Richard Butler

Richard Butler

DPReview Administrator
Lives in United Kingdom Seattle, United Kingdom
Joined on Nov 7, 2007

Comments

Total: 4520, showing: 1 – 20
« First‹ Previous12345Next ›Last »
In reply to:

Impulses: Is there actually a V60 rating? Like officially? Maybe the reason it's not being displayed is because it doesn't exist, because there's definitely several U3/V30 UHSHI cards that can manage sequential writes at double and triple their rating...

The ones that can do 60MB/s aren't even that expensive (90MB/s for writes does gets a little more expensive).

Usually new standards are all about increasing headroom well in advance of actual need. Cameras haven't been fully leveraging UHS-II until pretty recently (granted my E-M5 II took advantage of it and it's over a year old).

The V60 (and V90) standards were [announced a year ago](https://www.dpreview.com/news/3305606776/new-sd-card-5-0-protocol-supports-up-to-8k-video-recording).

However, we've not seen any cameras wear the branding yet. To the best of my knowledge, that's because no cards that are *dependably* fast enough have yet been made: if the '90MB/s (peak)' cards could reliably write faster than 30MB/s, you'd expect the SD Card Association's members to push for there to be a way for them to show that off.

Link | Posted on Feb 24, 2017 at 22:55 UTC
On article Modern Mirrorless: Canon EOS M5 Review (1605 comments in total)
In reply to:

TheFan: Am I the only one who can't find lens used at image quality compare ?

I downloaded full image but photoshop erase lens exif info.

Does anyone know which lens is used for image quality compare ?

Wherever possible we use native-mount 50mm lenses or their approximate equivalents, because they perform extremely well (both sharp and consistent across the frame) at the apertures we shoot them at.

We have found this gives us much more consistent results than trying to adapt lenses or use different mount versions of third-party lenses.

Link | Posted on Feb 24, 2017 at 22:40 UTC
In reply to:

CeleryBeats: Guys c'mon. A little effort in trying to put together some decently composed images that tell the story and significance of why and to whom medium format is relevant or why it is artistically a very high quality tool.

Not trying to be insulting here. But a little bit more effort could have been done here. Why not ask a qualified (medium format) photographer to do some shots for you? Would give a much better image for this camera and website overall. Wich is still the most popular in the world i would imagine.

As the headline states, these are our **first** shots (and among the first independent shots to be published).

While it would be nice to have the time to shoot more and better images (which we will be doing), the demands of immediacy and quality tend to be in tension with one another.

We don't know a great many 'qualified' photographers in Yokohama, where we were first given access to the camera, so we shot what we could.

Link | Posted on Feb 24, 2017 at 20:14 UTC
On article Sources of noise part two: Electronic Noise (237 comments in total)
In reply to:

neonlight: I do not agree with your comment that the size of pixels does not matter. As you mentioned, light is random and noise in the image is dependent on having reasonable to large numbers of photons landing on the sensor diode. Larger pixels will collect more light than smaller pixels, which will tend to reduce noise from random number variations in the light source.
However, electronic noise sources within a pixel are largely down to how well a chip has been designed and how capable the manufacturing process used to make it is.

RMKO - Quite the opposite. Pixel size only tells you about pixel level results. Since there's nothing to suggest that people are viewing or printing images larger, just because pixel counts have risen, doesn't is make more sense to talk about the effect on the final image?

As you point out, any two pixels of the same size will get the same amount of light for the same exposure. However, the larger sensor will have more of them, which means either greater resolution or cleaner images (since you combine multiple pixels when you downsize, which decreases the effect of noise).

Focusing on the effect on the whole image shows that pixel size makes little difference to overall image noise whereas sensor size does.

If a larger sensor has better image-level noise than a smaller sensor, regardless of whether it has larger pixels or more of the same-sized pixels (and is better by a very similar amount), then doesn't that point to it being the larger capture area that has the dominant effect?

Link | Posted on Feb 24, 2017 at 02:04 UTC
On article The effect of pixel size on noise (112 comments in total)
In reply to:

Irakly Shanidze: Physics here is very simple. One pixel does not care how much light the whole sensor gets. All it cares about is what it personally gets to deal with its own signal/noise ratio.

Yes, you observe something that is real (there seems to be more noise on a smaller sensor with the same size pixels), but misinterpreting this result. The amount of noise is the same PER PIXEL, but since the resolution is less, there is less useful information PER PIXEL. Another words, the noise is not more, it is larger compared to the image itself, that is why the picture looks noisier. Not because there is less light on the sensor.

Each pixel may be an individual detector but unless you always output at 100%, you end up combining data from neighbouring pixels when you print or display.

So, the pixel-level results of the D810, D810 crop region and D7000 all look the same. However, as soon as you scale (and hence combine pixels) the D810 full image down to the same scale as the smaller region captures, you get to see the effect of the addition light. You've captured the same amount of light at each pixel (with the same pixel-level SNR) but this means improved SNR performance over the whole image.

