Archive | Techsposé RSS feed for this section

Pentax’ PLM and DC lenses explained

2 Mar

DC (“direct current”) is a kind of focus motor that has been used in Pentax lenses for some time. While it is quiet, it’s not entirely silent. Pentax’ most recent 55-300mm lens features a new type of motor branded as PLM. Here is what Pentax representatives had to say about it in an interview:

The PLM design can quickly activate and allow for faster focusing, but the lens element must be low weight due to limited power (torque).

The DC motor can generate high power (torque) with deceleration mechanisms, which is better for lenses with larger focusing elements. A DC motor can be driven at high speed, but there is an issue that a little sound is generated.

Generally, we use the what we feel is the best focusing mechanism for each specific lens design.

It would be reasonable to suspect that the 55-300mm’s new optical formula and narrower aperture were needed to allow the faster, silent motor to be used, although it should be noted that, like the recent 18-50mm kit and non-kit zoom lenses, the new 55-300mm lens is collapsible to a somewhat smaller size, with the difference between collapsed and uncollapsed size being more pronounced in the 18-50mm.

In the interview, the representatives went on to explain that they do not expect to see PLM in a large aperture lens any time soon, instead putting their money on researching other kinds of motors as well as algorithms to improve autofocus.

Advertisements

Pentax KP, affordable DSLR for low light

28 Jan

Ricoh just took the wraps off the Pentax KP, a rather compact magnesium alloy body DSLR with a 5-stop, 5 axis stabilisation system and ISO 819,200. The body is not particularly beautiful, but inside it waits a noise reduction co-processor that, by first appearances, rivals DxO’s PRIME software, but acts instantaneously. (If you’ve used DxO’s engine, you’ll know what I’m talking about.)

05_kp_black_2040

The Pentax KP
(Source: Ricoh promotional materials)

Technical expose:High ISO images are noisier on average than low ISO ones. The noise cannot be neatly compressed, increases file size and slows down the process of writing files to storage (SD card in most cases). This can affect frame rates. In order to keep the frame rate up, Pentax used to apply noise reduction in RAW from ISO 3200 (several models including at least the K-5, K-5 II, K-5 IIs, K-30 and K-50). This was not configurable and led to mushy images that did not respond well to further noise reduction using other methods. So the way to work with these cameras was to underexpose ISO 1600 by up to three stops, depending on your need – not the best idea if you want to maximise colour tonality, but it got the job done.

screen-shot-2017-01-27-at-12-32-51

Where the magic happens:
the accelerator unit.
(Source: Ricoh promotional materials)

But from samples I’ve seen, the new co-processor, dubbed in the latest press release a “state-of-the-art accelerator unit”, renders such concerns obsolete. In the samples, ISO 6400 looks rather clean, and I’m curious to take a closer look at ISO 12,800. Several in the Pentax community have commented that they might delay their entry into full frame based on this camera’s performance, so the pressure is on for Pentax to bring the accelerator unit and high ISO performance to an updated K-1 full frame camera.

screen-shot-2017-01-27-at-12-37-31

Pure edge detection
in Live View
(Source: Ricoh promotional materials)

Launch price for the Pentax KP is going to be 1100 Euros/Dollars; the Pentax K-70 with similar performance up to ISO 102,400 (also has accelerator unit) is about 400 Euros/Dollars cheaper.

08_kp_silver_2040

Display articulation
(Source: Ricoh promotional materials)

Conclusion: The Pentax KP’s combination of five stops of stabilisation with state-of-the-art in-camera noise reduction will save you money on bodies, lenses, and software, as Pentax leapfrogs the competition. Bring on the night!

PS: Check out the separately posted ISO 819,200 sample image to see what it can do!

PPS: Also check out this comparison of ISO 102,400 after noise reduction in the KP and K-70.

Low light photography? Affordable? Look no further!

