Dpreview - fact, or fiction?

[brians.]

<p>

<p >Dpreview has caused me no end of grief and relief. I would never buy a camera without checking their review first. Their reviews are still the most comprehensive and objective of any I've found. However, the affect of different lenses in their tests is barely mentioned and real-world results are often different. What ticks me off most about Dpreview is some of the participants in their forums. Instead of discussing or ignoring a post, some participants slam the photographer for lack of experience. That sort of thing bugs me because everyone works at a different level and novice views can often be very revealing, despite what the pros like to profess. That said, Dpreview has galleries that rival Photo.net. Right now, I am avoiding Dpreview because of some unfriendly conversations in their forums. At least for now, I'm avoiding the place.</p>

</p>

[Landrum Kelly]

<p>"The fact that these guys have London as a place to shoot sample photos doesn't diminish their efforts."</p>

<p>How wonderfully magnanimous of you! I'm sure that everyone else was absolutely sure that something done in New York or San Francisco would have been much better.</p>

<p>The site has not given your pet camera the raves you are sure it deserves, and so you trash the site. The credibility of the site is unaffected by your reaction.</p>

<p>--Lannie</p>

[allardk]

<p>Tests can be objective, if conducted right. Conclusions drawn from them are subjective. DPR never made a secret of that, and I think they do a pretty good job at explaining why they think a certain camera is good or not. You don't have to agree. At least their reviews don't suffer from as many errors as <a href="00SPqP">this one</a> , as a random example.</p>

[danielleetaylor]

<p>My complaint with DPReview is that they judge noise on a strictly per pixel basis. Per pixel the 50D has more noise than the 40D. In print, where 50D images are enlarged less than 40D images, the 50D shows less noise. They did the same thing with the 5D mkII review when comparing it to the D700. (Although in that case you might need to print very large to show any real differences.) They basically said the D700 was a bit cleaner, but I bet in print it's not, again due to the enlargement factor. We're admittedly talking about small differences, but at least with the 50D there were endless discussions here, and they all made me want to scream "just try some prints!"</p>

<p>DxO has the same discrepancy in their tests, but they've at least acknowledged it with an article on the issue. I've seen this else where in comparisons of bodies. I think you need to keep track of per pixel issues, but also resize lower resolution images up to the highest resolution image in the test and see how things stack up when enlargement issues are involved, because enlargement counts in every single print. (Be careful with DxO scores because their tests ignore total image resolution to a greater degree than DPReview. A 6 MP camera can score higher than a 24 MP camera.)</p>

<p>Along those same lines, I've seen reviews where people compare cameras without optimally processing each image. They will then declare A the winner when B really is a match. Optimum USM, for example, varies based on things like pixel density. DPReview partially addresses this in their RAW comparisons, and they seem to have a keen eye when it comes to noticing when things really are equal or would be with a little processing.</p>

<p>In terms of detail the 50D is picking up a little bit more high frequency detail than the 40D, which is to be expected, but nothing that would make a user toss their 40D and run right out with a credit card. (Especially in this economy.) Feature/body wise, I think the 50D is pretty significant, but again it all depends on your finances.</p>

<p>While I'm reviewing reviews, I noticed that DPReview made some comments about AF in the 5D mkII review. (I forget if they did on the 50D or not.) The only thing that bothers me here is that they have no formal test procedures to judge this, and I think we all may be subject to a little number and marketing bias here. I've seen countless threads where users agonize over getting the 50D (9 point) or D300 (51 point), complaining that Canon 'cheats' users by not putting their 45 point AF in the xxD or 5D series, with the D300 AF touted as "professional" based on point count. Where am I going with this? There was a recent thread on photo.net where a Nikon user asked about the different AF modes, and most responses from Nikon users advised him away from the full 51 point mode towards the 9 or 21 point mode! They expressed that too many points just leads to confusion whenever the AF module has to distinguish between any real clutter, and that the better your tracking skills as a photographer, the more you find fewer points to actually be more precise.</p>

<p>This has been my exact experience and I use center point only 99% of the time. So for me the most professional AF module is the one which is most sensitive, especially center point, and can most quickly acquire/track the target, which depends greatly on the lens in question. But no site is doing any formal tests to determine something like that. (While I'm at it, I do think Canon needs to widen the AF area on the xxD and especially the 5D, and add a few more points. But this is primarily for the narrow situation of BiF against a blue sky. And for people who find it more convenient to choose a point when on a tripod than to focus/recompose.)</p>

<p>Any way...I generally like DPReview and think they are one of the more consistent and reliable testing sites. But I thought I would take the chance to nit pick ;-)</p>

