Had an extremely brief chat with an audiologist who specificialises in auditory processing as she was trying to get out the door to work this morning. The view was that in theory one could introduce changes to timing, intensity etc in software that would introduce spatial cues, and then introduce those to the ear canal through different drivers. However, what would happen perceptually after that would be profoundly impacted by the behaviour of the individual cochlear and neurology, all of which is habituated to individual pinna effect. In other words you bypass the pinna and all bets are off. Presumably this is why iems are such and enigma in the first place in terms of user preference. That obviously doesn't immediately rule out the possibility that one could develop a more sophisticated approach to spatial cues in software, but it's not obvious to me how splitting out the channels within the iem itself adds any value.
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Qudelix finally released their (AME?) T71 multi channel IEM....Do we care?
- Thread starter tjf
- Start date
somebodyelse
Major Contributor
- Joined
- Dec 5, 2018
- Messages
- 4,669
- Likes
- 4,113
This is why the good spatial systems are based on individual measurement, either with in-ear mics or modelled from scans of the head/torso and pinna. The step down from that is an average model for the population. Rather like the Harman targets, it may work well if you are close to the average, or not if you are an outlier.
- Joined
- Mar 1, 2020
- Messages
- 107
- Likes
- 499
You can absolutely do this - and it works decently well with over-ear headphones (link including a whitepaper demonstrating the effectiveness of this method).
But with in-ear headphones, different speakers do not have any spatial cues - it doesn't matter where inside the IEM the speaker is located, all IEMs play the sound directly into the ear canal, and the ear canal itself does not have any directional effects. Meaning that any sound source that is placed behind the ear canal entrance (EEP) does not carry spatial cues anymore.
Which in turn means that instead of acoustically summing the signals of each speaker you can just as easily sum the signals of each channel electrically and play them over a single speaker, with the end result being the same (assuming the speaker has negligible distortion)
I have received the T71 IEM. I had some doubts about Qudelix's "Surround" claims about this IEM... As it is today, the T71 dongle + T71 IEM combo appears to be a very interesting 4-way active IEM... but with no particular surround sound benefit.
In their forum, Qudelix reported the cutout frequencies as 150, 1500, and 6500 Hz. They hard-coded, and cannot be adjusted in the software (App). I suppose it is to protect the drivers, especially if you were to feed the BAs with low-frequencies.
The App gives you a set of "SPK PEQ" for each one of the 4x drivers, for both L&R or individual L and R IEMs. Each SPK PEQ is a set of up to 10x filters to choose from LPF, HPF, PEAK, LSHELF, and HSHELF. You can also adjust the HEADROOM and PRE GAIN, and MUTE the individual driver, both L&R or individual L and R. In that particular case, the pre-gain is equivalent to a gain adjustment for each one of the 4 active amps.
In recent Qudelix SW versions, both 5K and T71, "SPK EQ" is intended for "Speaker" FR correction--well the IEM in that case. You adjust the frequency response of a given set of IEMs to whatever "target" you want. It's also where the "AUTO EQ" preset are applied if you choose that option and import settings from AutoEQ. But you also have a "USER EQ" section, with another set of 10x PEQ (or GEQ for non-T71 IEMs). The idea is that every IEM would be adjusted to the same target in the "SPK EQ" section, and the USER EQ section would be available to adjust your own preferences based upon, for example, the type of music you are listening to... Independent L&R PEQ is not available for "USER EQ" (it should IMO, as my left and right ears are different...).
Going back to the T71 IEM, if my maths is correct, you have 9 sets of 10x PEQ filters (90 filters !!!) to play with: one set of ten filters for each one of the four drivers; which can also be L/R independent; plus one set of ten filters for USR EQ.
Example (Chrome extension):
This what I did (Filter type / Freq / Gain / Q) in this example:
> Low-shelf / 125 / +6.0 / 1.414 for both DD drivers
> Peak / 1000 / -3.0 / 1.414 for both BA1 drivers
> Peak / 4000 / 6.0 / 2.000 for the Left BA2 driver
> Peak / 4000 / 4.0 / 3.000 for the Right BA2 driver
> Peak / 8000 / -6.0 / 2.000 for both BA3 drivers
I won't comment much on the IEM sound itself... First that would be subjective only, and to me at least, with all these PEQ possibilities, it is largely irrelevant !!! As long as each driver distortion is acceptable (the fact that they use reputable Sonion BAs is a good sign, but not a guarantee...) the T71 can be adjusted to sound however one want...
The difficulty is to use all that stuff correctly, and without actual measurement capabilities, adjusting the T71 IEM to your own preference is going to be a daunting task !!! Just to make it more daunting, I also wish the App offered 'some' cutout freq. adjustment options, even if they were bound by some hard-coded limits...
But... I see nothing here about surround... I mean, it is possible that, in a future version, Qudelix develop a downmix SW where each of the four driver would be fed by some combination of surround channels, but I still don't understand how that would work: each driver appears to be dedicated to one freq. range, it's not "four wideband drivers in parallel". Even if it was the case, since each driver outputs in the same nozzle in your ear canal, I don't think it'll make any difference for the ear drum...
The T71 Amp/DAC has enough processing power to implement some fairly elaborate downmixing algorithms, no question here. But it would work the same whether you have a regular IEM (with passive filters if multi-drivers) or the T71 4-way active IEM.
