Richard Butler

Richard Butler

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

Comments

Total: 4004, showing: 41 – 60
« First‹ Previous12345Next ›Last »
In reply to:

ThorstenMUC: Nice, that Sigma is more aiming for the APS-C mirrorless market... maybe there will be an X-mount version of this lens soon.
Ok - Fujifilm's lens lineup for APS-C is a much harder competitor, than Sony's few dedicated APS-C lenses - but since the optics are already there and it would "only" mean to exchange the mechanical mount (easy) and the firmware (that'l be the main effort) I would expect it's a viable step.
Competition is always welcome.

Part of the problem may be that Sony made the details of its mount (presumably including the AF protocols) available to third parties. To the best of my knowledge, Fujifilm hasn't done this, so Sigma would need to reverse-engineer the AF system, which I can imagine is a significant investment.

Link | Posted on Sep 13, 2016 at 18:15 UTC
In reply to:

great Javier: Ha ha, this lens got the gold award, but in the "Latest reviews list" is a Silver!

Yeah, that's a bug that's being investigated. It's a Gold award.

Link | Posted on Sep 12, 2016 at 21:57 UTC
In reply to:

zeratulmrye: Nice job! Looking forward to more lens reviews like this. It would be even better if you include M43/Fuji lenses into the comparison.

We certainly hope to. We'll be reviewing this lens for Micro Four Thirds, soon.

Link | Posted on Sep 12, 2016 at 19:29 UTC
On article What is equivalence and why should I care? (2222 comments in total)
In reply to:

Hansel300: /// "ISO and its role in clarifying and confusing
This means that, a Four Thirds camera with a 50mm f/2 lens at ISO100 should produce a JPEG of the same brightness as a Full frame camera with a 100mm f/2 lens at ISO100 and, set to the same F-number and shutter speed, even though its smaller sensor means it is receiving 1/4 as much total light."///

No its not ! A camera with lens - regardless of focal length - at F2 and ISO 100 will have the same shutter speed be it FF or M4/3's, and produce a similarly bright JPEG. Total light ? You're just confusing the issue ! Was this article written by Tony Northrup ?

Even a FF lens is over projecting and has a larger image circle to keep, therefore a FF camera doesn't capture "Total Light" projected by the lens - you see how stupid my point is ? THERE IS NO SUCH THING AS TOTAL LIGHT. A lens is just projecting an image over a sensor. DPR is talking about curtains, the more I open my curtains more "Total Light" will enter.

Thinking in terms of total light is contrary to the way that exposure is usually thought of: that's why the article only suggests using equivalence to understand the different potential capabilities of different systems and formats.

But, if you are willing to think about the size of the area available to capture the light, rather than just the light per unit area, you'll see a direct relationship between this and the performance differences you see between formats.

And no, it's not helped by online videos saying that manufacturers' F-numbers are misleading. That's nonsense. But total light is a demonstrably valuable concept if you want to understand the way different formats relate to one another.

Link | Posted on Sep 12, 2016 at 17:18 UTC
On article What is equivalence and why should I care? (2222 comments in total)
In reply to:

Hansel300: /// "ISO and its role in clarifying and confusing
This means that, a Four Thirds camera with a 50mm f/2 lens at ISO100 should produce a JPEG of the same brightness as a Full frame camera with a 100mm f/2 lens at ISO100 and, set to the same F-number and shutter speed, even though its smaller sensor means it is receiving 1/4 as much total light."///

No its not ! A camera with lens - regardless of focal length - at F2 and ISO 100 will have the same shutter speed be it FF or M4/3's, and produce a similarly bright JPEG. Total light ? You're just confusing the issue ! Was this article written by Tony Northrup ?

Even a FF lens is over projecting and has a larger image circle to keep, therefore a FF camera doesn't capture "Total Light" projected by the lens - you see how stupid my point is ? THERE IS NO SUCH THING AS TOTAL LIGHT. A lens is just projecting an image over a sensor. DPR is talking about curtains, the more I open my curtains more "Total Light" will enter.

F-numbers are useful specifically because any two lenses with the same F-number transmit (ignoring slight transmission losses), the same amount of light **per square mm**. And the ISO standard (or, at least that SOS bit) *by definition* means that the camera has to do whatever it takes to ensure that your JPEGs come out the correct brightness, regardless of whether you're using an iPhone or a Medium Format camera.

That's literally what the statement you quoted is saying.

