[Mmurin]

I tried to reply your test with RMAA, as explained here

https://www.head-fi.org/threads/the...urements-in-post-3-tutorial-in-post-2.800208/

with the mobile loaded with sennheiser ie80 (small load)

Guess what, same result as you with all players, i.e.roll-off beyond 18kHz. I would say that this falls beyond audible frequencies.
One notable exception: aimp (audiotrack method) rolls off way before 18 kHz, and shows better stereo crosstalk values, compared to the other players.
I guess what we are seeing here is some software tuning.

I also tried a test for the V20 with higher demanding heaphones and a 75 ohm impedance adapter in the measurement loop.
Same frequency response as with stock player.
There where signs of amplifier suffering (high distorsion), but with better crosstalk values.
I guess this is a hint the V20 is optimized for high impedance, high efficiency headphones.

Interesting thing 1: the only value seriously affected when switching the dac off was stereo crosstalk, which probably means that really the V20 makes a difference for high impedance loads.
Interesting thing 2: I compared the results with my old lumia 820, using foobar for windows phone. The test shows a marked roll-off similar to aimp for the V20, but if the wasapi option is chosen the rolloff frequency increase, as in the other players for the V20.

Conclusions:
1- the DAC - software interface problems are more common than we think
2- still not sure this HF rolloff at 18 +kHz is audible
3- the roll-off for Win phone foobar without wasapi is clearly audible
4- I definitely would like some feedback from the more expert members of this community
5- probably the V20 is best performing with high impedance, high efficiency headphones.

I will reply to myself because I have some serious updates.
I repeated the tests with RMAA test audio file encoded first in opus.ogg (160 kbps) and then in .mp3 (320 kbps), for both my LG V20 and the old Lumia 820.
The overall readings degrades, especially the ones connected to the stereo tests.

The frequency response instead becomes completely flat!!!

For the V20 this happens with all music player except AIMP, which when playing the mp3 still gently roll-off the highs gradually, starting from 10 kHz.
One exception: Rocket player reading are overall worse, which leads me to believe it just does not use the DAC.

It is also interesting that thanks to the degradation coming from the compressed files, the measurements of the Lumia are on par. with the ones of the V20.
What does it means? Is it only RMAA black magic, or really any mobile is just good enough to play compressed audio file on easy to drive IEMS like the IE80?

I will post soon my measurement hoping for somebody to shed some light on this mystery.

 

[tienbasse]

Thank you for the information, but I kinda fail to see what point you're trying to make here...
Most people here enjoy the V20 immensely, roll-off or not, which is usually a sign of good performance (that and the 226 pages of this thread).

People, as usually, fail to see that a mass-produced item induces some variability, despite the QC testing that is certainly being done.
Some people will get units with lesser performance, some other with better performance, because V20 is not a hand-made 3000$ DAP. So don't spend too much time testing your V20 with the hope of reaching universal conclusions, it is relatively pointless and it certainly won't convince people who are satisfied with their V20 (as I am).

To add to it, the highest part of treble is certainly the most difficult part of the spectrum to maintain versus load, and there are a number of dedicated DAP that are not really doing a better job than V20. Plus, treble perception is also VERY person-dependant (sibilence, hiss, brightness due to various treble frequencies), so measurement is not a substitute for critical listening, and the IEM will affect the result immensely.

 

[fjhuerta]

I will reply to myself because I have some serious updates.
I repeated the tests with RMAA test audio file encoded first in opus.ogg (160 kbps) and then in .mp3 (320 kbps), for both my LG V20 and the old Lumia 820.

First thing - it's vital to do this test using a FLAC, WAV or uncompressed audio. Using a compressed file means you probably don't have any high frequency content to begin with. Try your test with a FLAC. It's quite possible your results will be different. I bet 160kpbs audio cuts off at around 17 KHz, and the 320KBPs won't have any data above 20-22 KHz.

What does it means? Is it only RMAA black magic, or really any mobile is just good enough to play compressed audio file on easy to drive IEMS like the IE80?

