Photo Of The Day By Ryszard Lomnicki

Today’s Photo Of The Day is “Tuscany” by Ryszard Lomnicki. Location: Val d'Orcia, Italy.
Photo By Ryszard Lomnicki

Today’s Photo Of The Day is “Tuscany” by Ryszard Lomnicki. Location: Val d’Orcia, Italy.

See more of Ryszard Lomnicki’s photography at camerapixo.com/photographers/ryszard-lomnicki.

Photo of the Day is chosen from various OP galleries, including Assignments, Galleries and the OP Contests. Assignments have weekly winners that are featured on the OP website homepage, Facebook, Twitter and Instagram. To get your photos in the running, all you have to do is submit them.

The post Photo Of The Day By Ryszard Lomnicki appeared first on Outdoor Photographer.

Nvidia CEO believes AMD has suffered a ‘great loss’ and Intel is playing GPU catch-up

Nvidia just posted some impressive financial results, and the company is certainly in a buoyant mood, with the CEO taking the time to divulge his thoughts on the recently-revealed AMD and Intel partnership in laptop CPUs, as well as the defection of a key executive from the former to the latter.

The Q3 fiscal results were certainly strong, with Nvidia notching up a record revenue of $2.64 billion (around £2 billion, AU$3.45 billion), an increase of a third compared to a year ago. The firm made big gains with data centers, but surprisingly also experienced a big jump in gaming revenue with a 25% increase year-on-year – flying in the face of analysts’ expectations.

After boasting of bulging coffers, chief executive Jensen Huang talked on the subject of Raja Koduri leaving AMD to become Intel’s senior VP of the Core and Visual Computing Group, with a remit to deliver ‘high-end discrete graphics’. Yes, discrete graphics solutions, not integrated (on-processor) affairs.

As Tom’s Hardware reports, Huang commented: “Yeah, there's a lot of news out there… first of all, Raja leaving AMD is a great loss for AMD, and it's a recognition by Intel probably that the GPU is just incredibly important now.

“The modern GPU is not a graphics accelerator, we just left the letter ‘G’ in there, but these processors are domain-specific parallel accelerators, and they are enormously complex, they are the most complex processors built by anybody on the planet today.”

He goes on to point out that this is exactly why “every major server around the world has adopted Nvidia GPUs.”

Graphic detail

So, there are a couple of things here. First of all, and obviously enough, it’s not surprising that the Nvidia CEO wants to paint Koduri’s departure as a bad thing for AMD, and AMD’s graphics cards, in terms of it being a ‘great loss’.

Also, on the Intel side of the equation, Huang focuses on the company’s need to drive forward with graphics processors as a critical one. And this likely reflects the fact that the mentioned discrete GPUs Intel talked about in its press release welcoming Koduri into the fold is more about targeting heavyweight arenas such as AI and machine learning, rather than anything to do with gaming.

In other words, Intel doing discrete graphics is certainly big news that will make big waves, but not in terms of consumer graphics cards.

Note that Intel has tried its hand at discrete graphics cards in the past – or had brief flirtations would perhaps be a better way of putting it – but you get the sense that it’s truly a serious drive this time around.

Furthermore, Huang took time to comment on AMD and Intel teaming up to make laptop processors with integrated AMD graphics, news which broke earlier this week.

His somewhat rambling comment on the matter was: “And lastly, with respect to the chip that they [Intel and AMD] built together, I think it goes without saying, now that the energy efficiency of Pascal GeForce and the Max-Q design technology and all of the software we have created has really set a new design point for the industry, it is now possible to build a state of the art gaming notebook with the most leading edge GeForce processors, and we want to deliver gaming experiences many times that of a console in 4K and have that be in a laptop that is 18mm thin.

“The combination of Pascal and Max-Q has really raised the bar, and that's really the essence of it.”

In short: Nvidia’s rivals need to do something, because the firm’s latest advances with Max-Q are pushing the notebook graphics envelope so much.

Strong words all round then, but given its current form, Nvidia is unlikely to be short of confidence. Particularly when looking to a future in which graphics processors are key to the likes of supercomputers and cutting-edge fields such as AI and machine learning.

  • These are the best graphics cards you can buy in 2017

Fujifilm X-E3 review

The X-E3 is the latest X Series mirrorless camera from Fujifilm to get the company's 24.3MP X-Trans CMOS III sensor, and replaces the ageing X-E2S.

While the likes of the X-T2 and X-T20 take on a DSLR-style design, with a raised, central viewfinder, the X-E3 follows the model of a classic rangefinder-style camera in a similar vein to the X-Pro2.

