Richard Butler

Richard Butler

DPReview Administrator
Lives in United Kingdom Seattle, United Kingdom
Works as a Technical Editor
Joined on Nov 7, 2007
About me:

Richard graduated as a scientist but had a lot more fun writing and shooting for his university magazine. A number of years spent variously as a reporter, writer and editor on science and engineering titles combined his knowledge of science with his interest in images and words. But it was spotting the connections between emission spectra, white balance and all the nonsense he'd taught himself playing around in Photoshop that helped kindle an interest in digital photography. Searching for a camera led to him discovering DPReview and Richard was recruited by Phil Askey in 2007. He's been combining his love of photography, communication and attention to detail (pedantry?) ever since.

He has unusually strong opinions about lenses for the APS-C format.

Comments

Total: 7068, showing: 1 – 20
« First‹ Previous12345Next ›Last »
On article Panasonic Lumix S1/S1R added to buying guides (26 comments in total)
In reply to:

anticipation_of: I'm sure you're well aware of this, but it feels like there's an unusually large review backlog right now. Lots of partial reviews but not a lot of finished ones. And the time gap between cameras being announced/released and cameras having their full reviews posted seems like it's gotten longer too. Postings like this one feel like a bit of a tease at the moment.

The goal is to reassure readers (such as the OP here) that we're working on the reviews, rather than sitting silently until they completely finished. It's really that simple.

Link | Posted on Mar 22, 2019 at 22:56 UTC
On article Fujifilm GFX 50R Review (1709 comments in total)
In reply to:

mostlyboringphotog: "Thankfully, above ISO 1600, the GFX 50S applies no additional amplification,..."
Not sure what is there to be thankful of?
If amplified ISO 6400 blows high light too much, shoot in manual with ISO1600.
If scene was not backlit and ISO 6400 is not blowing highlight, by not amplifying the signal, two stops of SQNR is being lost.
X-T3 applies amplification to ISO25600; what should we do? Be unthankful?

My understanding is that it would: noise inherent in the system would reduce the value of capturing it with greater precision. There's not necessarily a benefit in using 8192 values to characterise the brightest stop of data, vs using 2048, if there's lot of variance in the signal you're trying to encode.

But this is not my area of expertise, so I'll defer to the Sci/Tech forum on this.

Link | Posted on Mar 22, 2019 at 21:46 UTC
On article Fujifilm GFX 50R Review (1709 comments in total)
In reply to:

mostlyboringphotog: "Thankfully, above ISO 1600, the GFX 50S applies no additional amplification,..."
Not sure what is there to be thankful of?
If amplified ISO 6400 blows high light too much, shoot in manual with ISO1600.
If scene was not backlit and ISO 6400 is not blowing highlight, by not amplifying the signal, two stops of SQNR is being lost.
X-T3 applies amplification to ISO25600; what should we do? Be unthankful?

ie: if we assume F4 was needed for sufficient DoF and 1/250th was needed for movement-freezing purposes, then ISO 6400 may well be the 'correct' ISO in terms of giving a JPEG with middle grey in the correct place. This particular scene requires more highlight retention than the standard JPEG, so what's the correct response?

In a camera that adds more analog gain with each ISO step, you'd need to drop the ISO to reduce gain or reduce the exposure (which would increase photon shot noise). As an alternative, many cameras have a DR mode that reduces the gain/ISO relationship by one stop (or one/two stops in the case of Fujifilm), then has a tone curve to incorporate the additional highlights into the JPEG. In fact setting the Fujifilm to ISO 6400, DR400% would have achieved pretty much what our processed image does, out-of-camera.

Link | Posted on Mar 22, 2019 at 18:59 UTC
On article Fujifilm GFX 50R Review (1709 comments in total)
In reply to:

mostlyboringphotog: "Thankfully, above ISO 1600, the GFX 50S applies no additional amplification,..."
Not sure what is there to be thankful of?
If amplified ISO 6400 blows high light too much, shoot in manual with ISO1600.
If scene was not backlit and ISO 6400 is not blowing highlight, by not amplifying the signal, two stops of SQNR is being lost.
X-T3 applies amplification to ISO25600; what should we do? Be unthankful?