The differences on the LG V10 were very, very subtle, but they were there. They were slightly apparent when you did a quick A/B comparison. When compared directly on RMAA, you could easily see them. Try making the comparison with a FLAC, changes are more obvious.

People, as usually, fail to see that a mass-produced item induces some variability, despite the QC testing that is certainly being done.
Some people will get units with lesser performance, some other with better performance, because V20 is not a hand-made 3000$ DAP.

WHAT??!? You do know variability is quite a bit higher on hand built equipment, don't you? There's NO way people will get lesser perfomance with a mass produced product like the LG V20. Not one. Every chip, motherboard, circuit board trace is identical. Every single LG V20 will perform the same. It's as if thinking that, since the LG V20 is not hand made, its performance varies. It doesn't!

So don't spend too much time testing your V20 with the hope of reaching universal conclusions, it is relatively pointless and it certainly won't convince people who are satisfied with their V20 (as I am).

If you are happy with your V20, more power to you. I don't think you'll be convinced your unit performs poorly, when you like it. That's fine, I think.

To add to it, the highest part of treble is certainly the most difficult part of the spectrum to maintain versus load, and there are a number of dedicated DAP that are not really doing a better job than V20.

20 KHz is a walk in the park for practically all analogue circuits. A 384KHz DAC has theoretically 172 KHz more bandwidth. No amp I have ever tested has any issues getting to at least 30 KHz, not even old tube amps. Unless you really screwed up in your amp design, there's no way a headphone will present such a load as to generate a roll off such as the one we are looking at. This is not an "analogue filter", 6dB/Oct roll off. It's more similar to a brickwall digital filter. It doesn't have to do with a problem in the analog (amplifier) section of the V20, which is the only one that would be load dependent. It seems the audio subsystem is resampling audio. That's NOT good. Something is weird with at least two LG V20s.

If you are fine with not having a high resolution device, when the box states it is, that's fine.. but I bet some of us are curious as to why this device is not living up to its claims.

 

[Mmurin]

I will reply to myself because I have some serious updates.
I repeated the tests with RMAA test audio file encoded first in opus.ogg (160 kbps) and then in .mp3 (320 kbps), for both my LG V20 and the old Lumia 820.
The overall readings degrades, especially the ones connected to the stereo tests.

The frequency response instead becomes completely flat!!!

For the V20 this happens with all music player except AIMP, which when playing the mp3 still gently roll-off the highs gradually, starting from 10 kHz.
One exception: Rocket player reading are overall worse, which leads me to believe it just does not use the DAC.

It is also interesting that thanks to the degradation coming from the compressed files, the measurements of the Lumia are on par. with the ones of the V20.
What does it means? Is it only RMAA black magic, or really any mobile is just good enough to play compressed audio file on easy to drive IEMS like the IE80?

I will post soon my measurement hoping for somebody to shed some light on this mystery.

I will reply to myself because I have some serious updates.
I repeated the tests with RMAA test audio file encoded first in opus.ogg (160 kbps) and then in .mp3 (320 kbps), for both my LG V20 and the old Lumia 820.
The overall readings degrades, especially the ones connected to the stereo tests.

The frequency response instead becomes completely flat!!!

For the V20 this happens with all music player except AIMP, which when playing the mp3 still gently roll-off the highs gradually, starting from 10 kHz.
One exception: Rocket player reading are overall worse, which leads me to believe it just does not use the DAC.

It is also interesting that thanks to the degradation coming from the compressed files, the measurements of the Lumia are on par. with the ones of the V20.
What does it means? Is it only RMAA black magic, or really any mobile is just good enough to play compressed audio file on easy to drive IEMS like the IE80?

I will post soon my measurement hoping for somebody to shed some light on this mystery.

I will reply to myself because I unraveled the mystery.
I tried to produce an MP3 at 48kHz, and the frequency response with AIMP was completely flat up to 20 kHz.
Flac with the stock player also is flat up to 20kHz.
The only file which rolls off (from 18 kHz onward) on the stock player is the wav which comes straight out of RMAA.

