Richard Butler

Richard Butler

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

Comments

Total: 5294, showing: 81 – 100
« First‹ Previous34567Next ›Last »
In reply to:

Thomas Traub: part 2 of my post: Or in other words: if there is a szene with a very wide dynamic range (very very bright and very very dark areas) with - let's say 2^20 light-values (EV) and a sensor is able to capture 15 of the 20 light-values (let's say from value 3 to value 18 only for example to understand) than this information = this dynamic range can be saved as a 8-bit, a 12-bit or a 14-bit file. EVERY FILE CAN DESCRIBE THE BRIGHTES AND THE DARKEST PIXEL CORRECTLY!!! AND IT DOES NOT MATTER WHETER IT IS A 8-BIT, a 12-BIT OR AN 14-BIT-FILE!
The 14-bit file only has more information than the 12 bit file and this more information than the 8-bit-file BETWEEN the darkest and the brightest pixel the sensor can capture! (that's why we prefer for post-processing more bit-deep or more information)
THE BIT-DEEP OF THE FILE ITSELT SAYS NOTHING ABOUT THE BRIGHTEST AND THE DARKEST PIXEL A SENSOR CAN CAPTURE AND NOTHING ABOUT THE DYNAMIC RANGE OF A CAMERA!!!

Dynamic range is the brightness range you can properly describe, between the brightest tone and the noise level you find useable (SNR = 1 is the engineering cut-off).

In a linear file, the Raw values are not distributed the same way that they are in a processed file, such as a JPEG. The upshot is that bit depth mainly ends up being a measure of how much precision you can describe the darkest couple of stops of your capture with.

If you run out of precision, the imprecision of the capture of that data becomes *quantization error*: itself a type of noise. In other words, your lack of means to correctly describe the shadows becomes your DR cut-off.

Link | Posted on Sep 1, 2017 at 17:41 UTC
In reply to:

steelhead3: I have a question about Sony's 11/7 compression....does it affect DR being only 11 bits or is the native 12 or 14 bit retained for DR purposes?

That's an interesting question. The key thing is to recognise the role of [both steps in the compression process](https://www.dpreview.com/articles/2834066212).

The first step is to apply a compression curve to the data, which drops it from the 14-bit signal coming from the sensor down to 11-bits. Most of the information thrown away is in the highlights where, in a linear capture, you have a vast excess of precision.

In the highlights, although the signal to noise ratio is excellent, the [absolute level of shot noise](https://www.dpreview.com/articles/8189925268) is highest. This means you can throw away a lot of this data without it mattering (what Nikon does with its 'visually lossless' compression, I think).

So this first drop to 11-bits has little impact, I believe. It's the localised compression and the drop to 7-bit deltas (difference values), that causes the problems. As such, you get a reduction in DR in high-contrast regions.

That would be my understanding, anyway.

Link | Posted on Sep 1, 2017 at 17:32 UTC
In reply to:

krikman: Article is total misunderstandig of light, mathemathics and photography concept.
Best possible dynamic rangre achieved with 1-bit image. There is absolute light and absolute darkness.
Adding more bit depths to image only increases number of possible colors between Blacks and whites of given picture, i.e number of colors and tones.

Raw capture is linear, so the values in the file (number of tones) are not distributed the way they would be in a processed file.

Adding an additional bit means every stop in you image is described with twice the precision, but this only makes a significant difference in the shadows, where you don't already have a massive excess of precision.

Link | Posted on Sep 1, 2017 at 17:22 UTC
In reply to:

photoMEETING: I can't follow the line of arguments in that article.

DR is depending on two factors:
1. Full well capacity of the sensor [e-]
2. Readout noise of the A/D electronics [e-]

There is a formula to calculate DR from this two values.

Nothing about bit depth so far. :)

Another story is how many distinctive brightness levels are beeing saved in that given physical range of the sensor. The more bits per pixel you have, the more different shades of gray you can differentiate.

I can easily imagine an image of a sensor with a DR of only 8EV, but an 16 bit output with a rich differentiation of the brightness shades.

There can also be an image with a 14EV DR, but an 8bit output only. Many 8-bit-JPGs, developed from a ISO 100 NEF file out of a Nikon D750, are pretty good examples!

Probably, the artice has to be revised.

