Neutron HiFi DAC V1 Review

[CedarX]

Oh dang. Thank you so much for that! Saves me from buying another one!

It's off-topic, so I'll try to limit my comments...

I also have some hearing loss on my right ear. I have tried separate L/R EQ, and although I may discover some "new" soundstage clues, it's not that simple... I think my brain somewhat compensates for that right hearing loss, and if I try to R-EQ it, it sounds weird... So mixed results for me

One complain about the Q5K implementation: the independent L/R is only available in the "SPK EQ" tab. That's the section where you would EQ your HP/IEM to a "target", one that mostly works for you, and compensate for any L/R imbalance of the transducers as needed. It's also where the "AUTO EQ PRESET" are applied if you choose to use that option -- All logical up to that point.
What is not quite logical is to use that "SPK EQ" tab to compensate for hearing L/R imbalance. The "USR EQ" tab is where you would EQ your personal preferences from the "target" you've been using using. So, the process should go like this: EQ the considered HP/IEM to "target" > SPK EQ. Correct measured L/R HP/IEM imbalance > SPK EQ. Adjust personal preference on top of the "target" > USR EQ. Correct L/R hearing imbalance > ??? SPK EQ ??? Why? This should be doable in the USR EQ section, it's not related to the HP/IEM but to the user....

I submitted a change request in the Qudelix forum, but I don't think it got much traction. This is an example where Neutron could make a difference if they wanted to implement a similar, separate USER / HW EQ....... Granted it's probably a minor topic!

 

[Robbo99999]

It's off-topic, so I'll try to limit my comments...

I also have some hearing loss on my right ear. I have tried separate L/R EQ, and although I may discover some "new" soundstage clues, it's not that simple... I think my brain somewhat compensates for that right hearing loss, and if I try to R-EQ it, it sounds weird... So mixed results for me

One complain about the Q5K implementation: the independent L/R is only available in the "SPK EQ" tab. That's the section where you would EQ your HP/IEM to a "target", one that mostly works for you, and compensate for any L/R imbalance of the transducers as needed. It's also where the "AUTO EQ PRESET" are applied if you choose to use that option -- All logical up to that point.
What is not quite logical is to use that "SPK EQ" tab to compensate for hearing L/R imbalance. The "USR EQ" tab is where you would EQ your personal preferences from the "target" you've been using using. So, the process should go like this: EQ the considered HP/IEM to "target" > SPK EQ. Correct measured L/R HP/IEM imbalance > SPK EQ. Adjust personal preference on top of the "target" > USR EQ. Correct L/R hearing imbalance > ??? SPK EQ ??? Why? This should be doable in the USR EQ section, it's not related to the HP/IEM but to the user....

I submitted a change request in the Qudelix forum, but I don't think it got much traction. This is an example where Neutron could make a difference if they wanted to implement a similar, separate USER / HW EQ....... Granted it's probably a minor topic!

In software I use left & right independant EQ to channel match headphones (not my hearing), which I do from measurements. There's some argument that it's best not to correct for hearing loss & hearing imbalance in headpones - so from my own use case per channel EQ is about correcting the headphone to be perfectly channel matched.

 

[dmitrykos]

This is an example where Neutron could make a difference if they wanted to implement a similar, separate USER / HW EQ....... Granted it's probably a minor topic!

Do you mean to have possibility to use EQ (EQ 1) for normal use (user, personal preference) and simultaneously another EQ (EQ 2) for let's say hearing correction? If it is so, then in DAC V1 you can activate 10-band PEQ + FRC. FRC is the same 10-band PEQ, so it can be used for correction needs:

I think my brain somewhat compensates for that right hearing loss, and if I try to R-EQ it, it sounds weird... So mixed results for me

Of course. The same happens with vision, if one eye has some issue with for example damaged part of retina of the eye then brain is trying to compensate that and replaces the artefact with a replica of the part of the image from a healthy zone of another eye. Our brain is the most advanced machine when it comes to compensation works.

Few years ago I had a Neutron Player user who experienced very unpleasant sensations by one ear from a specific range of frequencies. Neutron's per-channel PEQ made it possible for him to use Notch filter for one channel for that specific ear and cancel that range of frequencies making the music more comfortable to listen to. Therefore, it is very personal. Maybe just correction of frequencies, like boosting some, would not make much difference as brain already got used to compensate sound image for that ear.

 

[CedarX]

If it is so, then in DAC V1 you can activate 10-band PEQ + FRC. FRC is the same 10-band PEQ, so it can be used for correction needs:

I understand it can be done (can be done on the Q5K as well).