So, for instance, have a look at DxO's SNR 18% data (which is primarily dictated by efficiency and shot noise):

Viewed at a pixel level, then yes, larger individual pixels produce better results. However, scale them to a common size and almost all contemporary sensors perform very, very similarly to other sensors the same size.

Link | Posted on Feb 23, 2017 at 23:29 UTC
On article First Look: Fujinon MK18-55mm T2.9 cine lens (64 comments in total)
In reply to:

MrBrightSide: I understand the motivation for producing this lens, but it's still it's ironic that in this era of evermore precise touch-to-focus and electronic Aperture controls that there is still such a high demand for ultra low-tech lens technology.

osv - I'm fully aware of that. Note that I didn't say anything about autofocus.

The feature on the GH5 sees you define the focus distance before you start shooting (either by manual or autofocus). When you're shooting, 'Focus Transition' drives the lens to those predefined distances and then stops. It doesn't check or attempt to acquire focus, it simply moves in a controlled manner to the distance you requested.

Link | Posted on Feb 23, 2017 at 20:25 UTC
On article First Look: Fujinon MK18-55mm T2.9 cine lens (64 comments in total)
In reply to:

VictorTrasvina: Perhaps Fuji might consider giving us an affordable S35 4K Video Camera? Like the GH5? And if you could already use your primes ?? Well :)

It's possible, I suppose (I don't know how much translational movement is needed to provide effective IS).

However, I'm not sure IBIS quality is considered effective enough for professional video work, so it may be a level of cost and complexity that the audience doesn't want to pay for (given the majority will already have some kind of stabilization solution in place).

Link | Posted on Feb 23, 2017 at 19:59 UTC
On article Sources of noise part two: Electronic Noise (237 comments in total)
In reply to:

neonlight: I do not agree with your comment that the size of pixels does not matter. As you mentioned, light is random and noise in the image is dependent on having reasonable to large numbers of photons landing on the sensor diode. Larger pixels will collect more light than smaller pixels, which will tend to reduce noise from random number variations in the light source.
However, electronic noise sources within a pixel are largely down to how well a chip has been designed and how capable the manufacturing process used to make it is.

I'm pretty sure this article doesn't talk about pixel size (because it's only talking about the effects on the whole image).

Larger pixels do produce less noise at the pixel level, precisely for the reason you state. However, as soon as you consider the whole image, by viewing images at the same size, pixel size [doesn't make much difference](https://www.dpreview.com/articles/5365920428/the-effect-of-pixel-and-sensor-sizes-on-noise/2).

Link | Posted on Feb 23, 2017 at 18:58 UTC
On article First Look: Fujinon MK18-55mm T2.9 cine lens (64 comments in total)
In reply to:

VictorTrasvina: Perhaps Fuji might consider giving us an affordable S35 4K Video Camera? Like the GH5? And if you could already use your primes ?? Well :)

I don't think high-end video cameras tend to have IBIS and I'm pretty sure Fujifilm has said the X-mount wasn't designed with enough space to allow it (I could be misremembering this: it could be that they've not designed their lenses to have larger-than APS-C coverage), so I wouldn't bet too much money on that.

Not sure what crop factor you're referring to.

Link | Posted on Feb 23, 2017 at 00:59 UTC
On article First Look: Fujinon MK18-55mm T2.9 cine lens (64 comments in total)
In reply to:

VictorTrasvina: Perhaps Fuji might consider giving us an affordable S35 4K Video Camera? Like the GH5? And if you could already use your primes ?? Well :)

It's pure speculation, of course, but I doubt Fujifilm has developed the X-mount version of this lens solely with X-T2 owners in mind.

Link | Posted on Feb 23, 2017 at 00:10 UTC
On article First Look: Fujinon MK18-55mm T2.9 cine lens (64 comments in total)
In reply to:

MrBrightSide: I understand the motivation for producing this lens, but it's still it's ironic that in this era of evermore precise touch-to-focus and electronic Aperture controls that there is still such a high demand for ultra low-tech lens technology.

Kharan - it's certainly possible to make focus moves programmable: the Panasonic GH5 allows you to pre-define three focus distances and a tap on the screen prompts it to rack to whichever point you choose. There are five speed options for how rapidly in makes those changes, too.

However, that's one feature on one camera. This lens is designed for the way the majority of videographers expect (and have learned) to shoot, today. It may not be the most futuristic response, but it's far from anachronistic, just yet.

Link | Posted on Feb 23, 2017 at 00:09 UTC
On article Leica SL Review (774 comments in total)
In reply to:

marike6: Review is much too short for a Leica camera, and the snarky "Conclusion" section, the bit about "if you have your heart set on a camera with a red dot on it" is kind of obnoxious.

Lan - that's useful feedback. I can't guarantee we'll go back to listing the full menu system (they take a long time to write and are not widely read), but there certainly will be reviews that have more extensive coverage of the controls.

Link | Posted on Feb 23, 2017 at 00:01 UTC
On article Leica SL Review (774 comments in total)
In reply to:

marike6: Review is much too short for a Leica camera, and the snarky "Conclusion" section, the bit about "if you have your heart set on a camera with a red dot on it" is kind of obnoxious.