Yes, DR is determined by full well capacity (the number of electrons that can be accumulated before clipping) and read noise (the noise floor of 'hmm' coming from the camera's circuitry).

What the article says is:

"bit depth is primarily about how much of your camera's captured dynamic range can be ***retained***."

It doesn't say the bit depth determines your dynamic range, just that it acts as a cut-off. The end of the article, where it discusses the use of 14 bit files on cameras with sub-12EV DR wouldn't make sense if I was claiming bit depth *determines* DR.

And, re gradation, please read the section about the implications of linear capture. In a processed file, your description would be absolutely correct, but in a Raw file with linear data, it doesn't work that way.

Link | Posted on Sep 1, 2017 at 17:06 UTC
In reply to:

mpgxsvcd: How does BIT depth increase Dynamic Range if it is not capturing any more data?

I'm not saying but depth *increases* dynamic range, I'm saying it increases your capacity to retain it, **if** your sensor is capturing it in the first place.

The thing I was trying to make clear in the last couple of paragraphs was that, if your sensor doesn't capture more than 12EV of DR, and additional bit depth is pointless.

Link | Posted on Sep 1, 2017 at 15:46 UTC
In reply to:

Petr Klapper: Theory sounds good and then there's a practical test:

https://photographylife.com/14-bit-vs-12-bit-raw

Of course 14bit gives you peace of mind that you're doing maximum possible, but even though the explanation above sounds dramatic, the real difference is still questionable even in extreme examples.

That article illustrates the point I was trying to make rather well, I think: there'll be little difference unless your scene exceeds 12 stops *and* you then process in such a way that you can see the very deepest shadows.

Link | Posted on Sep 1, 2017 at 15:34 UTC
On article Olympus OM-D E-M10 Mark III Review (560 comments in total)
In reply to:

Glen Barrington: Hmm! I'm impressed by the specs, and the thought of a new E-M10 appeals to me. But I didn't particularly like the jpg images. They seemed a little flat, washed out and not that sharp as a rule. I'm not too upset about it being a 16mp sensor, as I think the current 16mp sensor is pretty good overall, and I like the results I get from my 1st gen E-m10.

I like the idea of an improved 5 axis IS system very much.

I will keep an eye on this camera, but I don't think I will become an early adopter.

It's one of the [pork pies](https://en.wikipedia.org/wiki/Pork_pie) I made recently.

Versions with the Noise Filter turned off and sharpening increased to +1 have now been added.

Link | Posted on Aug 31, 2017 at 22:57 UTC
On article Olympus OM-D E-M10 Mark III Review (560 comments in total)
In reply to:

Brianroger: Nice shot of the pork pie and English mustard...

To which I respond: Beer Cheese.

Link | Posted on Aug 31, 2017 at 22:32 UTC
In reply to:

pharles: As an old guy who has in the past shot 135, 120 and 4X5, the whole issue of differences in DoF, grain (noise), IQ between the formats was instinctive. I knew that when I got the 4X5 camera out, i was going to have to use tilts to make up for the shallow DoF of the format. Lenses for the format were f4.5 or f5.6 at their widest. We see today the lenses built for APS-C sensors usually have slightly wider maximum apertures compared to full frame.

So Richard, when you're talking "equivalence" are you simply trying to update a concept, that us old film using geezers knew instinctively, for the younger, digital only crowd? Or is there more to it that I am missing?

"Pumping up the ISO in low light to maintain DoF with shutter speed and all of a sudden, we're not better than an APS-C setup."

Precisely! At any point where two systems (again: sensor size + aperture), can be set to equivalent settings, there is no benefit to the larger format.

If you match depth-of-field, angle of view and shutter speed, you'll be using matching aperture diameters on both systems. You'll get the same total light during the exposure and (differences in sensor performance aside), you'll get very, very similar images.

Equivalence is not about big sensors being better: it simply helps show if/when a larger sensor *and* larger physical aperture allow access to additional light/better IQ.

Link | Posted on Aug 31, 2017 at 22:20 UTC
On article Olympus OM-D E-M10 Mark III Review (560 comments in total)
In reply to:

Glen Barrington: Hmm! I'm impressed by the specs, and the thought of a new E-M10 appeals to me. But I didn't particularly like the jpg images. They seemed a little flat, washed out and not that sharp as a rule. I'm not too upset about it being a 16mp sensor, as I think the current 16mp sensor is pretty good overall, and I like the results I get from my 1st gen E-m10.

