[pefoto01]

Имах предвид не ДПР,а dpr,но преводача прави проблем!

[RolloR]

hey guys maybe it's just a bit out of focus?

try to compare it to Pentax k-x and k-7 they're available too, and look a LOT sharper

[eigelb]

hey guys maybe it's just a bit out of focus?

try to compare it to Pentax k-x and k-7 they're available too, and look a LOT sharper

K-x : 12MP sensor taken at F8 vs 16MP sensor at F9

[Edgar_in_Indy]

They would have to use a prime lens though. A zoom would be too hard to adjust focal length perfectly, except if they were using it at either end - which isn't usually the sweet spot of a zoom...

The Sigma 50mm f/1.4 would be a good choice then. A very well regarded lens and it's available in Canon, Nikon, Pentax, Sigma, Sony and Four Thirds mounts.

[pinholecam]

I'm intrigued by this remark regarding the fudging of results through f-stop variation (f9 vs f5.6 ), so I did some shots myself at ISO12800 (JPEG).

Certainly not defending DPR (which I don't trust at all), but I really want too see how bad the selection of f-stop can be.

Camera was mounted on a tripod.
Lens is a A50/1.4 with hood on. (Sorry don't have a FA50/1.4)
Focusing was MF using LV at x8 mag.
NR OFF




f5.6


f9



I can't fault the use of f9, though I do find fault with the sharpness reported on DPR.
Of course the results can be affected by may things like a lack of hood and poor focus, so we will never be sure.
Repeat : I certainly think I see better than what DPR has put up.

[JohnBee]

The edge haze on the DPR samples can be seen at ISO80,200 etc. etc.

[Edgar_in_Indy]

Looking at the images on the DPR review, I've noticed something else strange. The Canon 60D images seem to be properly exposed, while the other images, especially the K-5, seem to be ever so slightly over-exposed. This makes the K-5 images look slightly washed out and less dynamic than the Canon since the blacks aren't as black. I don't know if this was done on purpose, but it definitely makes the Canon look better than the competition when viewed side-by-side.

Looking at the EXIF data, the Canon image was captured at f/8 at 1/640 sec, while the Pentax was f/9 at 1/400 sec. I don't know if those settings should, in theory, come up with identical exposure, but even if that's the case, the fact that they're using two different lenses would still throw some unpredictably into the equation since different lenses will capture different amounts of light, even at identical f-stops.

It just seems awful convenient that the Canon benefited from the differences in exposure...

[pinholecam]

If comparing the shots at f9 alone with NR OFF and NR on Auto (which would be at Medium setting).

f9 ISO12800 NR OFF


f9 ISO12800 NR Auto (ie. Medium)




f9 ISO12800 NR OFF


f9 ISO12800 NR Auto (ie. Medium)



IMO, NR off is better than NR Auto.

How I wish I did not throw away that empty bottle of Bailey's and Martini .....

[Edgar_in_Indy]

I agree. I definitely prefer the image with no noise reduction.

[mattdm]

I presume they chose the aperture at which each camera/lens combo gave the best results in this specific situation, despite the earlier lens review's comments about the "sweet spot".

But the important point is: all the new cameras look pretty much amazing. I don't think there's any point nitpicking between them on image quality. It's the other things: handling, features, and the rest of the system that matter.

(I can't believe lack of AF microadjust didn't make the cons list, for example.)

[er1kksen]

Anyone remember when DPReview turned red for a day as part of an obnoxious 60D advertisement? Color me unsurprised.

Imaging Resource uses the Sigma 70mm f2.8 macro in their tests, widely regarded as one of the absolute sharpest DSLR lenses available. It'd be nice if DPReview did the same. Use of an aperture that's not diffraction-limited on a given sensor and some rigorous focus control would be helpful as well.

Another issue is the use of ACR. It's been known for a long time that ACR does NOT treat all RAW files the same. ACR conversions made Sony cameras look worse than they really were for a while, and the visible differences between the 60D RAW files and others shown look like black point and NR differences, which ACR could certainly have a hand in. Third-party no-nonsense RAW conversion platforms like Digital Camera RAW would be a more objective metric for comparing sensors, especially those with RAW conversion algorithms that are available to the public to look for inequalities (and even better, offer multiple algorithms to account for the fact that no single algorithm can be optimal for all sensors) rather than Adobe's usual smoke-and-mirrors, would be a much better metric for comparing the capabilities of different image sensors.

[Wheatfield]

Not only that the canon is $600 dollars cheaper than the k-5 body. What's up with that?
I am personally holding off on buying the k-5 until the price drops or there are more test results. I got so burned on the k-7 I am a little gun shy.
If i didn't have a couple of good sigma lenses for pentax I would change for sure.

Whats up with the price is that the K5 is priced higher than the Canon.
It's pretty simple.

How did you get burned on the K7? Were you expecting it to do laundry?

Good and Stigma lens don't really belong in the same sentence, so why not just change.
The world will be a better place if you do.

[Eruditass]

They should use a third party lens available in all mounts.

[Wheatfield]

Anyone remember when DPReview turned red for a day as part of an obnoxious 60D advertisement? Color me unsurprised.

Wasn't it Phil Askey who decided to compare the resolution of some Canon against the Pentax 6x7 and only managed to prove that a Manfrotto 055 tripod wasn't big enough to hold the 6x7 and (IIRC) an 800mm lens?
Like any idiot with half a brain should have known that right off the get go.
I guess Phil isn't just any idiot, he's special.

Anyway, I checked my mail archives,
I believe, from a conversation I had a couple of years ago that Amazon had bought DPReview, so they are going to have an automatic bias for whatever Amazon wants to flog.
My guess is that is Canon.
DPreview has pretty much been Canon's bitch for some time now.

[interested_observer]

As others have indicated a standard lens available with multiple mounts. However, body reviewers need to go one step beyond. Select a lens from a third party available in all (or most) mounts, and then have the lens manufacturer come up with a matched set of optics, that are as close as possible across the mounts. Apples to apples. Then with "very similar" glass across all the mounts, you can start to compare the output from each of the sensors and imaging systems.


Also, yes Amazon acquired d p r e v i e w . c o m in 2007 - here is the announcement.

• http://www.*************/news/0705/07051402amazonacquiresdpreview.asp
• http://www.google.com/search?q=amazon+acquires+*************&ie=utf-8&oe=utf...ient=firefox-a