Android resampling
Oct 22, 2019 at 2:55 AM Thread Starter Post #1 of 124

djyang0530

100+ Head-Fier
Joined
Feb 23, 2019
Posts
110
Likes
4
Location
.
I found out that whether i use mqa, 32bit 384khz at android mixer they resample does into 24/96? 16/96? So i have two questions.
1. What format do they resample into? Is it 16/96 or 24/96
2. For android os base dap how do they fix those problems? Like $3k dap like ak sp2000 is based on android.
If it's possible pls give me sources
 
Oct 22, 2019 at 7:10 PM Post #4 of 124
That's not the point...... yes resampled 16bit and 24bit 192khz can be easily heard
Well, android usually resamples to 48khz. If there's direct access to the DAC, and if it bypasses Android's audiuo systerm, it will resample to whatever the DAC can actually support.

According to A&K (https://us.astellnkern.com/products/a-ultima-sp2000)
"32bit/768kHz, Native DSD22.4MHz Support
SP2000 is the first Astell&Kern model to support 32bit/768kHz and DSD512 (22.4MHz) playback natively without compression or loss."

BTW - have you done blind testing to support the claim you can hear a difference between 16 and 24 bit 192khz? Because I'm pretty sure no human can.
 
Oct 22, 2019 at 8:58 PM Post #5 of 124
Well, android usually resamples to 48khz. If there's direct access to the DAC, and if it bypasses Android's audiuo systerm, it will resample to whatever the DAC can actually support.

According to A&K (https://us.astellnkern.com/products/a-ultima-sp2000)
"32bit/768kHz, Native DSD22.4MHz Support
SP2000 is the first Astell&Kern model to support 32bit/768kHz and DSD512 (22.4MHz) playback natively without compression or loss."

BTW - have you done blind testing to support the claim you can hear a difference between 16 and 24 bit 192khz? Because I'm pretty sure no human can.
Wt bitrate does it resample into? Yes i did and if you have a decent headphone and decent dac you can easily find out. 24/192 mostly gives more detail upper 8k compared to mp3 320 or even 16/44.1
 
Oct 22, 2019 at 9:03 PM Post #6 of 124
Wt bitrate does it resample into? Yes i did and if you have a decent headphone and decent dac you can easily find out. 24/192 mostly gives more detail upper 8k compared to mp3 320 or even 16/44.1
Well, that's all the info that A&K had, no mention on bits. On android as a whole, it depends on what the DAC supports. If the DAC supports 24-bit, it will keep the 24-bit depth.

So farm, I've found no evidence of any difference over 16/44.1 to anything higher. But hey, whatever pickles your gherkin.
 
Oct 22, 2019 at 9:19 PM Post #8 of 124
If you measure 16bit and 24bit 24bit gives about ~5 better dnr than 16bit file i think this could be an evidence
Not really. Is the difference audible? Instruments and recording have a finite dynamic range.
It would be like taking a colour photo In the infra-red wavelength, sure, the data is there, but you still can't see it, because it's outside your range.
I'm sure someone here who knows mire than me will cone along and clear things out.
 
Oct 22, 2019 at 9:29 PM Post #9 of 124
Not really. Is the difference audible? Instruments and recording have a finite dynamic range.
It would be like taking a colour photo In the infra-red wavelength, sure, the data is there, but you still can't see it, because it's outside your range.
I'm sure someone here who knows mire than me will cone along and clear things out.
I just answered the question about the evidence
Also dnr is Direct proportion with snr and that could be another evidence about 16bit and 24bit and about resample if they don't resample in 2 multiples the file lose part of it. And simply by doing blind abx test it provides to me that there is difference so idk why should i argue with this topic
 
Oct 22, 2019 at 9:44 PM Post #11 of 124
Well, you did come to the science forum...
But that's not the point that i was asking. it doesn't take any money to use bitperfect instead of going through android mixer or windows mixer even tho you can't feel that 5db dnr it's better to use instead of not using...... also at resampling it goes through digital filter for resampling not filters in dac so another filter makes changes on sound so..... it's better to avoid......
 
Oct 22, 2019 at 11:23 PM Post #12 of 124
Well, you did state that the difference was easily audible.
But anyway, I would imagine that A&K would modify Android to play anything natively without resampling, given it's supported by the DAC. Even my Fiio M6 does it up to whatever the DAC supports.
 
Oct 23, 2019 at 12:05 AM Post #13 of 124
I found out that whether i use mqa, 32bit 384khz at android mixer they resample does into 24/96? 16/96? So i have two questions.
1. What format do they resample into? Is it 16/96 or 24/96
2. For android os base dap how do they fix those problems? Like $3k dap like ak sp2000 is based on android.
If it's possible pls give me sources
there are 2 things to consider, what the DAC can do, what protocol is used to send the signal to that DAC. those are not matters strictly limited to android. even if the DAC is able handle a given resolution, the source of the signal has to know it and also has to be set to output that resolution. this is trivial on a DAP to handle internally as they would obviously install their own drivers and routines for the known internal DAC section. and maybe also for when the DAP is connected to an external DAC. I haven't followed the latest versions of android for that, but for years a device was set by default to 44 or 48kHz(you'd find that information somewhere in the device), and audio to a USB DAC would be limited to up to 96kHz. I think Audio Player Pro (android app) can bypass android entirely and output signal above that 96kHz limit(if the DAC can handle more).
 
