ABOUT THE MEASUREMENTS
The first measurements taken were a 1 watt/1meter response on tweeter axis of each speaker. One will be shown in red and the other in blue. This allows one to see the variation between the pair. Some were very consistent to one another and some were not. Then the one with the best, or what appeared to be the most accurate, was used for all other acoustic measurements.
From the measured on axis responses the best response of the two speakers shown by its self. Tweeter axis was chosen as a reference as it is often the reference design axis for the manufacturer and this is typically ear level on a normal stand.
Second is a cumulative spectral decay. This shows how fast the speaker dissipates stored energy and inertia. This will highlight any stored energy problems or uncontrolled break up somewhere in the speakers response. A long decay in a given area can be heard as a smear or ringing. It can cause notes to run together causing a loss of resolution or it can cause a harshness that can be very fatiguing. These measurements really shouldn't be compared to CSD's taken by others, but is very good in comparing all of these speakers that were all measured in the same way and with the same measurement system.
Next is the horizontal off axis response. These are taken at 0, 10, 20, 30, and 40 degrees off axis and stayed in line with the tweeter. These are seen in red, orange, yellow, green, and blue respectively. This plus some vertical off axis measurements can be a clue as to what the total in room response might look like. Most speakers faired within a normal range with this measurement. Some allowed the top end to drop out more quickly than others. Those speakers might be a little more forgiving of brighter rooms, but those with a wider dispersions might allow for a wider and more enveloping sound stage.
Then there is a vertical off axis response. There really is no accepted industry standard to this type of measurement. The vertical off axis response will certainly contribute to the total room response in a positive or negative way but some manufacturers put little value on it. This measurement can also tell you what the speakers will sound like if listening from above the tweeter axis from a typical standing position. Some people may want a consistent sound seated or standing. Some people could care less and will only listen when seating in front of the speakers.This measurement is also an indicator of how well the drivers are acoustically in phase. Some speakers don't really allow the drivers to be in phase very well (time arrival of one driver is not in sync with the other). Moving up or even down can allow one to see if this mis-alignment gets better or worse as driver distances change in relation to one another. If a speaker showed a mis-alignment in the vertical off axis then I may measure it below the tweeter axis or near the woofer axis to see if the alignment improves. Some speakers have been known to have a poor alignment if moving in the direction of the tweeter but a better alignment if moving in the direction of the woofer and may be preferred listened to with the speaker flipped over placing the woofer above the tweeter. These vertical off axis measurements were again taken at 1 meter with the first being on tweeter axis (red curve). Each additional measurement was made by moving the mic up 4" (orange, yellow and green).
Lastly was the impedance sweep. This can be very telling as to how hard or how easy of a load it will present to an amplifier and if it will present a consistent load over a wide range or a variable load over a wide range. Additional measurements are shown on some speakers that had built in adjustability in the response or that had a response issue that prompted further measurements to resolve questions that the on axis response may have left unanswered. Also some speakers had dual binding posts allowing for bi-wiring or bi-amping. This allowed the woofer and tweeter curves to be shown independently of one another and those were also included. Also, if grills were provided then an additional measurement was made with the grill on (in blue) and overlaid on a response with no grill (in red) so that the effects of the grill can be seen.
The Epos M12.2
This is the response on tweeter axis. The response is a little choppy but within normal industry standards. Baffle step compensation is minimal and it could give a thinner sound to the vocal region as it is several db down below about 700Hz and below. By contrast the peaked area in the 800Hz to 1000Hz range may be perceived as a brightness.
Since this speaker has dual binding posts it easily allowed the individual drivers to be measured. The crossover is very high and in the 5.5kHz range. Drivers are not in phase very well and the woofers response is canceling some of the tweeters output in the 8kHz to 16kHz range.
Since the drivers did not appear to be in phase very well from the tweeter axis the speaker was also measured above and below the tweeter axis to see if there was a height in which the drivers were more in phase. Here an on axis response (red) is overlaid with a response taken 4" below the tweeter axis (blue). We see less disruption between the drivers from this height and a smoother overall response. This would be a more ideal listening height for this speaker.