I like the idea of an improved 5 axis IS system very much.

I will keep an eye on this camera, but I don't think I will become an early adopter.

I believe it's a newer processor (hence the ability to capture 4K), but if there are any shots in the gallery you'd like to see with the Noise Filter turned down, I'd be happy to re-process and upload them.

Link | Posted on Aug 31, 2017 at 20:42 UTC
In reply to:

pharles: As an old guy who has in the past shot 135, 120 and 4X5, the whole issue of differences in DoF, grain (noise), IQ between the formats was instinctive. I knew that when I got the 4X5 camera out, i was going to have to use tilts to make up for the shallow DoF of the format. Lenses for the format were f4.5 or f5.6 at their widest. We see today the lenses built for APS-C sensors usually have slightly wider maximum apertures compared to full frame.

So Richard, when you're talking "equivalence" are you simply trying to update a concept, that us old film using geezers knew instinctively, for the younger, digital only crowd? Or is there more to it that I am missing?

Essentially, yes.

It's simply a way of getting a solid idea of the magnitude of difference you should expect between systems (sensor size + aperture), for those who haven't used multiple formats over multiple years, to the extent it feels instinctive.

In doing so, it also undermines the widely held misconception that smaller formats have a deep depth-of-field advantage.

For instance, it allows an easy way to tell whether a wide maximum aperture lens for APS-C might offer enough capability to avoid needing to replace your whole system with full frame.

Link | Posted on Aug 31, 2017 at 18:47 UTC
On article Action-packed: Sony a6500 review (1181 comments in total)
In reply to:

RubberDials: 'the lack of a second top plate control dial on a $1400 MSRP camera is hard to stomach.'

The Canon 1DxII has only one top plate control dial and it costs 6 grand. The Sony system is identical to Canon - one wheel on top and a thumbwheel at the back and yet Sony gets a sledging and I have never heard Canon criticised once in all the years they've been doing it. An example of some of the irrational thinking that underpins this review.

The style is reminiscent of the A7II test, where the archetype that the camera was being judged against shifted depending on the feature being assessed. Here we get lots of criticism on the handling as if the camera were being compared to a DSLR, then the claim that the reviewer is not impressed by the IBIS, where the camera is suddenly compared to Micro 4/3.

The reviewer knows he's being unfair, because he writes: 'Of course both the cameras mentioned use smaller sensors, which in theory should be easier to move around.'

The Canon system is not the same as Sony's. The key distinction being that both of the a6500's dials have to be controlled with the same digit. We'd criticise it on any brand, especially at this price.

The reviewer doesn't 'know he's being unfair,' he's acknowledging that it would be harder for Sony to achieve, to give context to the criticism. It's *entirely* reasonable to compare a camera to the best performance available in its peers: if you need IS, we should point out that you get better results from a Micro Four Thirds camera, if you want direct access ergonomics, you will find it in most rivals (not just DSLRs) at this price.

Given how often I read comments convinced we're being paid-off by Sony, it's entertaining to be accused of anti-Sony rhetoric. We're a review site, we test and use cameras and tell people what we find. If the criticism isn't relevant to you, discount it, but don't expect us to ignore what we believe to be shortcomings.

Link | Posted on Aug 31, 2017 at 18:40 UTC
On article Olympus OM-D E-M10 Mark III Review (560 comments in total)
In reply to:

Michiel953: Nice Dekerf bike! Is it Richard's?

It is.

This was the photo I mentioned. The forks are somewhat unusual (I built it assuming I'd swap them for some Enves or similar) but they're really comfortable on the dreadfully maintained roads of Seattle.

Link | Posted on Aug 31, 2017 at 18:13 UTC
On article Olympus OM-D E-M10 Mark III Review (560 comments in total)
In reply to:

Hellraiser: These m43 camera makers should really make their sensors go below ISO200, compared to modern APS-C these images (especially landscape) seem really really bad :S

Halving the base ISO would require you to significantly increase the full well capacity, which would mean a significantly different chip (and a design that outperforms all its peers).

Well, unless you lowered efficiency, but that doesn't help you (you'd get lower base ISO and need the extra light just to match the IQ of this sensor).

deep7: [ISO](https://www.iso.org/about-us.html) is not an acronym, so you don't add dots between the letters.

