
Fuji: Yeah, thank you very much!
DE: Thank you, yeah. It's so good to be able to talk to engineers, you know, and hear the details of the technology. I'm an engineer by background, never practiced as an engineer, but I always like being able to ask questions! Thanks so much for your time!
[quick links: Fujifilm X30 / X100T • Fujinon 56mm f/1.2 APD / XF50-140mm WR]