We're trying to make *all* reviews shorter: it's not a brand-by-brand decision.

We want to present all our existing testing and experience in the most digestible way possible. I'd like to think the quality and depth of the review isn't measured in pages (it's easy to write lots, it's much harder to condense lots of findings into a manageable package).

Link | Posted on Feb 22, 2017 at 20:04 UTC
On article Leica SL Review (774 comments in total)
In reply to:

Peter von Reichenberg: Well...one would excpect little more, right? Especially from Leica with its price tag and reputation. OK, let's see if reddotted fanclub will buy enough of them. Instead of being a trendsetter Leica just hardly tries to compete to Sony...disappointing. On the other hand - whatta shame for CaNikon losing a chance to bring second full frame mirrorless after Sony. I don't see anything special in these pictures many other cameras can't do. For me the biggest WOW goes to Fuji GFX 50S.

That's a little unfair. The SL is a beautifully engineered camera (in line with the company's reputation), with an innovative interface (which is uncharacteristically trend-setting), very good autofocus and a pretty good sensor. In almost every respect it's the most technically impressive digital camera the company has ever made.

Link | Posted on Feb 22, 2017 at 19:51 UTC
On article Leica SL Review (774 comments in total)
In reply to:

photoMEETING: "this is the first non-rangefinder 35mm full-frame digital camera Leica has made, and the company's first full frame mirrorless camera in the modern sense."

Both is true for the Leica Q, not the Leica SL.

It's not a rangefinder or a mirrorless (interchangeable lens camera). I'm pretty sure we haven't implied otherwise.

And I'm *really* not going to get into a discussion about whether we should retrospectively apply the term 'mirrorless' to rangefinders.

Link | Posted on Feb 22, 2017 at 19:50 UTC
On article Leica SL Review (774 comments in total)
In reply to:

photoMEETING: "this is the first non-rangefinder 35mm full-frame digital camera Leica has made, and the company's first full frame mirrorless camera in the modern sense."

Both is true for the Leica Q, not the Leica SL.

Nobody called compact cameras 'mirrorless,' since there were no significant numbers of fixed lens DSLRs to distinguish them from. We started using the term mirrorless solely as a less clunky synonym for 'mirrorless interchangeable lens camera.' It's not precise, it's not ideal but I've not seen significant confusion being caused.

Link | Posted on Feb 22, 2017 at 18:55 UTC
On article Sense and Sensitivity (5 comments in total)
In reply to:

Ruy Penalva: That was a good manner to not explain clearly what ISO means. Better, and simply, would be to say that ISO is the boost the software gives on the sensor to fit adequately aperture and shutter speed chosen in such a way as to result in an usable shot.

Such a table would be impossible. Industry body CIPA dictates that its members use either SOS or (for multi-area metering mode) REI, neither of which has a fixed amplification/brightness relationship.

All modern cameras that I know of are able to create their ISO settings using any combination of amplification and tone curve to produce the desired output brightness. For instance, Canon's Highlight Tone Priority mode uses one step less hardware amplification to produce each of its ISO settings, compared with standard mode.

Link | Posted on Feb 22, 2017 at 16:19 UTC
On article Leica SL Review (774 comments in total)
In reply to:

photoMEETING: "this is the first non-rangefinder 35mm full-frame digital camera Leica has made, and the company's first full frame mirrorless camera in the modern sense."

Both is true for the Leica Q, not the Leica SL.

We've always used 'Mirrorless' as shorthand for 'Mirrorless Interchangeable Lens Camera,' so, while the Q doesn't have and mirrors, we wouldn't consider it to be mirrorless.

Link | Posted on Feb 22, 2017 at 16:07 UTC
In reply to:

Ebrahim Saadawi: Such great lenses with great prices. So bumped I can't use them on my 5DIV or anything other than the restrictive E mount.

They do state E-mount/short flange distance "enable the compact, lightweight design" so these lenses would be considerablly different if made for DSLRs, is the most likely case scenario.

Still. A bit sad. Fujinon Zooms for Still glass prices! and it's f/2.8 constant. Bummer.

The flange distance adjustment is for tiny calibration corrections to ensure the lens is perfectly focusing on the sensor plane. It's not going to correct for the >20mm difference between the E-mount flange-back distance and that of an SLR mount.

Link | Posted on Feb 22, 2017 at 09:11 UTC
On article First Look: Fujinon MK18-55mm T2.9 cine lens (64 comments in total)
In reply to:

bokesan: "The lens also includes precise distance marks." - I wonder what they do about the variance in E-mount flange distance. Up to 0.1mm difference seems to be in spec, and I've seen notably different infinity focus positions with the same lens on different a7 bodies. Shimming?

I think it's slide 10 that describes how the back of the lens can be adjusted, relative to the sensor, to ensure correct back focus.

Link | Posted on Feb 22, 2017 at 08:56 UTC
Total: 4520, showing: 1 – 20
« First‹ Previous12345Next ›Last »