[Pentaxor]

Some time ago (I cannot remember if it was after the K20 or the K7 DPR review), people on that site also complained about the FA50 usage. One of the editors did respond that they had also tested the D-FA50 macro and its results were not any better, hence they used the FA50. Not that I am vouching for that claim though....

Also, with these higher-megapixel sensors, f9 is already encroaching on the diffraction limits. I think the reason they use that aperture generally (although IIRC they used f8 for the 60D) is that, as you can see, they are not very good about getting the exact right focus, and this gives them more margin for error.

maybe next time they should use the DA*55/1.4 instead. if they say that it's still not any better, I would suggest them to go and see an eye doctor.

[falconeye]

DPReview deserves a break here. All of us would have thought that a FA50 or DFA50Macro are good enough to compete.

But with today's high Megapixel sensors and 100% measurbation, tiny differences matter and DPR's own results show that e.g., the FA50 is outperformed by e.g., a Sigma 50/1.4. Moreover, the FA50 didn't DPR-test sharpest in the center.

And without any conspiracy, DPR may just have more lenses and more time at their disposal when they test Canikon. I can assure you that careful testing is very time consuming. I'm sure though that with all the vocal Pentaxians, DPR now uses as much time for a Pentax

BTW, DPR seems to have posted new crops for the K-5.

[mikemike]

DPReview deserves a break here. All of us would have thought that a FA50 or DFA50Macro are good enough to compete.

But with today's high Megapixel sensors and 100% measurbation, tiny differences matter and DPR's own results show that e.g., the FA50 is outperformed by e.g., a Sigma 50/1.4. Moreover, the FA50 didn't DPR-test sharpest in the center.

And without any conspiracy, DPR may just have more lenses and more time at their disposal when they test Canikon. I can assure you that careful testing is very time consuming. I'm sure though that with all the vocal Pentaxians, DPR now uses as much time for a Pentax

BTW, DPR seems to have posted new crops for the K-5.

If they wanted to test new camera bodies they should test with a lens that is available cross-platform like the sigma. Its not like they tested with several different lenses, with just the kit lenses, or even the most popular 50 mm lens for each platform so something is does seem fishy about why they chose those lenses and those apertures.

[ManuH]

If they wanted to test new camera bodies they should test with a lens that is available cross-platform like the sigma.

In theory this is ideal. In practice different copies may not perform the same. It goes to show how difficult it is to do controlled testing.

[falconeye]

In theory this is ideal. In practice different copies may not perform the same. It goes to show how difficult it is to do controlled testing.

You could use Zeiss glass as it ships with the sample's MTF plot (test protocol) obtained from an optical bench which isn't camera specific.

Or being DPReview (they should have the funds by now), they just aquire an optical MTF test bench used on ebay Or build one, it's not that hard (and a lot easier than Labo FNACs AF.C simulator I wrote about in another thread). I heard photodo isn't using their's anymore

[geauxpez]

Sounds like the K-f shots on DPReview were re-shot. I'm sure that this will bring a whole new round of pixel-peeping and conspiracy theories.

[sjwaldron]

They should use a Tamron Adapt-all lens or something and just manual focus it. That would factor out the sharpness aspect they say are not part of their review.

[JohnBee]

They should use a Tamron Adapt-all lens or something and just manual focus it. That would factor out the sharpness aspect they say are not part of their review.

That sounds like s a great idea in fact!

[Pentaxor]

They should use a Tamron Adapt-all lens or something and just manual focus it. That would factor out the sharpness aspect they say are not part of their review.

atleast the copy variation dilemma is solved. we'll be able to see how these cameras really work with a single test lens.

[glanglois]

You could use Zeiss glass as it ships with the sample's MTF plot (test protocol) obtained from an optical bench which isn't camera specific.

Or being DPReview (they should have the funds by now), they just aquire an optical MTF test bench used on ebay Or build one, it's not that hard (and a lot easier than Labo FNACs AF.C simulator I wrote about in another thread). I heard photodo isn't using their's anymore

Now you're asking for rigor and reliability (consistency) from a testing site.

