JMlab Utopia loudspeaker Measurements part 2
At 42" from the ground, the tweeter axis is a little high for a typical ear height in a typical listening chair. Fig.5 shows how the speaker's response changes as the measuring microphone is moved above and below that axis. (The off-axis responses are normalized to the on-axis response, which means that the latter appears to be a straight line.) It can be seen that the balance doesn't change significantly between +5 degrees and -10 degrees ref. the tweeter axis, meaning that JMlab has sensibly optimized the crossover design for seated listeners. In the lateral plane (fig.6), the Utopia offers wide, even dispersion, though again there are a few more ripples in the mid-treble region, presumably due to reflections of the tweeter's output from the edges of the wide baffle. Note that the region above the 1100Hz on-axis peak fills in to the speaker's sides, which might correlate with J-10's finding that the speaker worked best when it wasn't toed-in all the way to the listening position. (Note that the data points in this graph are more sparsely spaced than is usually the case. This is due to the fact that the speaker is too heavy to be measured on my usual turntable; the off-axis responses had to be obtained by manually moving the speaker.)
Fig.5 JMlab Utopia, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 20 degrees-5 degrees above-axis; reference response; differences in response 5 degrees-10 degrees below-axis.
Fig.6 JMlab Utopia, horizontal response family at 50", normalized to response on tweeter axis, from back to front: differences in response 90 degrees-5 degrees off-axis; reference response; differences in response 5 degrees-90 degrees off-axis.
In the time domain, the Utopia's step response (fig.7) indicates that the tweeter, midrange units, and woofer are all connected with the same positive acoustic polarity, but that the speaker's presentation (on its tweeter axis) is not time-coherent. Considering that Jonathan found the Utopia to sound so transparent, I was surprised to find that the waterfall plot on its tweeter axis (fig.8) showed some mid-treble hash. However, this measurement is susceptible to early reflections, which also give rise to delayed energy; I suspect that this is the reason for this hash, rather than resonances per se.
Fig.7 JMlab Utopia, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).
Fig.8 JMlab Utopia, cumulative spectral-decay plot at 50" (0.15ms risetime).
However, note that while there is no ridge of delayed energy associated with the on-axis peak at 1100Hz, which might explain why it appeared to have only a small effect on the Utopia's perceived balanced, there is such a ridge associated with the smaller response peak just under 4kHz. I would have thought this to have been audible, perhaps as a slight "etching" of recorded detail. I do note that J-10 wrote that "too great a toe-in sounded a touch zingy on certain material, even as focus improved...," which might well correlate with the presence of this resonant mode. Again, it appears that experimenting with toe-in will minimize its subjective effect.—John Atkinson