[allardk]

<p><em>DxO has the same discrepancy in their tests, but they've at least acknowledged it with an article on the issue. </em><br>

I see that complaint a lot. It's only true if you only look at the single number that comes out of the test. If you use the 'compare cameras' feature, you can select print or screen. Look at this comparison between 40D and 50D. (Is it ok to post screenshots from DxO here?, if not let one of the moderators remove the images and viewers go <a href="http://www.dxomark.com/index.php/eng/Image-Quality-Database/Compare-cameras/(appareil1)/180|0/(appareil2)/267|0/(onglet)/0/(brand)/Canon/(brand2)/Canon">here</a> )<br>

The only complaint you could have about this is that they use a simple square root to get from one to the other, ignoring the correlations you get from demosaicing, but that's nitpicking indeed. Though i'd like to see a systematic study on that one (not just the single blog entry on dpreview).</p><div></div>

[hughw]

<p>When one is considering a substantial investment in the hundreds or thousands of dollars then in my humble opinion it would be folly to take direction from only one source. The beauty of living in the "internet age" is that we can seek and review information on any product from a myriad of sources. Dpreiew is just one of those sources. When i am choosing equipment , I spend weeks /months on the web reading reviewa and comments from a wide range of sources and if possible i try to get my hand on the product for personal feel of what it can do. I find DP review to be unbiased and usually very accurate. But I dont rely on them exclusively, sites such as Imaging Resource, PopPhoto, Shutterbug, Photonet also inform my decision. There is no one holy grail site of Equipment review.</p>

[danielleetaylor]

<p>Allard K,</p>

<p>Don't those graphs illustrate the point that DxO also measures on a per pixel, or resolution normalized basis? That was my nitpick, that you have to take into account total image resolution and therefore enlargement factor.</p>

<p>http://www.dxomark.com/index.php/eng/Insights/More-pixels-offsets-noise!</p>

<p>Again, differences will be small between the two either way. Like I said, I'm nit picking.</p>

[dejansmaic]

<p>Huh?</p>

[stephen_asprey2]

<p>My experience of DPreview is positive. De Lenzer, I think you're feeling a little deflated concerning their review of the 5DMk2. The negatives they talk about concern the body (lack of competitive weather sealing - a fact confirmed by the recent trip by Luminous landscape people) and the fact that 21mp are always going to generate more noise than 13. What it has revealed is that there is a bigger gap between the top prosumer Canon (5DMk2) and their full pro range...the 1Dxxxx, than say between Nikons D300/700 and the D3. But there is also a bigger price difference too.<br>

The 5DMk2 weaknesses have been thrashed to death here over the past week. But that gives Canon something to correct for the 5DMk3.<br>

We should all get less precious and work harder to produce beautiful images, regardless of camera brand. Now if Nikon could get their pro lens range broadened, then I personally will be happier.</p><div></div>

[Nick D.]

<p>I think DP had the point, that there no reason for Canon 40D users to ditch they cameras and run to buy 50D and I agree with them. As for the ratings it's just one person opinion, 3 years ago I bought Fuji S3 PRO based on the samples from DPreview, camera was rated "above average", colors are wonderful, dynamic range unsurpassed even now, when I need speed I use Nikon D300.<br>

BTW, all sample pictures from new cameras looks the same to me in image quality and dynamic range no matter what brand of camera was used.<br>

It is just my opinion:) </p>

[liquidstereo]

<p>The reviews are in my opinion not useful after the page 5 or so.<br>

They have no ability to separate useful from non-useful. Or they choose not to do so. Additionally, their reviews have the stains of the reviewers. I.e. they bring their own experience and biases and they don't acknowledge this.</p>

<p>Cheers!</p>

[ralph_jensen]

<p>DPReview calls them as they see 'em. Yes, they're fallible, but they're also fairly thorough and increasingly transparent about their methods and choices. (Sean, to me they only get useful <strong>after </strong> page 5 or so! --because that's where they move beyond the press releases and product shots to actual testing of performance. Every review site has "the stain of its reviewers" - <em>in fact, almost every review site on the web is essentially the opinion of one person</em> - but DPReview publishes more bylines thanks to its larger staff, so it's easy to think their reviews are more "subjective" and less "objective.")</p>