27 Jan

Ricoh just took the wraps off the Pentax KP, a rather compact magnesium alloy body DSLR with a 5-stop, 5 axis stabilisation system and ISO 819,200. The body is not particularly beautiful, but inside it waits a noise reduction co-processor that, by first appearances, rivals DxO’s PRIME software, but acts instantaneously. (If you’ve used DxO’s engine, you’ll know what I’m talking about.)

05_kp_black_2040

The Pentax KP
(Source: Ricoh promotional materials)

Technical expose:High ISO images are noisier on average than low ISO ones. The noise cannot be neatly compressed, increases file size and slows down the process of writing files to storage (SD card in most cases). This can affect frame rates. In order to keep the frame rate up, Pentax used to apply noise reduction in RAW from ISO 3200 (several models including at least the K-5, K-5 II, K-5 IIs, K-30 and K-50). This was not configurable and led to mushy images that did not respond well to further noise reduction using other methods. So the way to work with these cameras was to underexpose ISO 1600 by up to three stops, depending on your need – not the best idea if you want to maximise colour tonality, but it got the job done.

screen-shot-2017-01-27-at-12-32-51

Where the magic happens:
the accelerator unit.
(Source: Ricoh promotional materials)

But from samples I’ve seen, the new co-processor, dubbed in the latest press release a “state-of-the-art accelerator unit”, renders such concerns obsolete. In the samples, ISO 6400 looks rather clean, and I’m curious to take a closer look at ISO 12,800. Several in the Pentax community have commented that they might delay their entry into full frame based on this camera’s performance, so the pressure is on for Pentax to bring the accelerator unit and high ISO performance to an updated K-1 full frame camera.

screen-shot-2017-01-27-at-12-37-31

Pure edge detection
in Live View
(Source: Ricoh promotional materials)

Launch price for the Pentax KP is going to be 1100 Euros/Dollars; the Pentax K-70 with similar performance up to ISO 102,400 (also has accelerator unit) is about 400 Euros/Dollars cheaper.

08_kp_silver_2040

Display articulation
(Source: Ricoh promotional materials)

Conclusion: The Pentax KP’s combination of five stops of stabilisation with state-of-the-art in-camera noise reduction will save you money on bodies, lenses, and software, as Pentax leapfrogs the competition. Bring on the night!

PS: Check out the separately posted ISO 819,200 sample image to see what it can do!

Craving full-frame? Read this first.

3 Jul

Reading this article might save you some money. No, I’m not sponsored to write this, so you may safely proceed. All you’ll get is technical insight and honest opinion.

Perhaps like many photographers, you’ve thought about going for a full frame camera. Perhaps you’ve heard about shallow depth of field, low light shooting and noise. Maybe you’ve heard that full frame is “one stop faster” or “one stop brighter”.

In this article, I’ll cover one of the reasons for going full frame – more light, and how much sense it makes, especially for your bank balance.

Frontal product photo of a black camera against blueish white background.

Ever wanted a Nikon D800E? Read on. (Image credit: Jastrow)

By going from APS-C to full frame – probably the most common move – you gain approximately one stop of light, i.e. you get twice as much light over the whole of your sensor. It would be approximately true to say that shooting ISO 200 on full frame, you get the same noise as shooting ISO 80 on APS-C. So you can shoot faster shutter speeds and, over the whole sensor, get the same noise. However, if you shoot 16 megapixels on APS-C, then shooting 36 megapixels full frame, you will get the same PER PIXEL noise, at the same ISO. So far, so good.

Small rectangle denoted APS-C next to larger rectangle denoted

Relative size of the two major DSLR sensor formats (Copyright breakfastographer / Chriusha (Хрюша) / CC-BY-SA-3.0)

So if we shoot the same apertures, we gain a stop of light. So we can ask the question, how much does it cost to gain a stop of light?