Everything I've read and been told suggests that quantization noise isn't a problem until you get down to the last few bits. The magnitude of photon shot noise is greater than the sampling precision, so you're oversampling most of the tones of your image (this is how Nikon is able to lossily compress its Raws without any visible loss of data: the highlights are so monstrously oversampled that you can compress it down a lot without any meaningful loss).

I think your 6dB figure would only apply to a perfect signal, rather than one that's inherently noisy. The Sci/Tech forum will be able to discuss this in much more detail than I can and I'd be interested to see their responses.

The ISO standard is based on middle grey rendering, so it's always a stumbling block when trying to discuss ETTR (which is the implication of what you're saying), and what 'correct' exposure is. That's essentially a separate issue...

Link | Posted on Mar 22, 2019 at 18:49 UTC
On article What is equivalence and why should I care? (2496 comments in total)
In reply to:

TheTomx: Sorry mate wrong - either you really don't understand sensor tech or you just stick to old rules.
based on this one an old mobile and a new mobile should have the same DOF with f1.8 but guess what they don't.
why ?

1. Pixel Tech changed - modern sensors have layers.
2. 20MP FF and 50MP FF don't have the same light exposure (that's why A7R vs A7II) have different ISO limitations
3. F is only one factor - what about how the lense is built. ?

There are ways of trying to address and work around the lower per-image light capture of small sensors (such as taking multiple exposures, as you say), but this is addressing the very basic 'how much light does your sensor get to create an image' and that's not inherently changed.

Link | Posted on Mar 22, 2019 at 17:40 UTC
On article Fujifilm GFX 50R Review (1709 comments in total)
In reply to:

mostlyboringphotog: "Thankfully, above ISO 1600, the GFX 50S applies no additional amplification,..."
Not sure what is there to be thankful of?
If amplified ISO 6400 blows high light too much, shoot in manual with ISO1600.
If scene was not backlit and ISO 6400 is not blowing highlight, by not amplifying the signal, two stops of SQNR is being lost.
X-T3 applies amplification to ISO25600; what should we do? Be unthankful?

If you shot ISO 1600 manually (using the same exposure values as the metered ISO 6400 exposure), you'd get the same Raw data but a non-usable JPEG. That seems less useful.

The reason this has been done is that there is no further SNR benefit to applying more amplification: you'd just be pushing more captured data to clipping without any benefit for doing so.

Link | Posted on Mar 22, 2019 at 17:24 UTC
On article Panasonic Lumix S1/S1R added to buying guides (26 comments in total)
In reply to:

anticipation_of: I'm sure you're well aware of this, but it feels like there's an unusually large review backlog right now. Lots of partial reviews but not a lot of finished ones. And the time gap between cameras being announced/released and cameras having their full reviews posted seems like it's gotten longer too. Postings like this one feel like a bit of a tease at the moment.

We're aware of the problem. The problem is when so many cameras get launched at the same time (EOS RP, S1, S1R, E-M1X, a6400, GR III. X-T30, Q2). That's significantly more cameras than we have camera reviewers and each camera takes time to review, even once we get a final-spec model in.

We're working on three of those cameras and hope to publish reviews on them as soon as we can so that we can get to the others, but we don't have final, reviewable versions of them all yet.

Link | Posted on Mar 22, 2019 at 17:12 UTC
On article Canon EOS RP review in progress (645 comments in total)
In reply to:

keepreal: I have no interest in mirrorless as I prefer an OVF but still follow articles to learn where the leading digital equipment is going.

This is the first full frame mirrorless camera that to me makes sense. Its weight is very reasonable and relatively speaking, so is the price. Perhaps this is the first where we are going back to the bulk of 35mm film cameras, if so very welcome.

Unfortunately this is undermined by the few lenses for it, probably most that will follow. The 24-70mm and 24-105mm zooms are too big and heavy. Why cannot they be only slightly bigger and heavier than equivalent lenses of the film era? A focus motor need not affect that much at all. And finally, why cannot the designs avoid distortion like they used to? Software correction is a poor alternative to that, limiting you to which software is suitable.

If you correct an aberration with glass, you risk introducing other aberrations: lens design is a series of trade-offs (it's not as simple as: more glass = better correction but higher price).

