Posts: 363
Threads: 11
Joined: Jan 2015
Reputation:
31
The current update as per previous updates has often elicited various comments about the change to sound quality and to the behaviour of the Spark volume control.
The latest update has found me with the time and inclination to perform some simple measurements.
I hope they prove useful and provoke discussion but, hopefully, not argument. They are, after all, based on my devices, positioned in my environment were they are located for normal listening.
Measurements made via Room EQ Wizard (REW V5.12) with a 256k log sweep from 15Hz to 20,000 Hz at -12 dB.
A UMIK-1 mic, with its’ 90 degree calibration file, was used pointed directly at a single Phantom at a distance of 30cms.
The attached images show the SPL responses for the measured frequency range with variable smoothing applied.
Image 1
Measurements for each Spark volume setting from 35 to 60, in 5 unit intervals, were made for the pre and post Dialog/Phantom 1.4.1.12 update.
Each pair of measurements were very similar to each other but, for clarity, only three are shown.
[attachment=747]
Image 2
Measurements for each Spark volume setting from 35 to 60, in 5 unit intervals, were made for the post Dialog/Phantom 1.4.1.12 update.
It can be seen that the Spark volume units can be considered to be decibels and that the scale is linear, at least from Spark volume 35 – 60.
[attachment=748]
Image 3
Measurements for Spark volume setting 50 were made with/without Night mode engaged.
[attachment=749]
Image 4
Measurements for Spark volume setting 50 were made with/without Night mode engaged.
Experimentation with various high pass filters indicates something close to a Butterworth 6dB/octave with cut off frequency of 45Hz filter is a close match.
[attachment=750]
Image 5
Measurement for Spark volume setting 50 was made with the mic at the main listening position.
This shows a more accurate representation of the Phantoms’ bass capability and how the high frequencies are ameliorated.
[attachment=751]
Posts: 492
Threads: 95
Joined: Dec 2014
Reputation:
9
Thanks. Very intersting. I'll take some time to figure out what I'm seeing but a few questions.
30cm away is helpful to know. Is the Phantom in the middle of a room or near a wall? if in middle of room, about how large the room? On the floor or stand? How high the stand if on stand?
I'm wondering about the dip at the 130hz or so area. Perhaps room interaction? Also at your listening
Finally, one of the questions some including myself are asking is how the sound changed from prior to update and after update. Many including myself noticed better highs, air, resolution and balance. I really wondered how they did that if they did. My guess is that they accentuated the upper frequencies and detuned the bass. To me this is the most interesting question of all - how they changed the frequency response to give that sense. I doubt that they can make the drivers actually respond faster but it just feels faster. Transients feel better.
Posts: 30
Threads: 0
Joined: Nov 2015
Reputation:
0
Do you use the Radiance XD to handle the audio delay?
Posts: 141
Threads: 7
Joined: Feb 2015
Reputation:
6
(06-Nov-2015, 21:32)alandbush Wrote: The current update as per previous updates has often elicited various comments about the change to sound quality and to the behaviour of the Spark volume control.
The latest update has found me with the time and inclination to perform some simple measurements.
I hope they prove useful and provoke discussion but, hopefully, not argument. They are, after all, based on my devices, positioned in my environment were they are located for normal listening.
Thank you for taking the time, this is great!
Do you have any thoughts about what is going on in the 100-200hz area? Is that where the crossover to the side woofers is?
Sent from my iPhone using Tapatalk
Posts: 77
Threads: 5
Joined: Feb 2015
Reputation:
1
(06-Nov-2015, 21:32)alandbush Wrote: The latest update has found me with the time and inclination to perform some simple measurements. Really nice post, thanks a lot for the graphs! Great work! 
If you still have a little bit more time and inclination, it would be nice if you could repeat just the volume level 50 test in another room, or, if in the same room, just turn your phantoms to the other wall, so we could try to get an idea of how much the room/orientation would affect Phantom's performance.
The bass so high matches what some people hear, music with too much bass... probably the room is reinforcing the bass too much in your case... What's your opinion about it? Does the bass sounds right to you?