If you upgrade from an 18-55/3.5-5.6 APS-C kit lens to a 17-50/2.8 APS-C premium zoom, you gain 1-2 stops of light, and it might cost you 150 Eurodollars if you buy used. Or for about the same price, you could upgrade to a 17-70/2.8-4.5, which gives you almost exactly one stop advantage over the kit lens, and a bit of extra range.

If you further upgrade to an 18-35/1.8 zoom, you might pay about 600 Eurodollars used or a good offer new, and you would gain a stop over the 17-50/2.8 or over two stops vs. an 18-55/3.5-5.6 kit lens. But you shorten your range by about one stop of teleconversion. Don’t worry if that sounds technical – just remember that 35mm is 35mm and not 50, 55 or 70mm.

Black zoom lens pointing upwards with hood attached, and lens cap lying next to it, label up. Zoom ring indicates 17-24-35-50. White background.

Another possible upgrade path? The Tamron 17-50/2.8 (Image credit: Christian Fischer/CC-BY-SA-3.0 Unported)

If low light shooting, reduced noise or faster shutter speeds are what you’re after, and you’re still shooting with an APS-C camera and kit lens, you owe it to yourself to first invest the 150 to see if the extra light is worth it. This will put you in a much better position to judge whether you want to invest about ten times that amount (or more) to get to a full frame camera and appropriate lens. (Note that a full frame camera with an f/4 lens offers no light advantage over an APS-C camera with an f/2.8 lens!)

There are other points to consider, of course, some of which I’ll touch on briefly. A full frame camera generally offers greater sharpness due to both an increase in resolution and easier manufacture and assembly of appropriate glass. The full frame format is more resistant to diffraction, typically tolerating f/11 rather than f/8 (APS-C) or f/5.6 (Micro Four Thirds) – note that these assumptions only hold for certain pixel pitch, i.e. a 36 megapixel full frame camera is just as sensitive to diffraction at the pixel level as a 16 megapixel APS-C camera. (Other assumptions apply, such as having comparable filter stacks in front of the sensor.) However, over the entire frame, the full frame camera is more tolerant. The flip side is that you need to shoot at those smaller stops to reach the same depth of field, meaning that in landscape shooting, there is no full frame advantage in available light, for the same positioning and framing.

Furthermore, full frame has greater limitations when it comes to designing zoom lenses – you may notice that a common superzoom lens specification for full frame is the 28-300mm lens, whereas on APS-C, the limit has been pushed to 16-300mm – the same versatility in focal length as a 24-450mm on full frame!

Similarly, compare the Sigma 18-35/1.8 to its full frame sibling, the 24-35/2. The smaller lens is both faster and, depending on your point of view, slightly more versatile. (The comparison is hampered by the full frame lens being the equivalent of a 16-24mm on APS-C – not an easy comparison!)

So a full frame camera is best suited to the shooter who knows what situations they want to cover and what their preferred focal lengths are. If you prefer spontaneous shooting and versatility, it is likely that a full frame camera will not make you happy.

Arguments about achieving shallow depth of field more easily on a full frame camera deserve a separate article.

Update 2/2/2017: Also check out my newer article, Low light photography? Affordable? Look no further! for the latest development, or check out a sample image.

Pentax: That thing with the pin

14 Aug

Sigma currently is discontinuing some of its lenses, and seems to be targeting K-mount as the first mount to do so. It was always clear that Sigma would update its long zoom range to its new Sports line, which, if nothing else, is distinguished by granting access to Sigma’s Mount Conversion Service, which gives users the option to keep the lenses they buy when they change to a different supported camera system. The idea is that you could buy a Canon lens and take it with you when you move to Pentax, Sony, Nikon, or Sigma, at a fee and a few weeks’ waiting time (during which, if you were a professional, you could rent a replacement lens from somewhere, or buy for the interim and sell the worse copy when your first one gets back from Sigma).