Software correction can alleviate (or at least, provide more options for addressing) some of those trade-offs. So yes, sometimes it means you can make a simpler, lighter, maybe cheaper optical design. But it also means that by under-correcting things like geometric distortion (which can be digitally corrected with minimal cost to IQ) you can then concentrate on correcting other aberrations, potentially giving a better result overall.

It means 3rd party s/w makers have to work harder to fully support a system, but should everyone be denied potentially better lenses to make their lives easier?

Try looking at it back-to-front: the SLR design arguably limits the options open to lens designers because they have to optically correct the preview, even though that may not give the best final image.

Link | Posted on Mar 22, 2019 at 16:58 UTC
On article Canon EOS RP review in progress (645 comments in total)
In reply to:

marc petzold: DPReview should also show cheaper alternatives, DSLMs, which means clearly: Sony A7 Mark II here! :)

A fine evening, and good light.

Hi marc - You pointed out that the a7 II has the same viewfinder, but is older. I was just trying to highlight that the mark III still uses the same panel. I was trying to add to what you were saying, not contradict it.

Link | Posted on Mar 21, 2019 at 16:21 UTC
On article Canon EOS RP review in progress (645 comments in total)
In reply to:

chavim: Richard, the Canon page lists this camera as having weather-sealing.

Can you clarify why DP Review mentions lack of weather-sealing?

Thanks

I'll ask Canon.

Link | Posted on Mar 21, 2019 at 16:12 UTC
On article Canon EOS RP review in progress (645 comments in total)
In reply to:

Arun H: The 6D mark II viewfinder has 0.98% coverage?

And 71x magnification: you just get to see a tiny fraction of the scene in incredible detail.

Sorry, that's exactly the kind of typo I make and then just can't see.

Link | Posted on Mar 21, 2019 at 00:45 UTC
On article Canon EOS RP review in progress (645 comments in total)
In reply to:

marc petzold: DPReview should also show cheaper alternatives, DSLMs, which means clearly: Sony A7 Mark II here! :)

A fine evening, and good light.

T3 - The RX100 series (excluding the VI) might be a better example. The three models in the a6x00 series were always separated by price, not adjusted later to make room for the newer ones (a6400 directly replaces a6300, according to Sony).

marc - We completely agree about there being a mismatch between the RP and the lenses currently offered. Your concerns about UI very much reinforces what I mean about the RP and a7 II being suited to different groups of people (only one of which is likely to use MF adapted lenses).

I still don't see the appeal of uncompressed Raw over the 14-bit lossless compression that the RP offers, though. It has a lossy option, too, but you don't have to use it. The RP, a7 II and a7 III all have 2.36M dot OLED viewfinders, by the way.

Link | Posted on Mar 20, 2019 at 23:14 UTC
On article Canon EOS RP review in progress (645 comments in total)
In reply to:

marc petzold: DPReview should also show cheaper alternatives, DSLMs, which means clearly: Sony A7 Mark II here! :)

A fine evening, and good light.

marc - I don't think our positions are very far apart at all. I wrote that the RP is a collection of familiar components (it *is* the 6D II sensor, though I wouldn't expect a meaningful drop in DR).

I also agree that the RP's video (and its battery) specs are "pretty low by contemporary standards."

The a7 II's Raw performance *will* be better than the RP, especially in terms of dynamic range at low ISO (though you need to turn Sony's compression off if you plan to make use of that). When I say the a7 II looks dated, I mean compared with the Mk III: no touchscreen, no 4K, the old, too-small battery...

But yes, the RP has the much more beginner friendly UI (that's why I say RP users would hate the a7 II and *vice versa*).

We also clearly agree that the lack of affordable lenses means the RP isn't the bargain it's intended to be. But given it was already launched for $400 less than any full-frame camera before it, we'll have to disagree that it should have been 40% cheaper again.

Link | Posted on Mar 20, 2019 at 21:07 UTC
On article 2019 Buying Guide: Best cameras under $1500 (114 comments in total)
In reply to:

MikeStern: X-T3 can be bought less than 1500$???
Then you make a pinhole lens from a milk carton?