Oct 23, 2019 at 6:25 AM Post #14 of 124
[1] If you measure 16bit and 24bit 24bit gives about ~5 better dnr than 16bit file i think this could be an evidence
[2] Also dnr is Direct proportion with snr and that could be another evidence about 16bit and 24bit and about resample if they don't resample in 2 multiples the file lose part of it.
[3] And simply by doing blind abx test it provides to me that there is difference, [3b] so idk why should i argue with this topic

Castleofargh (and Killcomic) have already answered your main point, Android will resample to the rate supported by the DAC, commonly in Android phones for example that would be 48kHz, although some are capable of 96kHz.

1. Theoretically, 24bit gives 48dB more dynamic range than 16bit. However, that's not the case in practice, there's no difference in audible dynamic range. This of course assumes that the 16bit and 24bit files are the same master but if they're different masters and additional compression has been applied to the 16bit version, then it would measure roughly 5dB lower dynamic range. However, that is NOT evidence 24bit has more dynamic range than 16bit AND as you're NOT playing different masters in this scenario, just converting sample/bit rates of one master, the audible dynamic range will be the same.

2. No, the only "part of it" lost, is the inaudible parts: Above the frequency limit and below the noise floor of human hearing. Also, not resampling at half the sample rate (multiple of 2) could sometimes make an audible difference with very old resamplers, 25 years or so ago but isn't the case today.

3. Countless ABX and double blind tests, over the course of 2 decades or more, have demonstrated no audible difference, which agrees with the objective measured differences being outside the limits of human hearing. The logical conclusion is that there must have been a fault with the ABX test you conducted, for example, you compared different masters (one with added compression for instance).
3b. You're right, you SHOULDN'T argue with this topic here in Sound Science forum, as you are arguing with well established/demonstrated science and you'd need some exceptional evidence to support your claims or be dismissed as a deluded audiophile making false claims!

G
 
Oct 23, 2019 at 6:06 PM Post #15 of 124
Castleofargh (and Killcomic) have already answered your main point, Android will resample to the rate supported by the DAC, commonly in Android phones for example that would be 48kHz, although some are capable of 96kHz.

1. Theoretically, 24bit gives 48dB more dynamic range than 16bit. However, that's not the case in practice, there's no difference in audible dynamic range. This of course assumes that the 16bit and 24bit files are the same master but if they're different masters and additional compression has been applied to the 16bit version, then it would measure roughly 5dB lower dynamic range. However, that is NOT evidence 24bit has more dynamic range than 16bit AND as you're NOT playing different masters in this scenario, just converting sample/bit rates of one master, the audible dynamic range will be the same.

2. No, the only "part of it" lost, is the inaudible parts: Above the frequency limit and below the noise floor of human hearing. Also, not resampling at half the sample rate (multiple of 2) could sometimes make an audible difference with very old resamplers, 25 years or so ago but isn't the case today.

3. Countless ABX and double blind tests, over the course of 2 decades or more, have demonstrated no audible difference, which agrees with the objective measured differences being outside the limits of human hearing. The logical conclusion is that there must have been a fault with the ABX test you conducted, for example, you compared different masters (one with added compression for instance).
3b. You're right, you SHOULDN'T argue with this topic here in Sound Science forum, as you are arguing with well established/demonstrated science and you'd need some exceptional evidence to support your claims or be dismissed as a deluded audiophile making false claims!

G
I just want to get my question answered..... with the sources like official android patch note or etc...... and for 16bit and 24bit in now a day many of song gets mastered at 24bit/96khz so when you downscale that into 16bit, aac, mp3 it losses many details that you could be also called dnr. With your observation human shouldn't hear any difference with recorded instrument playing and live instrument playing becuz 16bit/44.1khz is already above the maximum of human hearing but it's not like that. Btw 24bit gives more than 1.6m steps but 16bit gives only 65k steps that shows 24bit audio is much more closer to pure analog sounds. No difference and not audible is totally different topic we can find out that there is no difference with many measurements like thd snr dnr sinad etc but only way to prove "not audible" is doing those blind test and that might change through out times. At 90s people believe that maximum human eyes fps was 30 so they made tv that has upto 30fps but at 2019 we know that is false. It might be true that human can't hear difference with pure 16bit and pure 24bit but it's true difference story about getting resampled and going through digital filter (not for dac digital filter that's goes through during resampling) that can effect the sound somehow. You're focusing at the point of "audible" but I'm focusing on the fact that it's different or not. Android mixer is fixed to 48khz that if I'm listening to 44.1 or other songs that's not 2 multiple it's true that it's lossing some tiny part of it. Try this blind test you'll be able to tell the difference that has been resampled lossy and original https://hydrogenaud.io/index.php/topic,6651.0.html
 
Last edited:

Users who are viewing this thread

Back
Top