There’s nothing like measurements to really get an audiophile out of their chair and onto an Internet forum! My unreview of the DeVore Fidelity Orangutan O/96 loudspeaker generated a few comments and a lot of email, which is always entertaining, but it also elicited a response from John DeVore, the designer.
The problem with measurements is that they’re just that, measurements. Numbers can tell you a lot of things, but one of the things you can lose, and lose quickly, is any sense of context.
John’s note, included here, provides some of that.
DeVore Fidelity Orangutan O/96 Stereophile measurement comparisons
The three tables that follow are comparisons of the measurements published in Stereophile magazine between the DeVore Fidelity Orangutan O/96 and eleven other speakers. The three speakers in Table 1 were selected from reviews published during 2012 and represent highly-regarded full-range “conventional” speakers that are not particularly easy to drive with a low-powered tube amp. The same four measurements are used for each speaker. They were the three models that were at least the price of the O/96 and were reviewed most recently. The three speakers in Table 2 represent speakers that are intended to be easy to drive with a low-powered tube amplifier. I had to use some reviews published prior to 2012 in order to find three relevant examples, as this type of speaker is far less common than the “conventional” speakers in Table 1. Table 3 focuses solely on the horizontal dispersion measurement, which I will get to below.
All measurements were performed by John Atkinson and published in Stereophile Magazine. We make no claim other than that they have not been altered beyond fitting them on the pages.For the first two tables the same four measurements were used from each speaker review in order to show a more detailed comparison:
1-Frequency Response: Mr. Atkinson’s quasi-anechoic measurement of the frequency response of the speaker from 10Hz to 30kHz. A fundamental measurement that shows how evenly all the frequencies are reproduced by the speaker, as well as how high and low in frequency the speaker play.
2-Impedance: The impedance of the speaker that would be presented to the amplifier. The lowest value is noted in red for each speaker.
3-Step Response: A measurement of the time-domain performance of the speaker. An “ideal” speaker measurement would look like a right-triangle, with a vertical rise from “0” and the hypotenuse moving back down to “0”. An upside-down triangle generally means one or more drivers are connected in reverse-polarity. A sloping initial rise can indicate slower rise-time of a driver. Multiple spikes and uneven traces can indicate poor time-domain integration of drivers.
4-Waterfall Plot: This is also a measurement of the time-domain performance of the speaker. An ideal plot would be a flat “wall” along the very rear of the “3-D” graph, with no ridges protruding towards the bottom of the graph. Each ridge represents delayed energy released by the speaker.
Examining the Table 1 Frequency Responses, the Orangutan O/96 is either the best of the four speakers or tied for the best with the $41,000 Lansche Audio 5.1 with very even reproduction of the audio frequencies.
Looking at the Impedance column the O/96 with a minimum of 7.2 is clearly the best for a lower-powered tube amplifier. The Wilson and Lansche speakers are a very difficult load for any tube amp.
In the Step Response column again the O/96 displays the best response and is the only one to have an approximately right-triangle shape. The Wilson speakers reverse the polarity of the midrange driver and the other two show some lag after the tweeter rise.
In the Waterfall Plot we see all the speakers exhibiting significant delayed energy in the midrange region (the left-hand side of the graph). The treble of the O/96 is cleaner than both the $200,000 Wilsons and the Martens (noted by how far down the “wall” the hash lines and ridges start on the right and center of the graph). There is one ridge noticeable just below the clean treble region in the O/96 caused by some delayed energy from the woofer, though some of this is evident in all four of the speakers.
In Table 2 the DeVore Fidelity speaker is by far the best in the Frequency Response column. The closest second is the Avantgarde speaker with a self-powered woofer section, though there are clearly a number of problems visible with the integration of the tweeter and midrange drivers with the woofer section.
The Impedance column shows that these speakers are all far more easily driven than the speakers in Table 1, with the O/96 being the second easiest to drive after the 11 Ohm Voxativ.
The O/96 is again the best measuring speaker in the Step Response column, with both the JBL and the Avantgarde showing poor time-domain integration of the drivers as multiple vertical spikes in the response.
In the Waterfall Plot column the O/96 has far cleaner treble response than any of the other three speakers, and the midrange ridge is narrower and more controlled than the ridge in the Avantgarde.
Looking now at Table 3 we have the O/96 horizontal dispersion plot compared to similar plots from six other respected speakers. This is the measurement that has been cited as a weak point of a design like the Orangutans but in fact this Table illustrates that it is at least above average if not very good in this regard. What the perfect plot should look like is open to debate, but both Mr. Atkinson and Dr. Floyd Toole agree that the ideal off-axis response of a speaker should decrease smoothly with increasing frequency. This results in a plot that steepens gradually as you move to the right, higher in frequency. I only chose speakers that have similar or “worse” plots from respected companies that are also more expensive than the Orangutans. All of the plots have a dip to at least -18dB in the upper midrange, with the Wilson, Avantgarde and Klipsch dipping even lower. All the plots are significantly less smooth than the O/96.
What I believe is clear from these comparisons is that my design goal for the Orangutan O/96 of creating a speaker that is very easy to drive with a tube amplifier but that does not suffer from the usual problems that this type of speaker is known for, such as uneven frequency response, poor driver integration, significant coloration or difficult set-up has been achieved.
In fact, while the Orangutan is among the easiest to drive of the eleven speakers shown, it is also among the best measuring, including the difficult to drive “conventional” speakers. It is also the smallest speaker of the group by far, among the easiest to set-up in a room and less expensive than all but one.