So at a height of 4" below the tweeter axis we see the response of the pair of speakers. The speaker with the response shown by the blue line had a tweeter that was slightly more efficient than the one that produced the red line. With the speaker producing the red line being closer to accurate, it was used for the rest of the measurements.
The spectral decay is not too bad but there is some stored energy shown at each octave. The one in the woofer range could cause some smearing while the small resonances in the upper ranges might add a bit of ringing.
The horizontal off axis response shows a pretty even decay across the top end but quite a bit of loss across the board. These were all made at 4" below tweeter axis.
The vertical off axis is where things look a little out of whack for the Epos. Here we see the response on tweeter axis in red. Just 4" up (in orange) we see quite a bit of cancellation at 6.6kHz. Just 4" higher again we see and even deeper dip at 5.6kHz. Up an additional 4" and the heavy dip shifts to 4.6kHz.
The impedance looks pretty good and the tuning frequency of the rear firing port shows to be in the 55Hz range. There is one small bump just above 600Hz that coincides with some stored energy in the woofer range. This could be a woofer or damping material issue. Adding or changing the type of damping material used might solve this.
Finally here is the response on tweeter axis with and without the grill.
Here we see the on axis response of the pair of speakers. Clearly there are some differences in the output of the tweeters in this pair.
Here we see the response from the better of the two. There is a dipped area at what is likely the crossover point, but it maintains a fairly smooth response otherwise.
Spectral decay is fairly clean with just a little stored energy in the upper mid-range.
Off axis horizontal response is pretty good with quite a bit of output loss that is fairly even across the board.
Off axis vertical response shows some additional output loss in the 3kHz range but not too much more than the loss already inherent to the basic design.
The impedance is a fairly easy load to drive with a slightly higher load across the top end. The tuning frequency of the port is at 60hz.
The effects of the grill are very minimal with this speaker. It was a very nice grill design.
The first response curve shows the pair of review samples. Deviations between the two are within normal industry standards.
The on axis response is fairly smooth but the peaked area in the 700 to 1kHz range might add a touch of brightness or edge to the upper vocal range.
Shooting the individual drivers showed them to be well in phase and summing with one another just as they should. The crossover point looks to be in the 2.4 kHz range.
The spectral decay is fairly clean with just a few areas of stored energy showing up as thin ridge lines in the upper ranges and a little bit of a lump in the upper mid-range.
The Outlaw Audio speaker also had some three way toggle switches on the back of the speaker that would allow output adjustment to the bottom end and the top end.
The adjustment to the bottom end was called boundary adjustments. It had a minor effect on the response above 200Hz but was fairly significant below 200Hz.
Electrically it presented quite a bit of resistive load to the amplifier as the adjustments are made.
The more alarming aspect of the impedance has to be the dipped area at 3,375Hz where it dips to 3.2 ohms. Here is the impedance with the controls on both ends set to flat and with no adjustments.
The adjustments on the top end had a pretty profound effect lowering it by over a full db or raising it nearly 2db.
The adjustments to the top end also had some effect on the impedance up top as well. Increase the load and reducing the tweeter level raised the impedance slightly. Lowering the resistive load and raising the tweeter level also reduced the impedance to an alarming 2.9 ohms.
Horizontal off axis response looks fairly good.
Vertical off axis looks pretty good too. Drivers start to get a little out of phase once up to 12" over tweeter axis (remember only 1 meter away) but this is not too bad.
Effects of the grill were not too bad either.
Overall this is not a bad measuring speaker except for the really low impedance load on the top end. Listening was done with the levels set to the neutral positions as this was where they had the the most accurate response curve.
Measuring the pair showed quite a bit of difference between the two speakers. The woofer response of one of them was well off from the other.
Publisher's note: During the listening tests, it was determined that there was a physical problem with one of the Silverline's drivers.
Did this effect the performance of the Silverlines in the shootout? Yes. Definitely.
Here are the measurements for the properly working speaker:
The response stayed within an acceptable range but was very choppy. Most troubling was the peak in the 700 to 900Hz range.
Looking at the crossover response we can see that some of that choppiness comes from the tweeters response. It looks like it is using a low order (possibly first order) crossover. The actually crossover point is in the 3.2kHz range. The roll off of the top end of the woofers response is very smooth but in the 6.5kHz to 10kHz range it is cancelling some of the tweeters output.