Link | Posted on Aug 31, 2017 at 17:05 UTC
On article Olympus OM-D E-M10 Mark III Review (560 comments in total)
In reply to:

Bambi24: Can we stop pretending that colors or white balance in JPEG don't matter because "you can shoot RAW".

Do people not realize how many people actually shoot JPEG? Hint...the majority does. Everyone from casual shooters, to people who simply prefer JPEG, to news journalists not allowed to tinker with images, to sports shooters and wildlife shooters wanting access to the deepest buffer, shoot JPEG. RAW is not a solution to issues with JPEG. Canon can get JPEG right, many other camera keep messing it up, that is on them, it's not the fault because you shoot JPEG.

Not everyone has the time or desire to process thousands of images in Lightroom. Most people simply do-not-care, they have other things to do, more imporant or more enjoyable things. Many people do not even like the look of processed images.

A camera that can't get JPEG right is a bad camera for a lot of people. Having accurate JPEG colors and access to picture profiles is incredibly important for JPEG shooters.

We're not pretending it doesn't matter, otherwise we wouldn't have mentioned the degree to which it undermines the very idea of Auto mode.

We're just saying there it's not a lost cause, if the camera ends up applying circus-tastic levels of saturation: you can, in about three button presses, generate a less mad JPEG. But no, that doesn't negate the problem.

Link | Posted on Aug 31, 2017 at 17:00 UTC
On article Olympus OM-D E-M10 Mark III Review (560 comments in total)
In reply to:

FuhTeng: "Moving the mode dial from Auto to Art, I noticed the E-M10 III has another trick up its sleeve; you can now scroll through every Art filter on the camera along the bottom of the screen while your scene changes in real time." - congratulations I guess? This hasn't always been part of Oly's cameras where you change filter and it changes as you scroll? I like being able to go through the "filters" (or creatives styles on my Sonys) in real time and I've been able to do that for years.

Not in the 'ART' position on the mode dial.

In PASM modes of the last couple of generations of Olympuses (since the Art Filters were added to the list of Picture Modes), you could, if you had 'Live Control' as your control panel, but Carey's writing about the first-time user experience.

On the E-M10 III, when you select the 'ART' position on the mode dial, all your filter options are arrayed across the bottom of the screen, just like Instagram (probably not coincidentally).

Link | Posted on Aug 31, 2017 at 16:58 UTC
On article Olympus OM-D E-M10 Mark III Review (560 comments in total)
In reply to:

Brianroger: Nice shot of the pork pie and English mustard...

Thanks. Not a common occurrence in Seattle.

Link | Posted on Aug 31, 2017 at 15:34 UTC
On article Olympus OM-D E-M10 Mark III Review (560 comments in total)
In reply to:

h2k: Not sure if the article's concentration on full Auto serves the target group well.

Would have thought the target group uses more of Aperture Priority.

The shooting experience is focused on Auto, AP and Scene modes because some of the biggest changes made, relative to the Mark II, were in making these modes better and more accessible.

The conclusion of our full review will reflect both this and the perspective of people using the rest of the controls more extensively.

Link | Posted on Aug 31, 2017 at 15:30 UTC
In reply to:

mpgxsvcd: Is it a requirement that the card be a UHS-II card for the 400 MB/sec update? For example if you had a V60 UHS-I card, which is theoretically possible, would it be able to utilize the full write speed capabilities of the card in the GH5?

Personally (and this is primarily based on what Panasonic told us about the testing they've done), I wouldn't buy a card until that logo is on there. It *might* work but I wouldn't be confident enough to part with my cash. The V60 specs have existed for a while now, yet we've only seen a couple of cards wearing the logo.

Link | Posted on Aug 30, 2017 at 22:34 UTC
On article Leica TL2 first impressions (382 comments in total)
In reply to:

drummercam: "The camera's main settings have been split into nine sections". -- One of the 9 sections in that photo is "flash".

Just above that picture: "Meanwhile, the pop-up flash, which owners apparently said they didn't need, has gone".

Why is there a flash management section if no flash?

It's for when you've got something mounted on the hotshoe.

Link | Posted on Aug 30, 2017 at 19:27 UTC
Total: 5294, showing: 81 – 100
« First‹ Previous34567Next ›Last »