<p>Back to the original point of this thread, the 50D: Because they didn't say the 50D was head-and-shoulders above the 40D, DPR's really-very-positive review of the 50D (it had 30 "pros" and only 7 "cons," as I noted above) only encouraged the conspiracy theorists who think DPR has a built-in anti-Canon bias. But it's pretty hard to argue that after yesterday's glowing review of the 5DII!<br>

http://www.dpreview.com/reviews/canoneos5Dmarkii/page40.asp</p>

<p><em>To the one poster above who avoids DPReview because their forums are so stupid:</em> you're only hurting yourself if you stay away from the entire site because of the forums. From news to glossary/learning to reviews, most of the site is well worth at least an occasional visit. DPReview <em>forums</em> , on the other hand, are essentially the punchline of the online photographic community; many photographers (including me) take one look at the ridiculousness in the forums and never go back to those forums again.</p>

<p>But to discount the entire site because of the one part that isn't generated by the site's employees would be like discontinuing a subscription to a newspaper or magazine because it publishes letters to the editor that you disagree with.</p>

<p>There are plenty of other good forums out there, including this one!</p>

[chris_seary]

<p>"Test charts and timings aren't everything, in fact some would argue that they tell you little about what the using the camera would actually be like. I'd like to think I write from the viewpoint of a skilled, trained, experienced photographer actually using a camera for photography."<br>

This is just nonsense. Test charts are the only evidence based way to assess the sensor, and test charts are what are used at Canon when they design the camera.<br>

I read this prejudice against test charts whenever a problem with Canon is demonstrated, such as their problems with autofocus. Every time someone writes in about an AF problem, the Canon worshippers come out with the usual nonsense:</p>

<ul>

<li>user error</li>

<li>test charts are meaningless</li>

<li>try taking the UV filter off the camera</li>

</ul>

<p>Like it or not, Canon are falling behind. Two years ago no-one was even close to them. Now they've made the mistake of preferring more pixels to better ISO performance (which needs less tightly packed pixels). Bad move.<br>

The AF problem is a biggie, and far too many people are finding Canon autofocus unreliable.</p>

[roger_irwin]

<p>Dxomark would seem to find similar conclusions, that the pixel for pixel quality of the 50D is not significantly better than it's predessor the 40D.<br>

But there is <strong>one BIG difference</strong> in that the 50D has 50% more pixels. That seems to me to be a significant improvement overall. Plus the 50D goes forther at the high iso end.</p>

<p> </p>

[duane_mills]

<p>DPReview has built and maintained a solid reputation for providing consistent, balanced and fair testing for the various makes and models of all manufacturers. Simply because your brand of choice doesn't recieve a warm and fuzzy love-in is irrelevant and in no way reflects on the staff or reputation of DPReview. I work in an event staging/management company where in addition to shooting video (on a variety of Sony products), still photography is an important component of many jobs. There are three of us in-house with different camera gear, the dslr bodies include 1 X Pentax istD, 1 X Pentax K100D Super, 2 X Pentax K20D's and (recently) 1 X Sony A700. For each of these platforms we have found DPReview to provide a fair analysis of each body, its performance and limitations. As a Pentax shooter I found the recent DPReview conclusions of the new KM somewhat disappointing but fair and well founded. I personally and professionally don't feel that any inference that DPReview is inaccurate, biased or unfair is either warranted or deserved.</p>

[rich_evans]

<p>I usually take a quick glance at DPR every day just to see what's new or newsworthy. While I've found the reviews to generally be helpful for someone not totally familiar with a brand or technology (consumer) who may be looking for a new camera, for those who are already entrenched in a system, I find the reviews to be mostly just technical exercises, since I've pretty much already made up my mind with regard to what I might need from my system anyway. I know this is a Canon forum and I'm a Nikon guy, but I've also found that the DPR site is pretty Canon-centric as well, although the Nikon reviews more recently have been pretty spot on. Take any review for what its worth - I first look at the last page and then delve into any of the tests which I may feel are important to me, and then move on to the next site. Caveat emptor - more info is better than less....... --Rich</p>

[iqbal]

<p><strong>DP Review: A Fact</strong><br>

I have a 40D and then I bought a 50D, but I was not satisfied with the overall quality in most situation what was I getting from 50D. Moroever, there was additional requirement on hardware in terms of processing time and storage. I thought, there is something wrong with my copy of 50D. Then I came to know about DP Review on 50D. I wish I would have read it earlier (any way I returned my 50D and replaced it with a used 40D). DPR is quite upto par in its conclusion (in fact, 50D should have just been "recommended," not more than that); and reason is quite simple. There is some rule of physics that marketing people (who pushed for 15mp in 50D) can not just wish them out. As many said, given the size of the sensor, 12MP is the limit. Beyond that, there are more negative outcome than the positive. But marketing people know the weakness and lack of knowledge of a dominant majority of consumers; and they succeeded in alluring many consumers to buy a diminishing quality product, 50D, comapred to 40D. Consumers who believe that every new toy is better than the old one; and they have no money constraint. </p>