A full frame F2 lens transmits the same light per sq mm, regardless of what sensor you stick it in front of. It doesn't spread the same amount of light out over a wider area - that's why you can mount that same lens on a smaller format and still get the same amount of light per sq mm.

However, the larger sensor captures more of the light being projected. This is why the field of view is wider and this is why your images end up less noisy.

Link | Posted on Sep 12, 2016 at 17:14 UTC
On article What is equivalence and why should I care? (2222 comments in total)
In reply to:

dsumanchy: ”F-numbers and ISO are sensor-size independent. "

Respectfully, I disagree. Here are the crop sensor equivalence formulas which you can verify yourself with a real world test:

FX focal length = crop factor * lens focal length
FX aperture = crop factor * lens aperture
FX ISO = iso * crop factor * crop factor

To verify these are true grab a Nikon FX lens, a DX body and an FX body (Bodies of the same sensor generation are required). For this example I'll use 40mm focal length:

On the DX Body Shoot @ 40mm, f2.8 ISO 800
On the FX Body Shoot @ 60mm, f4.2 ISO 1800

Now compare images in lightroom, aperture, bokeh, noise, and FOV will be nearly identical! So no, bokeh (aperture), as well as ISO (noise) are not the same on FX and DX and thus sensor size does indeed affect both....when normalized to 35mm (FX) standard, which is the point of this article: equivalence.

Before arguing, do the test yourself and be prepared to raise an eyebrow, just like I did when I first performed the test...

The problem with the ISO standard (or, rather, the three different measurements set out in the ISO standard that you're likely to encounter) is that people think they have an inherent understanding of what it means.

The SOS part of the standard, for instance, is essentially a clumsy pastiche of the film ISO standard. And it does dictate that images should 'look the same' in terms of midtone JPEG output brightness for a given exposure and light level.

However, just as an ISO 800 sheet of medium format film will look better when printed to a given size than a piece of ISO 800 35mm film at the same print size, it tells you nothing about noise characteristics.

This may not be what you want ISO to mean, but that's essentially what it means in current cameras.

And, because ISO (SOS) means *x* JPEG brightness for *y* exposure values and *z* illumination level, regardless of whether you use an iPhone or a Phase One, it obscures the effect of sensor size.

Link | Posted on Sep 10, 2016 at 01:15 UTC
On article What is equivalence and why should I care? (2222 comments in total)
In reply to:

dsumanchy: ”F-numbers and ISO are sensor-size independent. "

Respectfully, I disagree. Here are the crop sensor equivalence formulas which you can verify yourself with a real world test:

FX focal length = crop factor * lens focal length
FX aperture = crop factor * lens aperture
FX ISO = iso * crop factor * crop factor

To verify these are true grab a Nikon FX lens, a DX body and an FX body (Bodies of the same sensor generation are required). For this example I'll use 40mm focal length:

On the DX Body Shoot @ 40mm, f2.8 ISO 800
On the FX Body Shoot @ 60mm, f4.2 ISO 1800

Now compare images in lightroom, aperture, bokeh, noise, and FOV will be nearly identical! So no, bokeh (aperture), as well as ISO (noise) are not the same on FX and DX and thus sensor size does indeed affect both....when normalized to 35mm (FX) standard, which is the point of this article: equivalence.

Before arguing, do the test yourself and be prepared to raise an eyebrow, just like I did when I first performed the test...

I think we're both taking opposite approaches to the same problem.

The point I was trying to make in the article is that F-number and ISO hold the same values, despite their effect being different at different sensor sizes. You're saying that they should change to reflect the changed effect.

However, like it or not, the way exposure has been indicated is intentionally such that F-number and ISO appear to work in the same way across multiple film formats (or sensor sizes in this instance).

This is the reason the very idea of equivalence ends up being so controversial, because it means stepping away from the per-unit-area way of thinking about exposure and considering a per-image perspective. It appears to run counter to everything people have learned, despite the effect (Large sensors produce cleaner images than smaller ones) being universally recognised.

Link | Posted on Sep 9, 2016 at 21:18 UTC
On article What is equivalence and why should I care? (2222 comments in total)
In reply to:

dsumanchy: ”F-numbers and ISO are sensor-size independent. "

Respectfully, I disagree. Here are the crop sensor equivalence formulas which you can verify yourself with a real world test:

FX focal length = crop factor * lens focal length
FX aperture = crop factor * lens aperture
FX ISO = iso * crop factor * crop factor

To verify these are true grab a Nikon FX lens, a DX body and an FX body (Bodies of the same sensor generation are required). For this example I'll use 40mm focal length:

On the DX Body Shoot @ 40mm, f2.8 ISO 800
On the FX Body Shoot @ 60mm, f4.2 ISO 1800

Now compare images in lightroom, aperture, bokeh, noise, and FOV will be nearly identical! So no, bokeh (aperture), as well as ISO (noise) are not the same on FX and DX and thus sensor size does indeed affect both....when normalized to 35mm (FX) standard, which is the point of this article: equivalence.

