Richard Butler

Richard Butler

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

Comments

Total: 4985, showing: 1 – 20
« First‹ Previous12345Next ›Last »
On article Action-packed: Sony a6500 review (1160 comments in total)
In reply to:

Zoom Zoom Zoom: I have a question (and maybe I just missed it on the review), but..

What lens was used on the test scene with the a6500? Somehow on these reviews, the issue on the test scene is extremely subjective because of the lens. I mean, for example look at the RX1, RX1-R and RX1RII. I dont know if it was the setup, or bad samples, or what.. but from corner to corner, their test scenes is a tragic blur and sharpness misery! But this is completely inaccurate! I have an RX1R and the Carl Zeiss 35mm f:/2.0 lens is pitch sharp on mine.. NOTHING to do with the IQ horror on the DPreview's test scene. So, what's up with that?

All this said, and the reason why I am curious as to what lens was used on this review of the a6500, is because I gotta say, the a6500 sure has among the best JPG output Ive seen. It certainly kills my RX1R's miserable JPG output and even when compared on this review to the a7RII, it totally rival's it in term of apparent detail resolving power. Amazing JPG engine!

If you click on the [ i ] tab underneath each close-up sample on the test scene, it should tell you which lens was used.

The RX1 lens doesn't produce a very field of focus, especially at relatively close working distances so it looks worse shooting a flat target than it would for more distant subjects arranged at different depths. There's not much we can do about this, other than mention in in the review text.

Link | Posted on Jun 28, 2017 at 00:38 UTC
On article Fujifilm X100F Review (774 comments in total)
In reply to:

K1000usr: @Richard Butler, it would be fun to see a comparative shootout with a Sony DSC-RX1* series camera. Not exactly apples to apples, but similar in many ways.

I am sure both cameras will outshine in some categories, some mutually exclusive.

Thanks.

The RX1 will produce better Raw image quality: there's no making up for the considerable difference in sensor size.

However, the X100F is significantly cheaper, slightly smaller and has much better battery life and prettier JPEGs, to my eye.

Link | Posted on Jun 27, 2017 at 19:18 UTC
In reply to:

marabunta: As of Monday, June 5, 2017, 279 pounds UK trades at $360.72 US and 319 Euros. The $300 figure is a wee bit off and a wee bit optimistic. Perhaps Yanks will be treated to a lower price do to market forces?

The story doesn't say that £279=$300. It says that £279 inc VAT would be roughly $300 *without tax*.

Link | Posted on Jun 6, 2017 at 21:21 UTC
On article iOS 11 brings photo updates to iPhone and iPad (5 comments in total)
In reply to:

donutlover: "we're curious how sharing and support from image editing apps will jive with the new format" -- I think the word you want is "jibe", not "jive" :-D

Do your apps not dance with one another?

Link | Posted on Jun 5, 2017 at 23:51 UTC
In reply to:

tlinn: I was all excited until I read the last sentence.

A fair number of smartphones are sold with prime lenses.

Link | Posted on Jun 2, 2017 at 17:14 UTC
On article Sigma SD Quattro H Review (692 comments in total)
In reply to:

KrampusClaus: looking at the top studio scene, using the middle wood print, which is the highest resolution that is obvious in this scene, the pentax k1 shows the correct texture in the upper left area, whereas both sigma show high resolution but INCORRECT processing of the diagonal and other texturing.

the nikon 810, sony a7rII both show extreme aliasing absent in the sigma, the pentax k1 is the best default cam - accuracy of texture that is.

as always refer to pentax 645z as the resolution ref, but the canon 5ds r showing nearly the resolution, but aliasing is bad in both.

it is so odd that all the sigmas have the diagonal the wrong way - the upper left small should be diagonal down rightward, so with a/some better algorithms, the sigma could be right there resolution wise, but for now - it would give screwy texturing for many things - of course this is the extreme @ 100% scrutiny but still the sigmas are wrong.

It's still aliasing. Foveon sensors aren't subject to the false colour that comes when Bayer sensors alias but they still mis-represent anything above the Nyquist frequency of their pixel count.

Link | Posted on Jun 2, 2017 at 06:31 UTC
On article Sigma SD Quattro H Review (692 comments in total)
In reply to:

Polacofede: would be nice that sigma put some effort on adding pentax K mount again... nothing to do with this review but well... how much does this camera costs?

If only the reviewer had mentioned that in the conclusion.

Link | Posted on Jun 2, 2017 at 04:58 UTC
On article Sigma SD Quattro H Review (692 comments in total)
In reply to:

KrampusClaus: looking at the top studio scene, using the middle wood print, which is the highest resolution that is obvious in this scene, the pentax k1 shows the correct texture in the upper left area, whereas both sigma show high resolution but INCORRECT processing of the diagonal and other texturing.

the nikon 810, sony a7rII both show extreme aliasing absent in the sigma, the pentax k1 is the best default cam - accuracy of texture that is.

as always refer to pentax 645z as the resolution ref, but the canon 5ds r showing nearly the resolution, but aliasing is bad in both.

it is so odd that all the sigmas have the diagonal the wrong way - the upper left small should be diagonal down rightward, so with a/some better algorithms, the sigma could be right there resolution wise, but for now - it would give screwy texturing for many things - of course this is the extreme @ 100% scrutiny but still the sigmas are wrong.