Posts: 492
Threads: 95
Joined: Dec 2014
Reputation:
9
To me, I would actually like MORE bass as the volume drops. So in the 15-30 zone for late night listening. In my other system with dual subwoofers, I don't drop the sub as much and it really fills out the music at low volume. But yes, at higher volumes, the bass is already quite strong. Don't know if it is too much - but it is definitely sufficient.
Posts: 363
Threads: 11
Joined: Jan 2015
Reputation:
31
07-Nov-2015, 01:41
(This post was last modified: 07-Nov-2015, 01:51 by alandbush.)
Hi guys and thanks for the comments and observations.
Sorry for the confusion that has been caused by the peaks and dips in the SPL responses.
I am guilty of not providing enough information as to how I expected the measurements to be interpreted.
I was NOT attempting to provide a measurement similar to that provided by some audio mags, eg. Hi-Fi News. I do not have the equipment nor the experience to do so and I suspect the side firing woofers make the whole process incompatible with other speakers.
I was attempting to provide data that could be used to compare firmwares and Spark volume changes regardless of the flatness of the frequency curve.
So my interpretations are as follows
Image 1 - indicates NO significant change in the SPL across the measured frequency range for most of the usable volume between the two firmwares, i.e. Any SQ change is due to some other cause.
Image 2 - indicates that Spark volume affects the measured frequency range equally and that it is linear
Image 5 - indicates, that in my MLP, the Phantom is very capable at producing bass volume down to at least 20Hz. This is undoubtedly due to the room size, speaker position and orientation as well as MLP etc. Etc.
Yes, the bass is fantastic but eventually becomes a little tiring. Fortunately, I use a MiniDSP to apply EQ.
Posts: 363
Threads: 11
Joined: Jan 2015
Reputation:
31
07-Nov-2015, 01:47
(This post was last modified: 07-Nov-2015, 01:48 by alandbush.)
(06-Nov-2015, 23:36)AlanD Wrote: Do you use the Radiance XD to handle the audio delay?
Indirectly. It cannot specifically delay the video but it is probably increasing the video processing and together with the TV processing I have little problem with lip sync.
Posts: 141
Threads: 7
Joined: Feb 2015
Reputation:
6
I see, thanks for the clarification!
In my room, the bass seems about right. Or at least very enjoyable and not overwhelming the rest of the music!
Perhaps the perceived change in sound is due, as you suggest, to some other aspect of processing - maybe how Spark is working as a player or how the DACs are rendering the file in the phantoms or something like that.
Posts: 492
Threads: 95
Joined: Dec 2014
Reputation:
9
(07-Nov-2015, 01:41)alandbush Wrote: Hi guys and thanks for the comments and observations.
Sorry for the confusion that has been caused by the peaks and dips in the SPL responses.
I am guilty of not providing enough information as to how I expected the measurements to be interpreted.
I was NOT attempting to provide a measurement similar to that provided by some audio mags, eg. Hi-Fi News. I do not have the equipment nor the experience to do so and I suspect the side firing woofers make the whole process incompatible with other speakers.
I was attempting to provide data that could be used to compare firmwares and Spark volume changes regardless of the flatness of the frequency curve.
So my interpretations are as follows
Image 1 - indicates NO significant change in the SPL across the measured frequency range for most of the usable volume between the two firmwares, i.e. Any SQ change is due to some other cause.
Image 2 - indicates that Spark volume affects the measured frequency range equally and that it is linear
Image 5 - indicates, that in my MLP, the Phantom is very capable at producing bass volume down to at least 20Hz. This is undoubtedly due to the room size, speaker position and orientation as well as MLP etc. Etc.
Yes, the bass is fantastic but eventually becomes a little tiring. Fortunately, I use a MiniDSP to apply EQ.
Alandbush, by using minidsp, I assume that you are using optical in to the Dialog if you are running stereo, or directly to a single Phantom?
I can't figure out how one can use miniDsp if one was using Spark to steam online services such as tidal or using Spark to play music in one's own server.
Thanks again for generating the chart. Very interesting to see that there is no boost in the bass when volume gets turned down. Some had speculated that earlier versions of Spark did that.
MtnGuy
|