So the conclusion is that this mess comes from the infamous resampler.
So basically I infer that with stockplayer and the other players audio files (except wav) are re-sampled to 48 kHz no matter what.
AIMP plays the native audio file without resampling, no matter what.

 

[tienbasse]

WHAT??!? You do know variability is quite a bit higher on hand built equipment, don't you? There's NO way people will get lesser perfomance with a mass produced product like the LG V20. Not one. Every chip, motherboard, circuit board trace is identical. Every single LG V20 will perform the same. It's as if thinking that, since the LG V20 is not hand made, its performance varies. It doesn't!

I happen to design medical devices for a living, and my team also transfers completed designs onto manufacturing lines. We work in the plant next to the lines, so let me explain why i say that.
The issue with mass-manufacturing is not built reproducibility, in which a proper manufacturing process will excel (assuming it is properly developped and followed over time, which I can tell you is not very often the case).
The main issues are quality control and process deviation over time.

When you mass-manufacture items, you want minimize costs, which also means you want to minimize lead time (= time from scratch to release). Unless you build luxury items (= not a smartphone, even high-end), quality control will be "optimized", meaning that systematic testing (= on 100% units) will be reduced to basic functional testing that can be automated. There may be more thorough additional testing, but since it is more time consuming and/or not automated, it will be done on a statistical basis (one in every XXX units).
End result: a small percentage of disfunctional units will escape QC and reach customers, obviously with issues that are small enough not to be discovered through automated QC and that arise from the slow small deviations that invariably affect processes over time.
These deviations should in theory be detected through proper in-process controls, but these tend to slow down processes, so they're reduced to basics, so yes, small s**t can still happen over time (often small enough not to affect main functions but big enough to degrade performance in unsuspected ways)!
And trust me s**t happens a lot for items with small margins and large scale production.

Hand-made high-end items may present more variations in terms of build, you are right, but each unit usually gets tested through and through since cost is less of an issue. Unless there was an actual design flaw, you will get a functional unit, and in the rare event of an issue, you will get aftersale service more easily than with a mass-produced item (especially a smartphone, where most manufacturers tend to play hide-and-seek).

Sorry for the wall of text, but people put way too much trust into industrial manufacturing. Especially for items that are not made to save lives or do not present significant risks for users (more risk = more norms to impose stricter release testing and rigorous design).

 

[sherazshery]

I also use HD600 with my korean LG G6. I have got a shorter cable for outdoor usage and love the comibnation.

i am also a proud owner of the LG G6 Plus. i am interested in buying a good open head phone such as Senn HD600 to up my game from superlux hd 668 B.

the hd 668 pairs good with my htc m7 but on G6 it is very harsh with no bass even after applying 3 layers of the famous tissue paper mode.

my fiio F5 iem works great with LG G6 and i like its combination very much ( even better in high gain mode)



whats your take on the G6 HD600 combo. will it be any good for me as i like fairly neutral and airy sound (neither a basshead nor a treblehead)

many thanks in advance for everything.

 

[Mmurin]

I did the tests for my own intellectual curiosity, and also because my subjective feeling, now proven right, was that not all music players sounded the same.
Also, this kind of roll off on certain types of music is plenty audible, e.g. metal music.

All in all I think the music experience in Android is quite a s**tshow. I did the same tests on the HTC10 (before it died), and on that one the differences were clearly audible.

And, aside from all this, there are not that many music players with a really well designed UI. Poweramp, Neutron, UAPP and most of the rest lack what I consider a basic feature: multiples queues (or auto playlist, call them what you prefer), with automatic bookmarking.

One shouldn't have to go through such an ordeal just to listen to music.

 

[fjhuerta]

Sorry for the wall of text, but people put way too much trust into industrial manufacturing. Especially for items that are not made to save lives or do not present significant risks for users (more risk = more norms to impose stricter release testing and rigorous design).

Ah, I see where you are coming from, thanks for the explanation.

Regardless, a variation from 96KHz to 20 KHz is not within the realm of tolerance. It's something else.

