Author: Günther Nubert
Oct 10, 2001, 12:00:00 AM
Response to the guestbook entry by Ali (8.10.01) and Stefan Redlich (9.10.01). A small statement (as suggested by Mr. Redlich) is unfortunately not enough to bring some clarity to the results of the tests mentioned. From conversations with journalists from various specialist magazines, I have gained the impression that the aim of the vast majority of test editors is to provide readers with information that is helpful when selecting hi-fi components. However, the test in ComputerBild 12/99 lacked EXACTLY THAT. I imagine that all of our customers will understand that we did not want to advertise our result, which NONE OF THE 10 TEST CANDIDATES exceeded. We are of the opinion that this was a blatant misjudgment and we believe we can prove it. After the test, I made several phone calls and wrote three (sometimes vehement) letters to the ComputerBild editorial team. Finally, we were invited to a meeting in Hamburg, followed by an invitation to the most important speaker developers in Germany to work on improving the test criteria in a committee. Some of the assessment criteria (such as the rather radical load test) used in the 12/99 test were subsequently revised. A kind of truce has also been reached and we are hopeful that future tests will be more realistic. I actually thought that time heals wounds and I wanted to forget this test as quickly as possible. But now, unfortunately, I have to dig out some facts about it AGAIN: 10 sets were tested, which in our opinion sounded somewhere between a VW Polo and an S-Class Mercedes (in a REALISTIC test they would have earned a rating range of around 1.5 to 4). 8 sets received a sound grade of 4.0 and 2 sets received a sound grade of 5.0. In the entire multi-page test, there was only one tiny reference to the sound quality of the adequate sets: ...all other speakers received adequate grades, with the systems from Teufel and Nubert making the best impression. Funnily enough, the two best-sounding sets were ranked 7th and 9th and the set that was described as lively but a little shrill was ranked 1st. Because sound differences were not included in the evaluation, the ranking was based on other characteristics, such as required amplifier power. Unfortunately, the physical relationships were turned upside down. The fact that our nuBoxes 460 and CS-3 with the SAME AMPLIFIER already achieve significantly higher volumes than the speakers that were rated 1 for the recommended amplifier power means that they should have been rated 1 for a LONG time. - But because they can also emit FAR MORE POWER without distortion, they were given a score of 5!!! - WHO WANTS TO UNDERSTAND THAT! - The mistakes made here were so great that the ratings of most of the speakers in the test field were reversed! Our nuBox 460 and CS-3 were rated 5 because they supposedly require extremely high amplifier power! - However, almost all of the speakers that require MUCH MORE amplifier power for the same volume and would have earned a score of 5 were given a score of 1! As a result of this distortion, some of the sets were pushed far up in the overall ranking! The assessment of SOUND INTERFERENCE DUE TO CLIR FACTOR, which is much more important than the required amplifier power and from which the second and third-placed sets in particular suffered, NO LONGER HAD THE WEIGHT to oust these sets from their places. What is a reader supposed to do with a test in which the TEST WINNER received the test result "sufficient" and "poor price/performance"? A sad thing for the informative value of the test: Correcting the errors in the recommended amplifier power would completely change the order of the rankings - although EVEN THEN the sound quality of the speakers would still not be decisive for the overall result! - A reference to this test has been available for several years in our Technik Satt booklet, which can also be downloaded from our web-page, in the Dolby Surround chapter... (construction details, last section) and commentary on measurement methods... (speaker load test, paragraphs 1 and 2). If the rated power handling of our speakers had been 20 WATT each with otherwise unchanged characteristics, this set would have risen from an initial score of 4.05 (even 4.13 after correction of a calculation error by ComputerBild) to 3.42 with unchanged evaluation criteria - this set would then have been the ONLY one to win the trophy with a (dream score?) overall score of satisfactory despite its only adequate sound! Günther Nubert