Before arguing, do the test yourself and be prepared to raise an eyebrow, just like I did when I first performed the test...

Equally, ISO* as it's usually used, takes the same light per unit area and turns it into the same output brightness.

What that exactly means in terms of how much light (signal) an image is made up from does vary with sensor size but in terms of the general effect achieved (output per input) ISO is constant across sensor sizes.

*Or the SOS section of the standard, anyway.

Link | Posted on Sep 8, 2016 at 23:18 UTC
On article What is equivalence and why should I care? (2222 comments in total)
In reply to:

dsumanchy: ”F-numbers and ISO are sensor-size independent. "

Respectfully, I disagree. Here are the crop sensor equivalence formulas which you can verify yourself with a real world test:

FX focal length = crop factor * lens focal length
FX aperture = crop factor * lens aperture
FX ISO = iso * crop factor * crop factor

To verify these are true grab a Nikon FX lens, a DX body and an FX body (Bodies of the same sensor generation are required). For this example I'll use 40mm focal length:

On the DX Body Shoot @ 40mm, f2.8 ISO 800
On the FX Body Shoot @ 60mm, f4.2 ISO 1800

Now compare images in lightroom, aperture, bokeh, noise, and FOV will be nearly identical! So no, bokeh (aperture), as well as ISO (noise) are not the same on FX and DX and thus sensor size does indeed affect both....when normalized to 35mm (FX) standard, which is the point of this article: equivalence.

Before arguing, do the test yourself and be prepared to raise an eyebrow, just like I did when I first performed the test...

The *effect* of the F number changes with sensor size: that's the entire point of this article (Which is why it does exactly what you suggest doing - shooting at equivalent exposure settings).

However, the focal length and the entrance pupil know nothing of the sensor placed behind them and are utterly unchanged. And the light per unit area projected by any two lenses with the same F number will be the same (to a good approximation - I'm not getting into T stops).

There's no such thing as full frame F-number, just an F-number. However, to the point that both you and the article are making: you can compare the effect of an F number on one sensor, to the effect that same F number has on full frame (or whatever other reference point you choose).

Link | Posted on Sep 8, 2016 at 22:44 UTC
On article Sony Cyber-Shot DSC-RX100 In-Depth Review (3 comments in total)
In reply to:

great Javier: So weird that there are not almost any comments in a such important camera.

I think the review pre-dates the ability to add comments.

Link | Posted on Sep 3, 2016 at 00:01 UTC
On article What is equivalence and why should I care? (2222 comments in total)
In reply to:

Luddhi: Is there a formula for calculating the horizontal field of view for a given focal length?
I ask because I end up with some odd focal lengths when adapting lenses and would like to be able to calculate the horizontal angle of view.

[In which case, try Wikipedia](https://en.wikipedia.org/wiki/Angle_of_view)

Link | Posted on Sep 2, 2016 at 17:23 UTC
On article What is equivalence and why should I care? (2222 comments in total)
In reply to:

emxgarcia: Just to further understand and please correct me if I'm wrong.
An f/1.2 full frame lens mounted on an APS-C body it not a truly so, maybe an f/1.6.
But, an f/1.2 MFT lens on an MFT body is f/1.2

Let's leave the word 'real' out of this.

But yes, equivalence lets you work out the capability that different camera **and** lens combinations give. So rather than just concluding 'Full Frame cameras are expensive,' you can conclude: 'this combination of camera and lens will give me *this much* capability for *this* much cost and *this much weight*, compared to *that* body and lens...'

Link | Posted on Sep 1, 2016 at 17:09 UTC
On article What is equivalence and why should I care? (2222 comments in total)
In reply to:

Luddhi: Is there a formula for calculating the horizontal field of view for a given focal length?
I ask because I end up with some odd focal lengths when adapting lenses and would like to be able to calculate the horizontal angle of view.