Whereas the X-Pro2 is aimed at professionals and keen enthusiasts, the X-E3 has been designed to appeal to a slightly broader audience, with more diminutive proportions and streamlined controls.

The specification hasn't been compromised to achieve this, however, with the X-E3 sharing an almost identical set of features with the X-T20, along with a few tricks of its own. 

Features

  • APS-C X-Trans CMOS III sensor, 24.3MP
  • 3.0-inch touchscreen, 1,040,000 dots
  • 4K video capture

As we've seen with every Fujifilm X Series camera since the X-Pro2, the X-E3 features the company's 24.3MP X-Trans CMOS III APS-C sensor, which delivers a decent boost in resolution over the X-E2S’s 16.3MP sensor.

This also means it gets a moderate increase in ISO range over its predecessor, with a native sensitivity range of ISO200-12,800 (compared to ISO6400 on the X-E2S), while the expanded range now runs to ISO100-51,200. And where the X-E2S was restricted to JPEG-only files at the extended range, the X-E3 supports raw files as well.

Like the X-E2S (and the X-T20 for that matter), the Fujifilm X-E3 features a high-resolution 2.36 million-dot OLED electronic viewfinder (EVF) with a 0.62x magnification, while the rear of the camera is furnished with a 3.0-inch touchscreen display with a resolution of 1,040,000 dots. 

Unlike both the X-T2 and X-T20, the X-E3 forgoes a tilt-angle display in favor of a flush-fitting design. The touchscreen functionality builds on the system seen on the X-T20 however, offering a greater depth of control. As on the X-T20, then, you can tap the display to acquire focus and trigger the shutter, as well as being able to swipe through and pinch-to-zoom when reviewing images.

There's still no direct control for navigating the menus, but the X-E3 does let you use flicking motions to activate pre-assigned functions, while you can also adjust settings in the Quick menu. 

The touchscreen functionality builds on the system seen on the X-T20, offering even a greater depth of control

As we've seen with other X Series cameras, there's Wi-Fi and NFC connectivity, but a first for the range is Bluetooth. Once you've paired the camera with your smart device and downloaded the accompanying Fujifilm Camera Remote app you should be able to easily transfer your images seamlessly, to share on social media.  

The X-E3 also offers 4K video capture (3840 x 2160) at 30p as well as Full HD (1920 x 1080) at 60p, with both formats supporting film simulations. There’s a 2.5mm jack input for a microphone, but not for audio monitoring.

Finally, there’s just a single SD card slot on the X-E3, and the camera doesn’t support the faster UHS-II cards.

  • The 10 best mirrorless cameras you can buy right now

Build and handling

  • Magnesium top and bottom plates
  • No weather sealing
  • Weighs 337g

While the X-E2 and X-E2s followed much the same design as the X-E1, the Fujifilm X-E3 is markedly different. Noticeably smaller than the X-E2s, the X-E3 takes the crown as the smallest X Series camera with a viewfinder currently available. Partner it with a couple of Fujifilm's neat f/2 primes, like the 23mm or 35mm, and you've got a great little camera system. 

Don't think the reduction in the X-E3's footprint has compromised its handling; there's a decent handgrip that's pretty much identical in size to the one on its predecessor, while the raised thumb rest on the rear of the camera makes it incredibly comfortable to hold one-handed. 

As we've come to expect with X Series cameras, the fit and finish are very nice. It may not be weather-sealed, but the magnesium alloy top and bottom plates, along with the machined dials, give the X-E3 a lovely premium feel. 

With the X-E3's body shrinking compared to the X-E2s, the built-in flash has had to be sacrificed, to be replaced by a pocket-sized hotshoe-mounted flash. Otherwise, the top plate of the X-E3 remains the same as on the X-E2, with shutter speed and exposure compensation dials complementing the shutter button and small function button. 

The shutter speed dial offers settings running from 1 to 1/4000 sec plus Bulb, Time and Automatic, with the exposure compensation dial running from -3 to +3EV. As we've seen with other X Series cameras, there's also now a ‘C’ setting on the dial, which lets you set compensation up to ±5EV using the camera's new front command dial. 

A subtle change is the arrival of a small rocker switch round the shutter speed dial, with the option to set the camera to a full Auto mode – perfect for new users who just want to start snapping and without getting bogged down wondering which aperture or shutter speed to use.

The four-way controller on the X-E2 has completely disappeared on the X-E3

The biggest changes, though, are at the rear of the X-E3, with a much more streamlined layout than we've seen previously. 