That's aliasing (a higher frequency than the pixel count can correctly render being mis-represented as a frequency within the sensor's domain), so it's not something that processing can correct. You could add an AA filter so that you don't mis-represent it, but you can't reliably get it right.

Link | Posted on Jun 2, 2017 at 03:15 UTC
On article Sigma SD Quattro H Review (692 comments in total)
In reply to:

Jim Evidon: The Foveon sensor is top of the class in sharpness, but it has pretty much stood still in other areas such as color output, WB, and range of light, etc. Add that to the asking price of over $2300 just for the body and it becomes a camera for the very few that love the Foveon output and must have the camera. Not everyone's cup of tea, I'm afraid. The DP Review photo examples ( see appendix tests part 6 ) , if accurate, demonstrate that the Sigma sd Quattro H runs out of noiseless imaging above ISO 400 which is pretty sad these days. Even my M9 goes to ISO800 before noise starts to creep in and that camera is several years and probably 3 camera generations old.
My Leica T and Olympus Pen F are still singing above ISO1600. Come on Sigma, I know you can do it. Try harder!

Hi Jim. Where are you seeing $2300 in the review?

Also, please note Johan Borg's use of the word 'basically.' The DNGs are 12-bit files, so are far from 'uncompressed JPEGs.' They aren't, however, completely raw (very few Raw files are) instead, as discussed in the review, they have the colors resolved by the Sigma's algorithms in the camera, since most Raw processing software can't do that for Foveon sensors.

Link | Posted on Jun 1, 2017 at 23:52 UTC
On article Sigma SD Quattro H Review (692 comments in total)
In reply to:

Shinsei: SIGMA should reform it by an E mount.

Probably not. They haven't made E-mount details fully public, just [licensed 'basic specifications'](https://www.sony.net/SonyInfo/News/Press/201102/11-018E/) to 'manufacturers of lenses and mount adaptors.'

Link | Posted on Jun 1, 2017 at 23:47 UTC
On article Sigma SD Quattro H Review (692 comments in total)
In reply to:

Stigg: No matter what minor differences there are it still produces ones and zeros assembled on a computer screen which can be later digitally printed to paper. To my eyes that look and feel is very obvious. It has its uses of course, but the claims of "film-like" are patently silly and nothing more than marketing claims. If you want a film-like look just use film. You'll love it!

Much harder to show on a website, though.

Link | Posted on Jun 1, 2017 at 23:07 UTC
On article Sigma SD Quattro H Review (692 comments in total)
In reply to:

Vamp898: Actually i think the High-ISO rating is quite unfair.

I know APS-C Cameras (like the A77) with less megapixels performing about the same level and getting better ratings.

Considering the MP Size and the Sensor size (51 Megapixel on APS-H) the Noise Level is actually pretty good and, at worst, on average level.

Sure there are no Bayer APS-H Sensors with 51 Megapixels but if, they wouldn't perform better.

GEONYC - The role of pixel size is usually insignificant in comparison to the effect of sensor size, so it's rarely relevant. That's why it's not often discussed.

Link | Posted on Jun 1, 2017 at 18:08 UTC
On article Sigma SD Quattro H Review (692 comments in total)
In reply to:

JackM: Providing only images processed from DNG in Adobe Camera Raw and no images processed from X3F in Sigma Photo Pro is a rather heavy handed editorial decision if you ask me. People who prefer this softer ACR look have no use for this camera at all in the first place. Would love some X3F raws to play with if you don't want to bother with SPP.

There's an X3F of the ISO 100 studio scene shot included specifically so that you can compare. However, it was not plausible to review the camera twice with two different workflows, so we concentrated on the one that broadens the sd Q H's appeal, rather than the one that existing users will already be familiar with.

Link | Posted on Jun 1, 2017 at 18:05 UTC
On article Sigma SD Quattro H Review (692 comments in total)
In reply to:

supersonic: The Sigma mount is, near enough, a Pentax K bayonet, with Canon compatible electronics (pins swapped), set at Canon registration length (44 mm).

Please feel free to correct.

Which is a shame, because Nikon, (46.5mm), Pentax (45.45mm) are longer.

If the Sigma camera had a Canon bayonet, instead of the narrower diameter PK size, then it would be so simple to adapt Nikon and Pentax, even Canon, and other lenses to fit, using readily available adapters.

Sigma could still sell their lenses, but in Canon mount.

I seem to remember others have modified Sigma cameras to fit a Canon bayonet. If this could be done to this camera (or Sigma would do it themselves), that would be so good, and open up so many other lens opportunities.

Of course, Canon might not be happy. But "third party" modification option might avoid that.

GEONYC - [Chevy Prizm](https://en.wikipedia.org/wiki/Geo_Prizm)?