So far, Sigma has updated or released a number of lenses for its Art line, which is also eligible for the programme, but not much was heard for Sports or Contemporary, its remaining two categories (Sports is for long lenses, Art for high quality lenses from moderate wide to portrait focal lengths, and Contemporary is presumably for everything else).

So when it comes to Pentax mount, people see that the K-mount bodies are selling like hotcakes, at least in some regions, but feet still get chilly whenever there’s a possibility that fewer lenses will be available in the near future. As a result of discontinuations, people then overpurchase lenses that they think they might need in the future, such that remaining stock can disappear quickly. This apparently happened with the 70-200mm f/2.8, 100-300mm EX, and 120-400mm lenses. In addition, some people called their Sigma representatives and were told different things. One speculation apparently was that the 70-200 was discontinued because only two to three of them were sold, per month, for K-mount in the US. Another person was able to place a “back-order” for one, suggesting it isn’t really discontinued.

Sigma have, in a recent interview, stated that supporting Pentax is more difficult than other brands because Pentax still controls aperture through a mechanical lever. This increases research and development as well as machining costs. In spite of this, Sigma have in the past kept the same price point for lenses in various mounts. From Pentax/Ricoh’s point of view, this circumstance may be seen as a competitive advantage, since at least historically, camera makers have sought to sell their own lenses rather than allowing third party lenses to be bought. On the other hand, having a smaller lens collection available for purchase may cut off sales entirely if people take it as a reason to stop buying into bodies.

It’s also true to say that the contact point between the camera’s and the lens’ aperture mechanisms is an additional source of “shutter” noise when shooting at less than widest aperture. Lenses on any TTL system – Olympus, Sony, Canon or other – ultimately need to move their aperture blades into position. (Btw, that’s one reason why cameras with an optical or hybrid viewfinder, such as the Leica M series or Fujifilm X100, are better at street photography.) However, its clear that Pentax’ additional aperture lever is a source of significant noise.

Not knowing what else may happen at Photokina, and whether something special is happening at either Pentax or Sigma to create these perturbations in Sigma’s lens supply chain, I would still like to ask the question why Pentax doesn’t include an additional aperture-control electrical contact on its next generation of bodies to allow third party lenses to be more easily made or mount-converted, leading to a larger available lens selection that could benefit the wider adoption of a set of bodies that for many applications are still the best available.

Will Canon eliminate viewfinder black-out?

13 Aug

Viewfinder blackout is one of the most upsetting limitations of current camera technology. So far, it affects all cameras that produce a viewfinder image through the same lens that is subsequently used for image capture.

Canon now has a patent on switching from the traditional phase-detect autofocus (PDAF) sensor to its dual pixel on-sensor PDAF. Like many patents, this is in itself so obvious that it doesn’t, in my opinion, justify a patent. However, Imaging Resource has mused whether this means that there is a sister patent that will explain how a viewfinder image will continue to be generated during such capture. Obviously, this would point at some kind of hybrid viewfinder, but Canon might, in the process, succeed in eliminating black-out. Ironically, one way to implement such a viewfinder would be through another additional pellicle or reflex mirror.

It certainly is bloody time that somebody eliminated black-out. Aptina’s sensors can be read out at 60 Hertz to record a full resolution image. In spite of Nikon’s 1 series supporting such a framerate, something is clearly wrong with the electronics design since even those cameras still have black-out.

One has to wonder whether Canon will eventually bowl itself out of the game with its apparently steam-punk approach to this problem. At the same time, one wishes that Nikon electronics and software engineers would get their hands out of their ***** and produce a better processing pipeline. I have no doubt the Germans could do it if they put their minds to it. Hey Leica, are you listening?

Source: Egami (Google Translate Link Warning: LINK)

The more megapixels, …

6 Oct

…the sharper your lens has to be.

…the more accurate your focus has to be.

…the faster your shutter has to be OR

…the better your image stabilisation has to be (IBIS, OIS or tripod).

…the bigger your hard disk has to be.

…the faster your computer has to be.