The display now butts up to the left edge of the camera, while the four-way controller from the X-E2 has disappeared. Instead, you can use the touchscreen control to flick right, left, up or down with your thumb to access four different functions or settings, which can be tailored to your preference in the menu. 

The X-E3 also benefits from the focus lever we first saw on the X-Pro2, enabling you to toggle quickly to the desired AF point, while you can also use it to navigate the camera's menu system if you prefer that to the touchscreen. 

As we've seen with other X Series cameras, the level of customization impresses, while the overall handling is very satisfying. The absence of a four-way controller isn't an issue, while there are plenty of body-mounted controls to make this a quick camera to operate.

Autofocus

  • 325-point AF
  • Eye-detection AF
  • 5 AF-C presets

The Fujifilm X-E3 uses the same 325-point AF system as the X-T20. This is broken down into 169 phase-detect points set out in a grid of 13 x 13 in the middle of the frame, with an additional two grids of 6 x 13 contrast-detect points on either side to make up the 325 focusing points. 

That's if you're using the X-E3's single-point AF mode. When you opt for Zone or Wide/Tracking, coverage drops to a still-decent 91-point AF arrangement. In this instance, there's a central 7 x 7 grid of phase-detect points.

For static subjects, focusing is swift – the X-E3 happily locked onto a variety of subjects under a range of lighting conditions, while switching to continuous AF saw a welcome improvement in focusing accuracy over the X-E2. 

This is thanks to a new AF algorithm that takes three parameters into consideration. These are Tracking Sensitivity (how long the camera waits before switching focus), Speed Tracking Sensitivity (determines how sensitive the tracking system is to changes in subject speed) and Zone Area Switching (whether bias is towards the center, auto or front).

With five presets to choose from depending on how erratically your subject is moving, AF tracking is much improved. Focusing speed could be a bit quicker (it's not quite a much for the likes of Sony's Alpha A6300), but it's a very solid performer. 

  • 10 best camera focusing techniques

Performance

  • 14fps burst shooting (electronic shutter)
  • 8fps burst shooting (mechanical shutter)
  • 350-shot battery life

Thanks to Fujifilm’s X-Processor Pro imaging engine the X-E3 takes 0.4 seconds to power up, compared with 0.5 seconds for the X-E2s, while the shutter lag remains the same at just 0.05 seconds. 

While the X-E3 isn't marketed as a action camera, it is capable of shooting at a very fast 14fps using the electronic shutter, or at a slightly more restrained 8fps if you'd prefer to use the camera's mechanical shutter – if you do, expect to be able to shoot 23 raw files or 62 JPEGs before the buffer slows up.

The X-E3 uses Fujifilm's tried and tested TTL 256-zone metering system, which in the main performs very well. As we've found before, when presented with a high-contrast scene it can sometimes underexpose the shot a touch, but this is easy to correct thanks to the well-positioned exposure compensation button – or you can simply lift the shadows in post-processing if you wish.

The electronic viewfinder is the same unit found on the X-T20. As we've found with that camera the viewfinder image is lovely and bright, and, thanks to the 2.36m dot resolution, rich in detail. 

Toggling through the X-E3’s settings is pretty painless when combined with the rear joystick

While there's no tilt or vari-angle positioning of the rear display – either would have come at the expensive of the X-E3's compact build – the 3-inch touchscreen is very good. It would be nice to see the menu system integrated into the touchscreen interface, but toggling through the X-E3's settings is pretty painless when combined with the rear joystick.

One area where there's room for improvement is battery life. At 350 shots per charge, it's certainly not bad for a mirrorless camera, but when compared to a DSLR rival such as the Nikon D5600 and its 820-shot battery life there's certainly room for improvement. It's nice to see a dedicated charger bundled in the box, while you can charge via USB as well.

The upgraded connectivity options of the X-E3 work well. It's easy to pair the X-E3 with your device, and provided Bluetooth is turned on on both devices an automatic, low-powered connection is established every time you fire up the Camera Remote app. 

Image quality

  • ISO200-12,800, expandable to 100-51,200
  • Film simulation modes
  • +/-5 EV exposure compensation in 1/3 or 1/2-stop increments

The Fujifilm X-E3 uses the same X-Trans CMOS III sensor as we've seen in the X-Pro2, X-T2, X-T20 and X100F, which means the image quality from the X-E3 is some of the best available from an APS-C-based camera.

