I find them quite valuable...
Sure. There is always value in such information than having such information. But the unintended consequences of the type of reviews here also make them suspect for practical use.
-Reviews have uncovered things broken (some manufacturers have addressed, others have ignored).
How many of them were really show-stoppers (let us stick to mainstream, known brand electronics)? Seriously. The problem is that the decapitated panther as a measure of "broken" gives both false positives and false negatives.
Yes, you can always say but it is up to the reader to figure out the exact situation, read the charts, read between the lines, etc and then make the decision. Real world doesn't work that way. It is like saying they shouldn't have rioted from what was said so not responsible because they should have understood better.
Is the HTP-1 "broken"? In theoretical or academical sense, perhaps. But has very little practical significance. Replacing that with an AVR as a pre/pro instead (because it measured fine without edge cases) but with serious compromises in functionality or features is a bad outcome of such adherence to a method. Because of certain level of sycophancy in the group (happens anywhere in any forum) negative feedback on the review are dismissed as someone being "butt-hurt" which reinforces the "absoluteness" of the review for new or casual readers.
The danger of the limitations of the measurements here are:
1. While everyone agrees, in theory, that this should not be the only criterion, people (especially newcomers and people who may be browsing that we don't even know) seem to make routine decisions on what to buy and what not to buy based on those measurements and often on bad interpretations of those or worse on what the panther denotes. This thread started with such an assumption about the Denon. Frequently see such statements of bad interpretations made. If you want to blame this entirely on the reader, then that is not entirely fair.
2. People often (as expressed frequently here) do not consider equipment that has not been measured (given the limitations if sampling here) which is also a form of false negative and will likely lead to sub-optimal decisions.
3. People often spend huge amounts of money for equipment that "measures well" either misled to think that it sounds better or that they would face some unknown horror if they didn't buy something at the top of the charts. This suffers from the same false positives as people only listening to top of the charts music (however objective the criterion might be).
-Provide helpful information in putting together a system. For example the performance of the output of preouts on AVRs. Some can provide output of 2V+ and some can do much less. With this info, you can pick an amp to match a particular receiver for best performance.
Sure, but that is like putting out a measured spec-sheet. But the reviews are more than that. There are subjective interpretations in the review (often not so obvious), there are over-all conclusions that eliminate equipment from consideration for criterion that would make no difference ("in this day and age we expect better performance") despite a better-fit for the requirement, etc.
-Receiver amp performance. There is a very large variation in amp quality in receivers. Look no further than the NAD T758V3 vs Denon X3700. Not only in quality, but quantity. Measuring 2 channel output and 5 channel output is quite helpful in determining if external amplification is necessary
What exactly in these two measurements would it result in a practical selection of one over the other or whether to use external amp. How do the SINAD or power envelope differences fare in comparison to how much heat each puts out trying to drive less than efficient speakers? Were the difference in audible range? Eventually, it all boild sown to would have expected better (but not it is inadequate).
-DAC SINAD I think is important up to a certain point.. that point seems to be dependent on the noise floor of your room and listening volume. For dedicated home theater users in very quite rooms that listed near reference level, I wouldn't want something measuring with a SINAD of 70. In a living room with a high noise floor, where listening level is 20-30db less than reference it probably wouldn't matter much.
This is always argued in the extremes. How many of the currently manufactured measured equipment has DAC SINAD in the 70s? How about 85+? How about 90+? How about 95+? SINAD chasing encourages the other extreme in fact often at cost of features.
-Measurements are starting to get to be part of shopping for more and more users. Even looking at AVSForum, people shopping are now considering measurements in their shopping decisions. In most posts there, when someone asks what to purchase there are generally multiple referenced to tests here. The value is if customers are using bench tests as part of their purchase decisions, manufacturers take notice and put more effort in the performance of their products.
In theory, yes. It is the unintended consequence of false positives and false negatives that makes them not very suitable in practice.
I am all for manufacturers/reviewers making more measurements and designing better equipment. That isn't onconsistent with my statement on the form of reviews here with its limitations being of use in a practical sense now.
This isn't to say that bench tests like provided here should be the sole determining factor in a purchase decision. Things you have mentioned like reliability, features, EQ, stability are in many ways just as important. But I think dismissing bench tests/reviews such as Amir is doing is also a mistake. Just my opinion of course. YMMV.
This has the danger of going into the false binary choice of Amir or no Amir and drawing the usual critic vs sycophant debate. So, I object to your characterization as being irresponsible.
My statement was a very specific one. Of its use in practice for making equipment selection decisions (all the CYA caveats for its use notwithstanding). This doesn't mean it has no value. Don't artificially force that binary choice.