Link | Posted on Jun 1, 2017 at 18:03 UTC
On article Sigma SD Quattro H Review (692 comments in total)
In reply to:

The Davinator: I want to like it...but even at iso 100, the noise is obtrusive.

I dialed the sharpening down in my sample images (lower amount, finer radius) but I left it at defaults in the test scene, since that's the standard process for the test scene.

Both DNG and X3F shots of the test scene are available (the X3F is the Raw download when 'Hi Res' is selected), so that you can draw your own conclusions.

Link | Posted on Jun 1, 2017 at 17:59 UTC
On article Sigma SD Quattro H Review (692 comments in total)
In reply to:

Graham Gibson: I agree with the overall sentiment of the review, but the image tests really leave a lot to be desired from a comparison standpoint. The RAW DNG processed images look terrible, like they were just using full default settings. Just compare the results to the JPEGs to see the real Quattro image quality. Clearly better results should be possible from the DNGs (and I have seen such from other sources).

Additionally, as others have pointed out, it would be hugely helpful to include X3F and X3I (SFD) shots of the same scene processed through Sigma's own software to compare to the DNG RAWs. Everyone interested in this camera is going to want to know which file format to choose and how the output of each workflow compares. You include a whole review section comparing X3F and DNG, so why not show us the accompanying studio scene?

The X3F of the studio scene is [available for download](https://www.dpreview.com/reviews/image-comparison/download-image?s3Key=6bdc4eacbe794bdd8ecda98ac8be1dfe.dng) from test scene.

The DNGs in the test scene are processed as similarly to all other cameras as possible (except the color response is set by Sigma, not Adobe) and are also available to download.

Link | Posted on Jun 1, 2017 at 00:39 UTC
On article Sigma SD Quattro H Review (692 comments in total)
In reply to:

xpatUSA: "photons of different colors have different amounts of energy and more energetic light would penetrate further into the chip."

OOPS, Richard!

12 million green, 6 million blue, 6 million red would be 24 million sensels in total, which is what you get in a '24MP' Bayer sensor. These would then be demosaiced, which would interpolate the missing values to give 24 million R,G,B pixles.

Link | Posted on May 31, 2017 at 23:53 UTC
On article Sigma SD Quattro H Review (692 comments in total)
In reply to:

Jorginho: No. I do not think this as a food landscape cam. It still seems to seriously struggle with dynamic range. The sky in the first pic just is washed out and strange. It is its achilles heel. If DR is not taxing at low ISO the detail and colours are really really good. But I hoped Dynamic range with the APS-H woul dbe seriously better as well as the tones in the sky. Not seeing it here.

If you mean the one with the train, that could just be my processing. It was a pretty challenging scene.

Link | Posted on May 31, 2017 at 22:35 UTC
On article Sigma SD Quattro H Review (692 comments in total)
In reply to:

mike earussi: I'm really sorry you were too lazy to get the best out of the camera by deciding to use DNG/ACR processing instead of using Sigma's own SPP software. There may be only a slight improvement from the H's regular raw file vs DNG, but there is a big difference when using it in SFD mode, which can only be processed using SPP and would easily rival any 50mp MF body both in DR and resolution, making it the cheapest "MF" camera on the market. If you're going to test a camera in the first place at least take the time to do it right.

Marco - I've used SPP at several points over the years (particularly during the DP1, DP2 and SD1 (M) reviews). Even with a new, fairly powerful computer, it's simply not practical for me to use for any significant number of images.

The need to shoot either X3F or DNG meant that I had to pick one when I was out shooting (rather than trying to constantly switch back and forth). Given the Quattro's likely position as a second camera for many people, we decided to test and review it as a camera that was going to be adapted to fit into existing workflows.

We shot and examined the studio scene in X3F but didn't want to review the camera twice, with two different workflows, one of which can't be compared fairly with anything else we test. We've included a 25.5MP and 51.2MP rendering of the studio scene from Sigma Photo Pro so that people can decide if they do want to go down the X3F route, but for us the interesting thing about the Quattro H is that you don't have to.

Link | Posted on May 31, 2017 at 22:18 UTC
On article Sigma SD Quattro H Review (692 comments in total)
In reply to:

mike earussi: I'm really sorry you were too lazy to get the best out of the camera by deciding to use DNG/ACR processing instead of using Sigma's own SPP software. There may be only a slight improvement from the H's regular raw file vs DNG, but there is a big difference when using it in SFD mode, which can only be processed using SPP and would easily rival any 50mp MF body both in DR and resolution, making it the cheapest "MF" camera on the market. If you're going to test a camera in the first place at least take the time to do it right.

mike - having looked at both SPP-processed X3Fs and ACR-processed DNGs, we focused the review on the camera as a DNG camera, since this significantly broadens its appeal (as discussed in [Carey's article](https://www.dpreview.com/opinion/9977248883/sigma-shoots-dng-raw)).

We probably should have shot the studio scene in SPF mode, in hindsight, but once that opportunity was missed it's hard to justify the time it would take to add it.

Link | Posted on May 31, 2017 at 22:11 UTC
Total: 4985, showing: 1 – 20
« First‹ Previous12345Next ›Last »