Detail is excellent, while those not wanting to spend time sitting in front of a computer will enjoy the X-E3's range of film simulation modes. With 15 effects to choose from there's a wealth of options available, with the likes of Velvia great for enhancing colors in landscapes, while mono shooters will enjoy the Arcos mode. 

Another area where the X-E3 excels is dynamic range. There's plenty of flexibility for recovering detail in the shadows and highlights of images – especially raw files, with about four stops to play with at low ISOs.

Available for both JPEG and raw files is the X-E3's Dynamic Range mode, with the highest DR400 setting preserving plenty of detail in both the highlights and shadows, though this comes at the expense of the base sensitivity available in this mode, with only ISO800 or higher available.

As we've found with other X Series cameras that use this sensor, the X-E3 handles image noise very well. At the lower end of the sensitivity range files look very clean, with no signs of luminance (grain-like) noise in our shots.

It's only really at ISO3200 that luminance noise becomes noticeable, and its organic appearance means it's not a big issue, while even up to ISO12,800 results won't be unusable – you'll need to be prepared to tweak files in post-processing if you're planning to shoot at this sensitivity though.

  • The 10 best cameras you can buy right now

Verdict

The Fujifilm X-E3 is a cracking little camera. The premium finish is complemented by an array of tactile controls and a touchscreen interface that make this camera a joy to shoot with. 

There are no complaints when it comes to image quality either – photos are rich and detailed, while the film simulation modes can add a lovely twist to your shots. And, thanks to the improved connectivity, it's possible to share striking-looking images straight out of the camera before you're even home.

We'd avoid pairing the X-E3 with some of Fujifilm's larger zoom lenses – it's really a camera to enjoy with some neat prime lenses – while the battery life means you might want to invest in an extra battery or two.

Fujifilm may have shrunk the camera, but it hasn't sacrificed performance, and the X-E3 is everything you'd want in a compact-sized mirrorless camera.

Competition

Processing For Neutral Fall Color Photos

Processing for neutral fall color
Left: RAW file as captured by the camera. Right: Final image after all adjustments.

Interpreting color in an image is our prerogative as photographers. Some will decry overly saturated colors as unnatural, and I’ll admit to sharing the opinion that it’s easy to go too far with color adjustments, but ultimately it’s up to you to decide what looks best for your photographs.

Whatever your preferences, a good place to start when developing an image is with neutral, balanced color and make further adjustments from there if desired. The following technique uses the curves tool to get to that neutral starting point. I generally prefer to shoot in RAW, so in this example I’ll be using Adobe Camera Raw, but you can use it with JPEGs, too, in any software application that offers curves adjustments.

This technique to achieve neutral fall color is part science and part art. The science comes from comparing and adjusting the RGB values of three target points in the image: one white point, one black point and one gray point. The art is in selecting those points.

The fundamental concept to understand is that when the RGB values for a given pixel are identical, you have a color-neutral pixel. For example, a pixel with values of R=255, G=255 and B=255 is pure white, while one with values of R=0, G=0, B=0 is pure black. By targeting three points in an image—one white, one black and one gray—and adjusting their individual RGB values to be as close to identical as possible (I aim for a variance of 3 or less), you’ll arrive at neutral color throughout the image.

Step 1: Make A Basic Exposure Adjustment

The first step—before selecting your black, gray and white target points—is to get the exposure in the ballpark of where you think it should be. This is especially necessary for processing RAW captures, as the camera hasn’t made any adjustments for you. Figure 1 shows the original capture in Adobe Camera Raw, and Figure 2 shows the result of bringing the exposure down to -1.69.

Neutral fall color, before processing
Figure 1. The original RAW capture in Adobe Camera Raw prior to adjustments.
Neutral fall color, after initial exposure adjustment
Figure 2. The image after making initial adjustments to the exposure.

Notice the clipping warning at the top left of the histogram in Figure 2. This is indicating that the blue channel is being clipped (losing data). Ideally, you won’t have any clipping; in this case I had to choose between clipping the red channel in the image’s highlights or clipping the blue channel in the midtones. I opted for the midtones because I was confident I could bring that back later. It’s not absolutely critical that you have no clipping, but keep an eye on those warnings and avoid it whenever possible.

Step 2: Targeting With The Color Sampler Tool

With the exposure set, the next step is to select a white, gray and black point in the image. To do this in ACR, you use the Color Sampler Tool, which looks like an eyedropper with a small target, found in the toolbar above the image. First, I selected a white point. Choose a spot that is as close to pure white with no detail as possible. Specular highlights are ideal.