EDIT: @oratory1990 already covered this in the previous post !!!
A potential benefit is IMD and phase distortion. This was the main Qudelix argument when they introduced the Qx-Over 2-way active IEM (2x identical 8mm drivers if I recall correctly, not sure if they were the same as the the T71 DD). Traditional Hybrid IEMs are build around a DD driver with a "down-slopping" FR, and BAs are used to "fill" and adjust the FR. With arrangement, the DD driver can exhibit IMD distortions in the undesirable/audible freq. range. By having steep active filters, the IMD concern is eliminated. Qudelix also claimed that passive filters could introduce phase distortions, an issue not present with active filters. This is well beyond my expertise, and I don't know enough to separate the marketing BS from the truth here... but the 4-way active T71 IEM may bring some benefit in these areas, IMD and phase distortion.
My take of the T71 IEM so far: super interesting from a technical side--a tinkerer dream !!!
But no surround credential IMHO...
In their forum, Qudelix reported the cutout frequencies as 150, 1500, and 6500 Hz. They hard-coded, and cannot be adjusted in the software (App). I suppose it is to protect the drivers, especially if you were to feed the BAs with low-frequencies.
The App gives you a set of "SPK PEQ" for each one of the 4x drivers, for both L&R or individual L and R IEMs. Each SPK PEQ is a set of up to 10x filters to choose from LPF, HPF, PEAK, LSHELF, and HSHELF. You can also adjust the HEADROOM and PRE GAIN, and MUTE the individual driver, both L&R or individual L and R. In that particular case, the pre-gain is equivalent to a gain adjustment for each one of the 4 active amps.
In recent Qudelix SW versions, both 5K and T71, "SPK EQ" is intended for "Speaker" FR correction--well the IEM in that case. You adjust the frequency response of a given set of IEMs to whatever "target" you want. It's also where the "AUTO EQ" preset are applied if you choose that option and import settings from AutoEQ. But you also have a "USER EQ" section, with another set of 10x PEQ (or GEQ for non-T71 IEMs). The idea is that every IEM would be adjusted to the same target in the "SPK EQ" section, and the USER EQ section would be available to adjust your own preferences based upon, for example, the type of music you are listening to... Independent L&R PEQ is not available for "USER EQ" (it should IMO, as my left and right ears are different...).
Going back to the T71 IEM, if my maths is correct, you have 9 sets of 10x PEQ filters (90 filters !!!) to play with: one set of ten filters for each one of the four drivers; which can also be L/R independent; plus one set of ten filters for USR EQ.
Example (Chrome extension):
This what I did (Filter type / Freq / Gain / Q) in this example:
> Low-shelf / 125 / +6.0 / 1.414 for both DD drivers
> Peak / 1000 / -3.0 / 1.414 for both BA1 drivers
> Peak / 4000 / 6.0 / 2.000 for the Left BA2 driver
> Peak / 4000 / 4.0 / 3.000 for the Right BA2 driver
> Peak / 8000 / -6.0 / 2.000 for both BA3 drivers
I won't comment much on the IEM sound itself... First that would be subjective only, and to me at least, with all these PEQ possibilities, it is largely irrelevant !!! As long as each driver distortion is acceptable (the fact that they use reputable Sonion BAs is a good sign, but not a guarantee...) the T71 can be adjusted to sound however one want...
The difficulty is to use all that stuff correctly, and without actual measurement capabilities, adjusting the T71 IEM to your own preference is going to be a daunting task !!! Just to make it more daunting, I also wish the App offered 'some' cutout freq. adjustment options, even if they were bound by some hard-coded limits...
But... I see nothing here about surround... I mean, it is possible that, in a future version, Qudelix develop a downmix SW where each of the four driver would be fed by some combination of surround channels, but I still don't understand how that would work: each driver appears to be dedicated to one freq. range, it's not "four wideband drivers in parallel". Even if it was the case, since each driver outputs in the same nozzle in your ear canal, I don't think it'll make any difference for the ear drum...
The T71 Amp/DAC has enough processing power to implement some fairly elaborate downmixing algorithms, no question here. But it would work the same whether you have a regular IEM (with passive filters if multi-drivers) or the T71 4-way active IEM.
EDIT: @oratory1990 already covered this in the previous post !!!
A potential benefit is IMD and phase distortion. This was the main Qudelix argument when they introduced the Qx-Over 2-way active IEM (2x identical 8mm drivers if I recall correctly, not sure if they were the same as the the T71 DD). Traditional Hybrid IEMs are build around a DD driver with a "down-slopping" FR, and BAs are used to "fill" and adjust the FR. With arrangement, the DD driver can exhibit IMD distortions in the undesirable/audible freq. range. By having steep active filters, the IMD concern is eliminated. Qudelix also claimed that passive filters could introduce phase distortions, an issue not present with active filters. This is well beyond my expertise, and I don't know enough to separate the marketing BS from the truth here... but the 4-way active T71 IEM may bring some benefit in these areas, IMD and phase distortion.
My take of the T71 IEM so far: super interesting from a technical side--a tinkerer dream !!!
But no surround credential IMHO...
Last edited:
Similar threads