The spectral decay is fairly clean other than the lump in the 700 to 900Hz range. There is clearly some stored energy there.
Horizontal off axis looks pretty good.
The vertical off axis shows some cancellation in the crossover range.
Moving down to the woofer axis we now see cancellation in the 2kHz range. It looks like this speaker has a very narrow window (vertically) where the response stays within an acceptable range.
The impedance is much higher on the tweeter end than the woofer end. Some amps will cause this impedance mis-match to yield an output mis-match. So the response may vary from amp to amp.
The grill on the Silverline was at stretched across a wood frame that was at least an inch thick. This cupped the drivers and caused a lot of interference from those reflections. This was the most disruptive grill of the group.
The on axis response was pretty choppy but one was worse than the other.
This was the best response of the pair. It looks like it had no baffle step compensation.
The spectral decay showed some pretty bad stored energy in a several areas, but most significantly in the range around 1kHz.
Off axis responses in the horizontal plain reveals that it would not be good to listen to these any further than 10 degrees off axis. Maybe it would be a good choice for an overly lively room as there is not a lot of off axis energy. Unfortunately it is not a smooth off axis response
Vertical off axis shows the drivers remain in phase almost as well as they were on axis but still rough all over.
The impedance shows a pretty easy load to drive and just one wiggle in the 55Hz range that looks like it could be an internal resonance.
In the grand scheme of things the grill had little effect in making the response any worse that it already was.
Shooting a response of the pair of LSA-1's we see pretty good consistency with some variation near the crossover region. This could be an inconsistency with the crossover parts or the drivers. Without further measurements it is hard to say, but the drivers themselves are very consistent outside the crossover region.
Here we see the on axis response of a single unit. The top end is a little hotter than the lower end and this speaker my be perceived has being overly bright in tonal balance.
Since the LSA-1 has dual binding post it was easy to shoot individual driver responses. We see a crossover point in the 3.1kHz range. We can also see that the drivers are not summing with each other very well (not in phase very well), and the woofers upper response is canceling the tweeters output in the 4.5kHz to 7.5kHz range. There is also a bump in the woofers response in the 3kHz range that could be some cone break up that is not being controlled by the crossover.
Looking at the spectral decay we can see some stored energy in the 1kHz range but it is fairly low in level. There is also some stored energy just above the crossover point.
Looking only at a spectral decay of the tweeter we can see that some of the stored energy just above the crossover point is in the tweeters response, but not all of it. It is really not too bad compared to most speakers.
Looking at a spectral decay of just the woofer we can see that some of the break up in the upper range is clearly coming from the woofer, but the ridge line present in the 1kHz range may be more of a problem.
The off axis horizontal response looks pretty good with just a slight peaked area in the 5kHz range that is not very prominent until 40 degrees off axis.
Not only are the drivers not in phase very well on axis but in the vertical off axis they become further out of phase causing complete cancellation in the vertical off axis. This is not a speaker you'd want to listen to while standing.
Moving the microphone to 4" below the tweeter axis (blue line) shows that the drivers are now more in phase, but the natural response of the two drivers coupling created quite the peak in the 4kHz to 6kHz range. If you look at the individual driver responses from above you can see that this response now more closely follows that of the drivers.
The impedance is a little on the low side at the woofer end dipping to 4.5 ohms. Tuning frequency is pretty low for a speaker of this size too. It is tuned to 35Hz.
Dropping the grill on had quite a bit of an effect in the upper range. Listening without it for critical listening my be greatly preferred.
THE UNITY AUDIOS
This response is somewhat typical of a small full range driver being run with no network on it. With no baffle step compensation circuit the output drops drastically as the longer wavelengths become more omni directional and wrap around the small box giving up half of its output in the opposite direction from the direction the speaker is facing. We see it in the response as a loss of output below 1kHz.
Overall response is pretty rough and some of those peaks in the 4 to 8kHz range look like they might be a real problem.
The spectral decay confirms the stored energy level in those peaked areas. This might equate to a considerable amount of ringing in that range.
Since this is a single full range driver the off axis is the same in all directions and just like any other driver its size it falls off rather quickly in the top end.
At least it is an easy load to drive.