My point was more about how to structure all these PEQ filters. I like where Qudelix is going: one group of 10x PEQ filters for the HP/IEM used to correct them to a (good) “target” (e.g. using AutoEQ); then another group to adjust this “target” to your personal preferences.

Both EQ groups should have independent L/R adjustment options… but for different reasons: HP imbalance correction for the 1st group, hearing imbalance/preference for the 2nd group.

 

[DanTheMan]

I still can’t believe that no one is talking about the coolest aspect of this device: Ambiophonics! The ability to do Ambiophonics with total control of the relevant parameters for 3D sound from a small or narrow stereo source with multiband parametric EQ is a Godsend! To me, V2 doesn’t need 4V as much as it needs to be Bluetooth and possibly LDAC capable. Yeah, you can do that from the app now and it’s only $8, but not with your favorite streaming service. You’re limited to internet radio or your own hard drive. I think everyone owes it to themselves to at least try the app and stream it to a Bluetooth capable stereo. You might just be surprised how big a small stereo Bluetooth speaker can sound with Ambiophonics let alone how awesome a real stereo can sound. The old Jawbone Jam Boxes used to have BACCH filters which was fun, but Ambiophonics has always kept a better FR balance to my ear. With this device, you really should be able to get the most out of your setup. You have control over all the parameters. I mean we are 12 pages deep people…

 

[d0j1n]

@dmitrykos My Neutron player just arrived. First of all congratulations on this beauty of a device. Configuring is also really nice and easy. It is so small and lightweight!! Just perfect for portable use. One thing made me wonder. I found presets from oratory in the frc-section and they look quite different from what I get from oratorys dropbox pdf-file. Could you or somebody else explain to me where the frc-values come from and how they differ from what I can get from oratory himself on his reddit full-list? Thanks in advance!

Greetings from Germany!

 

[dmitrykos]

Could you or somebody else explain to me where the frc-values come from and how they differ from what I can get from oratory himself on his reddit full-list?

I am really glad that you enjoying your DAC V1!

FRC presets are coming from the AutoEQ database maintained here:

GitHub - jaakkopasanen/AutoEq: Automatic headphone equalization from frequency responses

Automatic headphone equalization from frequency responses - jaakkopasanen/AutoEq

github.com

I did not look deep into how AutoEQ collects measurements but you could get in touch with the maintainer (jaakkopasanen) via the Issues section of that repo and report the issue, like the one you noticed.

NConfigurator keeps parsed AutoEQ presets in fc_presets_autoeq.xml file which is stored on Windows for example here: C:\Program Files (x86)\NConfigurator\bin\assets

Currently you can save/load one preset. But you you could also make your own copy of fc_presets_autoeq.xml file and edit it with a text editor by leaving presets which are needed for you or just add only your own.

 

[CedarX]

One thing made me wonder. I found presets from oratory in the frc-section and they look quite different from what I get from oratorys dropbox pdf-file.

I don’t have the DAC V1, so I’m not sure what the “frc-section” contains exactly, but if it is similar to other devices relying on the AutoEq database (e.g. Qudelix…), that would be Oratory’s measurements equalized to the AutoEq default target—an automated process based on AutoEq algorithms… Whereas Oratory’s pdf files are Oratory’s “curated” EQ—a manual or semi-manual process representing Oratory’s take on each HP/IEM.

The same kind of differences you’ll find comparing Amir EQ recommendations, @Maiky76’ ”take” on EQ, and AutoEq results if you upload Amir’s measurements into AutoEq and select one of AutoEQ targets…

 

[dmitrykos]

There are new updates to firmware and NConfigurator with new functionality and bug fixes.

Firmware:
Firmware 47 - Neutron Forum

NConfigurator:
NConfigurator 1.6.5 - Neutron Forum

With new update:

• It is now possible to force oversampling for platforms which do not support frequency switching by allowing only 384000 Hz for example (of course quality will depend on the resampling software of the app/OS). Earlier, DAC V1 could be limited to some max frequency, so with this modification there are more possibilities to play with this functionality.
• DAC V1's functionality to go into a deep sleep when nothing is connected to line-out became optional, see option Deep Sleep. There were many cases of confusion from DAC V1 users regarding this behavior, especially if one attempts to configure DAC V1 without headset attached (looked like DAC V1 was malfunctioning).
• Mic's output frequencies can be limited too.

In addition, many thanks to @d0j1n for noticing that NConfigurator's EQ GUI could not change filter type and parameter type (bug introduced in recent updates of NConfigurator). This issue forced the release of this update.