I find it stupid to list this. You simply can’t make it work without at least a kit lens. Imagine buying a TV without a power cord that costs a good chunk.

I'll delete it if you like, but there's no need.

There's no perfect way of doing this and it's useful for us to understand how people are interpreting it, so that we can try to improve things next time 'round.

Link | Posted on Mar 20, 2019 at 20:54 UTC
On article Canon EOS RP review in progress (645 comments in total)
In reply to:

0lf: bottom of page 3, ISO and connectivity : “and an option to have the camera choose that minimum shutter speed automatically. If you choose the last option, you can bias it either faster or slower than the default option, which is 1/shutter speed.”

1/FL

That would make more sense, I agree.

Link | Posted on Mar 20, 2019 at 19:46 UTC
On article 2019 Buying Guide: Best cameras under $1500 (114 comments in total)
In reply to:

MikeStern: X-T3 can be bought less than 1500$???
Then you make a pinhole lens from a milk carton?

I find it stupid to list this. You simply can’t make it work without at least a kit lens. Imagine buying a TV without a power cord that costs a good chunk.

This buying guide is based on the body-only price. When we considered factoring-in a kit lens, it just confused matters (some bodies aren't sold with specific kit lenses, and a brand with a better/more expensive kit lens ends up with its cameras looking poor compared with higher-end bodies bundled with cheap kit zooms).

Once that decision was made, one camera will always be the most expensive and will strain at the edges of the price category.

However you divide up a continuum, the examples on the boundaries draw attention to the challenges of doing so.

Link | Posted on Mar 20, 2019 at 19:12 UTC
On article Canon EOS RP review in progress (645 comments in total)
In reply to:

marc petzold: DPReview should also show cheaper alternatives, DSLMs, which means clearly: Sony A7 Mark II here! :)

A fine evening, and good light.

Just to clarify: it's included in the comparison because it's now cheaper but its MSRP is listed to make clear that it was originally meant to be a more expensive camera (though I was rather surprised, putting the comparison table together, how dated its specs now are).

I'd reiterate what we said in the text: anyone for whom the RP is a good fit will probably have a pretty miserable time with the a7 II. And vice versa, except that, in my own experience, anyone looking for a more involved camera would be better-off saving up for an a7 III, since it's *much* better than the a7 II.

I'd also suggest that Raw compression strategies are a red herring in this instance. Lossless compressed Raw (or even sensibly lossy compression) is usually preferable to uncompressed, which I'm pretty sure are the two options the EOS RP brings. But with these cameras it's the RP's sensor performance that's going to limit you in challenging situations, not the way its data is retained.

Link | Posted on Mar 20, 2019 at 19:02 UTC
On article Canon EOS RP review in progress (645 comments in total)
In reply to:

Swerky: In the studio test scene I noticed that the EOS RP crops are noticeably sharper than the ones of the 6D II! Then I remembered the RF 50mm f1.2 is mounted on the RP. In the exposure latitude and iso invariance tests, the RP is as noisy as the 6D II but the blacks remain black because the general colours shift towards green. On the 6D II it’s towards red and thus blacks turn to red.

The green/magenta shift tends to be a question of Raw profiling and the challenges of white-balancing very noisy data. I wouldn't read too much into that: just the noise level.

Link | Posted on Mar 20, 2019 at 18:15 UTC
On article Canon EOS RP review in progress (645 comments in total)
In reply to:

PhotoDiod: Carey and Richard, so you label a multitude of legitimate observations about the skewed tonality of EOS-R/P's skin color reproduction as "rumblings and grumblings about the color" and simply dismiss them with a highly scientific "whatever"? Way to test and review the camera!

The 'case closed' that you're inferring wasn't intended. We'll look to rephrase it.

Link | Posted on Mar 20, 2019 at 18:14 UTC
On article Canon EOS RP review in progress (645 comments in total)
In reply to:

marc petzold: DPReview should also show cheaper alternatives, DSLMs, which means clearly: Sony A7 Mark II here! :)

A fine evening, and good light.

I probably should have put a smiley on there.

Link | Posted on Mar 20, 2019 at 18:11 UTC
Total: 7068, showing: 1 – 20
« First‹ Previous12345Next ›Last »