What I'm worried is this. Android is known for resampling audio, and not in the best of ways. It seems to me, and to Mmurin, hi res audio (and even standard CD audio) is resampled to a 48KHz fs. This is bad, it'll seriously affect quality.

The weird thing is, my V10, S8+ and X5 don't do it. The V20 is the only one that does it.

And apparently, you need to use a very specific audio player to avoid it (if at all).

I just got USB Audio Player Pro with the experimental code to directly access the V20's DAC. As soon as I get my new V20, I'm going to do a tst and compare it to my wife's. I'll also try to downgrade its firmware, and see if something along the way is making the audio subsystem kick in.

 

[vladstef]

Ah, I see where you are coming from, thanks for the explanation.

Regardless, a variation from 96KHz to 20 KHz is not within the realm of tolerance. It's something else.

What I'm worried is this. Android is known for resampling audio, and not in the best of ways. It seems to me, and to Mmurin, hi res audio (and even standard CD audio) is resampled to a 48KHz fs. This is bad, it'll seriously affect quality.

The weird thing is, my V10, S8+ and X5 don't do it. The V20 is the only one that does it.

And apparently, you need to use a very specific audio player to avoid it (if at all).

I just got USB Audio Player Pro with the experimental code to directly access the V20's DAC. As soon as I get my new V20, I'm going to do a tst and compare it to my wife's. I'll also try to downgrade its firmware, and see if something along the way is making the audio subsystem kick in.

I've had great success with latest builds of Neutron player - go to their side and in their forums you have a list of settings that you need to select for their drivers to kick in - they bypass Android's nasty ways by upsampling to 24 bits instead of 44.1 -> 48 (this is far less audible, if at all). You also get bit perfect anything above 48kHz.

 

[Jwtencza]

Hi, I have a pretty basic question. I've researched online but am still not sure. Does the LG V2 equalizer work system wide or just with certain apps? In other words, if I am streaming Amazon prime music can I adjust the sound frequencies with the built in equalizer. I would be on a Verizon phone. I had the LG V10 and I remember that I had to use apps from the play store, and I was not very impressed with them. Thanks so much.

 

[cirodts]

A simple question, which player to use for maximum sound quality for v20?

 

[oliwek2]

There are different views on this matter, but usually people recommend UAPP (USB Audio Player Pro : https://play.google.com/store/apps/details?id=com.extreamsd.usbaudioplayerpro) or the stock LG player.

 

[cirodts]

I have jetaudio, I do not like the lg player because it does not select folders and sub folders.
Perhaps with the standard player lg you have better sound quality and with other players no?

 

[oliwek2]

I just got USB Audio Player Pro with the experimental code to directly access the V20's DAC.

Thanks for this sentence, I didn't know there was beta code for direct ESS DAC access from UAPP (with a custom HiRes driver). For people like me, read http://www.extreamsd.com/forum/showthread.php?tid=621
(people using it are not allowed to share the code themselves, just mail them)

I have jetaudio, I do not like the lg player because it does not select folders and sub folders.
Perhaps with the standard player lg you have better sound quality and with other players no?

With high resolution files (and good CIEMs) I had better sound than using for example Poweramp, yes. But some people prefer Poweramp from what I read here, so it's better that you test yourself with your hardware. The stock LG player interface is quite limited, that's also true. Never tested Jetaudio, but there are so many players on android now.

 

[Zildon]

Thanks for this sentence, I didn't know there was beta code for direct ESS DAC access from UAPP (with a custom HiRes driver). For people like me, read http://www.extreamsd.com/forum/showthread.php?tid=621

Hi, been using this UAPP direct driver for a couple of weeks and have really enjoyed listening to it. Although only using my ears it equals the sound quality of the LG stock player with fewer limitations such as reading cue files. I have also not experienced any interruptions/ interference when playing music in UAPP from notifications, app updates like i get from the stock player. You have to pay for it and then email the Davy the UAPP developer for the direct download to the beta programme but it seems the only workaround until LG V20 gets the Oreo update where android wont downsample.