It's usually easiest to use [this website](http://www.tawbaware.com/maxlyons/calc.htm), which has a good range of calculators

Link | Posted on Sep 1, 2016 at 17:05 UTC
In reply to:

ObelixCMM: On August 5th Sony released new firmware for Sony A6300 promising to fix overheating problems and DPR doesn't report it.
Today DXO One gets new accessories and firmware update, and we have article by DPR Administrator Richard Butler, talking about "significant" firmware update for camera that will never reach 10% of Sony A6300 sales.

I think it's fair to call a firmware update that adds Wi-FI capability (or, at least, engages it) 'significant.' DxO sent us a press release. Between the firmware and the accessories, it seemed worth covering.

By contrast, Sony didn't issue a press release and its description of the firmware explicitly said that the improved 'temperature stability' was in 'picture shooting mode,' rather than movie shooting mode that had attracted attention. That said, I'm surprised to find we didn't publish a story on it, since I remember it being discussed.

However, we have installed the firmware, tested it and updated our a6300 review conclusion (it can now shoot for 29:59 in cool conditions but can cut out before then in a warmer environment).

Link | Posted on Sep 1, 2016 at 00:59 UTC
In reply to:

privatebydesign: Really? On close examination the difference between the 5D MkIV, which you call "significant" and the 1DX MkII which you say "exhibits pretty low levels" is minuscule.

The link below is to an overlay comparison.

http://www.canonrumors.com/forum/index.php?topic=30636

Note that I said I was deferring to Rishi not because I trust him personally (though I do), but because, unlike me at that moment, he had access to both cameras and was, therefore, in a better position to know the right answer.

While I agree all writers need to earn trust (and, as I say, we try to provide enough of the files that you don't have to take our word for what we say), it's also reasonable to expect to be given a degree of benefit of the doubt. Starting off with the assumption that everyone is inherently untrustworthy and likely to be actively trying to mislead is just as perverse as having blind faith.

Link | Posted on Sep 1, 2016 at 00:39 UTC
In reply to:

privatebydesign: Really? On close examination the difference between the 5D MkIV, which you call "significant" and the 1DX MkII which you say "exhibits pretty low levels" is minuscule.

The link below is to an overlay comparison.

http://www.canonrumors.com/forum/index.php?topic=30636

I try to address everything that looks like an honest question that I see, but the time available is distinctly limited. I can't remember ever asking anyone to just trust me or us (I'm not saying I haven't, but, as a habit, I try not to).

Virtually everything we publish has downloadable files to allow comparison (something not possible in the instance of video of the back of cameras' screens, I'll concede). But, probably to a greater degree than any other site, we show our working, provide the files and are willing to discuss our methods, rather than expecting anyone to just trust us.

By all means PM me if you have any questions. But there's a reason I cut together the whole video of the cameras being panned, rather than just showing the illustration with the red lines: because it shows the difference more clearly. I don't fully know why the video doesn't make it look as dramatic as it is (and as the frame grabs show), which is why I tried to make clear I was speculating.

Link | Posted on Sep 1, 2016 at 00:34 UTC
In reply to:

photophile: Didn't Lytro already do this ?

Lytro only performed a single capture, the multiple images were rendered from that single capture. However, they made a radically different resolution/refocusability trade-off.

Link | Posted on Sep 1, 2016 at 00:15 UTC
In reply to:

Jonath: Just wondering, the 5D MkIV is the only camera in the default test camera selection with an AA filter, does this interact in anyway or influence the level of noise observed in these tests?

With the risk of being slightly wrong (either in concept or explanation), I don't think it matters.

So long as you capture the same amount of light, I don't think it especially matter which pixel collects which photon. The total noise will be the same, even if the precise distribution (which inherently has a significant random element to it), is slightly different.

Link | Posted on Aug 31, 2016 at 23:05 UTC
In reply to:

KW Phua: I think if Canon remove the AA filter, the detail and noise will be improved.

The detail level might increase, but so would the risk of false colour and moiré.

There's no reason to think that noise would be affected.

Link | Posted on Aug 31, 2016 at 22:32 UTC
In reply to:

photo_rb: In the camera selector dropdown is Canon 5D supposed to be the 5D MkIII?

No. The camera labelled 'Canon EOS 5D' is the Canon EOS 5D (we shot it for last week's Throwback Thursday article).

The 5D Mark III was one of the first cameras to be shot as part of the ISO Invariance test but pre-dates the introduction of the Exposure Latitude test.

Link | Posted on Aug 31, 2016 at 21:14 UTC
Total: 4004, showing: 41 – 60
« First‹ Previous12345Next ›Last »