Ummm .... OK, so it doesn't seem like much when put that way.

Investing in some Zeiss glass, with a nod to Zeiss in each review, sounds both practical and effective. Likely to put an end to much of the carrying on. Would save me at least ten minutes each day.

Is the lack thereof owing to unwillingness to invest, lack of cash flow (snort), or a disinclination to take the whole testing thing seriously?

[Asahiflex]

atleast the copy variation dilemma is solved. we'll be able to see how these cameras really work with a single test lens.

Actually Dpreview (and all other camera testing institutions) should order a very sharp "Adaptall" lens right now. The adapter could be quite simple as AF/automatic diaphragm is not needed at all. And, as far as I am aware, Adaptall adapters to any mount are available plenty. Maybe they could even order a custom lens at Tamron - one with a very high resolution.

Needless to say: I refuse to look at the crops of any camera until they use the same lens for each and every shot, as there are way too much variables involved. The fact that both Imaging Resource and Dpreview had to reshoot the K-5 samples proves my point.

[gazonk]

Needless to say: I refuse to look at the crops of any camera until they use the same lens for each and every shot, as there are way too much variables involved. The fact that both Imaging Resource and Dpreview had to reshoot the K-5 samples proves my point.

Sadly, that won't help as long as they use wrong apertures (IR used 5.6 for K-5 and 8 for D7000) or miss the focus (the K-5 IR shots are back-focussed)

[ksuwildkat]

The treatment of Pentax on DPR reminds me of subtle racism. I wrote a long analysis/rant about the subtle ways that DPR "discriminates against Pentax.

"As for AF speed, how does that speed work on a lens that doesnt AF - like the one DPR uses for its IQ tests for the D5000 - "Nikon D5000: Nikkor 50 mm F1.4G lens, Aperture Priority, Manual WB, Default Parameters. Standard Picture Controls, Active D-Lighting Off, NR Normal, JPEG Large / Fine"? Funny, DPR doesnt tell you that the lens they did the IQ test with HAD to be manually focused and then they had to switch to another lens to do the AF test. Talk about shady testing. Kinda like the testing on VR on Nikon lenses. For some strange reason, Nikon cameras with Nikon lenses are incredibly unstable with VR off. But put a Sigma or Tamron on the same camera and it recovers."

"Review of Nikon 18-200VR

It's no great surprise to see Nikon's VR system performing well here. Whilst it doesn't quite seem to deliver the 4 stops which Nikon claim in our hands, it comes pretty close, especially at 200mm. The key difference here is not so much the yield of critically sharp shots at slow shutter speeds, as the hugely increased chances of getting usable shots with only mild blur, which would be completely impossible in the absence of stabilisation.

Review of K200D VR

With anti-shake switched on the improvement is immediately visible although you still only get 100% sharp shots at 1/100th sec. On the plus side even at shutter speeds as slow as 1/6th sec you get 50% usable shots. The system gives you an advantage of approximately 2 stops but tends to work more efficiently at very slow shutter speeds. This is a reasonable result but not outstanding. It is also not even close to the the 4 stops that Pentax claims its anti-shake is capable of.

Now go look at the actual results from the test. According to their tests, at 50mm and 1/6th of a sec, the Nikon had a 10% chance of a "Sharp" picture. The K200 had a 30% chance. Across the board the K200 did better but in the "Conclusions" section the reviewer states "Efficient image stabilization (not as good as Pentax claims though)" while the Nikon gets "Very effective vibration reduction system, at least 3 stops benefit.""

Look at the cameras comparisons in reviews. if it is a Nikon, the next camera is a Canon. If it is a Canon, the next camera is a Nikon. If its a Pentax, the next camera is an Olympus or Sony. There might not even be a Nikon on the chart to compare. The subtle message - Nikon and Canon are in a "different category."

To their credit, DPR did redo the shots. But how come it took community outrage? It will be interesting how the K-5 review comes out on DPR since it will likely be one of the last. The reviews so far have been glowing. Do they dare speak the trurth and agree?