Next, I chose the gray point. This one is the trickiest of the three and is where the “art” of this technique comes in. Do your best to find an area that should be a neutral, midtone gray. White objects in shadow can work well.

Last, I selected the black point, a black spot on one of the trees that was also in shadow. Every image is different, and you may not get pure white or pure black. The technique will still work as long as you get close.

Neutral fall color, setting target points for white, middle gray and black
Figure 3. The Color Sampler Tool, target points and target point RBG values are outlined in red.

As you can see in Figure 3, the white point was nearly pure white, with a value of R=252, G=251, B=252. The gray point was less neutral, with values of R=106, G=106, B=103. The black point was less neutral still at R=31, G=26, B=26. Overall, though, this was good news, in that these values are already pretty close to neutral and I wouldn’t need to make major adjustments.

Step 3: Adjust The Gray Point

With the target points set, I switched to the Tone Curve tool. In ACR, it’s the second tab in the tool palette. I like to start by adjusting the gray point. With the tone curve tab open, select the “Point” tab. You’ll now see a graph of all three channels’ values, with the pure black on the bottom left and pure white at the top right, and a diagonal line connecting those points. The diagonal line is what you will click and drag to adjust each of the target points’ values.

Neutral fall color, midtown adjustment
Figure 4. The Tone Curve tool palette, Point tab, Channel dropdown and midtone adjustment point, and midtone values (top left) are outlined in red.

In this example, the gray point’s red and green values were identical at 106. It’s the blue value I wanted to bring up to match, so in the Channel dropdown, I selected “Blue.” Since I was adjusting the midtones, I clicked on the center of the diagonal line (in the middle of the grid) and dragged that point straight up until all three values, R, G and B, were identical at 106. Bonus—the blue channel clipping from Step 1 was corrected by this adjustment. That was it for the midtones in this example.

Step 4: Adjust The Black Point

Next, I adjusted the blacks. Since the only option is to increase the values at the black end of the curve, I needed to increase the output of the blue and green channels (both with values of 26) to get closer to the red channel’s value of 31. I started with the blue channel, clicking on the bottom left corner and dragging straight up until I got a value of 30. In doing so, the red channel also shifted to 30 (Figure 5). Don’t be alarmed when this happens. Adjusting one channel can affect the other channels’ values and the other target points’ values as well. This technique requires some tinkering back and forth—keep your eye on the values as you go. The goal is to get them as close as possible.

Neutral fall color, blue channel adjustment
Figure 5. Blue channel adjustments and resulting target point values.

With the red and blue values now both at 30, I switched to the green channel, clicked and dragged it up (Figure 6). I stopped at 29 instead of 30, because moving it further affected the red and blue channel values. Again, tinkering. With this adjustment, all three points now had RGB values at my desired variance threshold of 3 or less.

Neutral fall color, green channel adjustment
Figure 6. Green channel adjustment and resulting target point values.

Step 5: Adjust The White Point

Depending on your image, you may need to adjust the white point’s values as well. This is done with the opposite approach that I used for the black point. You can only decrease the output of the channels at this end of the curve, so you’ll pull down the two channels with the highest values to meet the lowest of the three by repeating the technique above, but instead clicking on the top right corner and dragging down with the appropriate channel selected. Since my values were already very close (Figure 5) at R=252, G=251, B=252, I was satisfied with the curves adjustments and skipped this step, and was ready to make my final enhancements.

Step 6: Do Your Thing

At this point with the technique, you’ll have an image that’s essentially color neutral. Now you can apply your artistic license to achieve your vision for the image, knowing that you’re starting from a balanced foundation.

In this example, I returned to the Basic tab of the ACR tool palette to make some final adjustments to the exposure (Figure 7). Using the slider controls, I increased the Contrast a small amount, modestly decreased the Highlights and Blacks, and considerably reduced the Whites, all the while keeping my eye on the histogram’s clipping warnings and dialing back adjustments when clipping occurred.

Neutral fall color, final exposure adjustments
Figure 7. Final adjustments to exposure.

As a final step, I clicked “Open Image” to leave Adobe Camera Raw and proceed to Photoshop, where I did a little rotating and cropping of the image to straighten the horizon line, as you can see in the before and after image at the opening of this article.

There are, of course, numerous approaches to image development, many of which are faster and less involved than this technique, but I come back to it time and again because I’m confident that whatever creative choices I make after the curves adjustments will be from a neutral starting point. I hope you’ll find it useful, too.

The post Processing For Neutral Fall Color Photos appeared first on Outdoor Photographer.