[mageproductions]

<p>Well, what an outflowing! Many of the replies are cogent, in the extreme, while relatively few are trying to pick a fight. That speaks well, indeed, of the crowd here.<br>

The waters have been stirred. Now they are all but opaque, owing to all the different points of view and the different conclusions, they have caused. <br>

Allow me to gently, and politely try to return to the original point, which was that Dpreview told me not to trade my 40D for 50D, because the image quality would not be "significantly" better.<br>

When I did make that trade I found the quality of the images to be not only significantly, but thrillingly better. In another post I included this image:<br>

<img src="http://www.MageProductions.com/cabin.jpg" alt="" /><br>

I mentioned the complete lack of fringing where the sky shows through the branches - a consistent and very phony-looking problem I had with every cam since the 20D - and that the overall detail was so very much better. I did not mention how the detail makes the image look far more real...<br>

You are seeing this image resized, to fit in this forum. Right click on it and take it home. Look it over. There's nothing special about it - every image I get looks this good! (This one is using a 70-300 IS at about 300 feet away).<br>

I am not cheauvanistic. I did not participate in the development of this cam - lol. If I did not like it I was going to sell it on ebay. The auction contains a 40D, instead.<br>

After all the smoke clears, Dpreview told me wrong.</p>

[CharlesBecker-Toronto]

<p>"After all the smoke clears, Dpreview told me wrong." not really-what is 'significant' to one person may not be to another. It is just the reviewer's opinion-you can be guided by it but you shouldn't base your final decision on it. What was a problem with the viewfinder in the Nikon D70 reviews was not a problem for me so I bought it and used it happily for 4 years. A review is a guideline and nothing more. cb</p>

[iqbal]

<p>(1) DP review (or that matter, any other publication) is not in the business to tell an individual what to do.</p>

<p>(2) The view of a photo on screen does not tell much, unless you have reasonable size prints.</p>

<p>(3) Massive consumerism (because I have money --actually borrowed from Chinese and Arabs-- and there is a new toy to play with) has created this global financial and economic crisis. Still we did not learn, and waiting when Canon will introduce 60D with even higher MP on the same size sensor.</p>

[doug_williams4]

<p>Being new to digital I found DPR a useful site because they showed enough pictures to allow you to get an idea of what kind of image the camera produced. While I didn't understand all the technical analysis I was able to use the site because it contained enough images for me to form an option. They also have images that you can print from which I found very useful. The only thing I had to watch out for was that the tests did not always use the kit lens and outher better lenses were used for testing. </p>

[cc_chang1]

<p>I wish people can be a little more objective (scientific) and fair.</p>

<p>De Lenzer stated: "When I did make that trade I found the quality of the images to be not only significantly, but thrillingly better. In another post I included this image: I mentioned the complete lack of fringing where the sky shows through the branches - a consistent and very phony-looking problem ..."</p>

<p>First of all, if you want to show that 50D is better than 40D, you should show the same image captured by both under the same conditions in order to demonstrate that the 50D is "thrillingly better." </p>

<p>The major issues that DPreview raised about the 50D are the increase of high ISO noise and reduction in dynamic range when compared to the 40D. Do you have any evidence that the 50D in fact does better than the 40D in these two areas?</p>

 

[josephbraun]

<p>There is no one stop resource for info..at least for me.. I gather info from many sources. Take what you like and leave the rest :)</p>

[Landrum Kelly]

<p><strong>What DPReview actually says:</strong></p>

<p>"Looking at the specification differences between the EOS 40D and our test candidate it appears you pay quite a premium for the 50D's extra megapixels and as we've found out during this review you don't get an awful lot of extra image quality for your money. The Canon EOS 50D still earns itself our highest reward but considering its price point and our slight concerns about its pixel-packed sensor, it only does so by a whisker."</p>

<p>I really cannot see why you have interpreted this as a negative review of the 50D. These remarks are not highly negative. To the contrary, they acknowledge the superiority of the 50D over the 40D. They simply raise the question as to whether or not the improvement is worth the extra money. That is for the buyer to decide, and I think that Phil Askey et al. are to be congratulated for raising the kinds of questions that we have come to expect from them.</p>

<p>--Lannie</p>

<p> </p>

[naturetrek]

<p>Reminds me of the Canon G3 vs. G5 war. Every disgruntled G3 owner was bashing the G5 just because it wasn't the "upgrade" they were waiting for, but nonetheless, was "just enough" better to piss them off :)</p>