FiiO M11Pro | Dual AK4497 | THX AAA 78 Headpone Amp | 5.15" 18:9 720P | Exynos 7872 | 3G Ram+64G Rom |
Jun 7, 2020 at 1:55 AM Post #1,246 of 1,931
Interesting. You think 384k/352.8k is wrong?

In UAPP if we set up bit perfect + paid the MQA in app fee, and with the M11 / M15 recent 0604 update UAPP is able to do the complete unfold - not just the 1st but the full unfold - which will show as 352.8k or 384k, it's broken?, or it's working better?

Perhaps that's just a way for FiiO to indicate that the full unfold is happening?

That's why I posted here my experience with the M15 0604 upgrade + UAPP streaming Tidal, I hoped the M11 Pro / M11 could also experience the same MQA improvement. :)

That looks like Tidal - so I went and updated my M15's Tidal, but I'm still only showing 88.2k.

I'm getting 384k in UAPP streaming Tidal witih bitperfect enabled, but not in even the updated Tidal...

Is there another setting in the Android Tidal app I need to set, or set in my Tidal account through my PC's Tidal app?
Tidal company needs to fix app compatible with Android 7.Now is playing 16bit, so I think Fiio gave us this beta firmware to polish it before all will be definitely ready from both sides. So 🙏 everyone, trie to post as many bugs you can so they can sort this out fast and we will be able to use the final release.
Sorry, I had to look it up - MQA is new to me - it turns out that 384k output of a full unfolding MQA renderer isn't even the highest possible, that being 768k:

From: MQA Playback
http://bobtalks.co.uk/blog/science-mqa/mqa-playback/

Q4. What are advantages of an MQA Renderer?
  • An MQA Renderer can provide a low-cost or lightweight upgrade for playback of music encoded in MQA.
  • Some Class I USB devices can only receive 96 kHz audio. In these cases, a Renderer is able to take in the MQA Core and complete the unfolding up to 384 or even 768 kHz internally.
So the M15 / M11 Pro / (M11?) MQA 8x unfolding output from UAPP streaming Tidal MQA's showing "384k" isn't a bug, it's a feature :)

Update: In this case it's unknown to us - so far - as to why FiiO is display 385k / 352.8khz instead of the source sample rate.
 
Last edited:
Jun 7, 2020 at 4:23 AM Post #1,247 of 1,931
Sorry, I had to look it up - MQA is new to me - it turns out that 384k output of a full unfolding MQA renderer isn't even the highest possible, that being 768k:

From: MQA Playback
http://bobtalks.co.uk/blog/science-mqa/mqa-playback/

Q4. What are advantages of an MQA Renderer?
  • An MQA Renderer can provide a low-cost or lightweight upgrade for playback of music encoded in MQA.
  • Some Class I USB devices can only receive 96 kHz audio. In these cases, a Renderer is able to take in the MQA Core and complete the unfolding up to 384 or even 768 kHz internally.
So the M15 / M11 Pro / (M11?) MQA 8x unfolding output from UAPP streaming Tidal MQA's showing "384k" isn't a bug, it's a feature :)
I think you are wrong as most of the MQA is 96k or 192k 24bit. Only few albums are 384k. It is not good result if all MQA playing 352 or 384 meaning is 44k x8 or 48k x8.
MQA could be even 16x.
But each track sould be docoded correctly.
 
Jun 7, 2020 at 4:32 AM Post #1,248 of 1,931
I think you are wrong as most of the MQA is 96k or 192k 24bit. Only few albums are 384k. It is not good result if all MQA playing 352 or 384 meaning is 44k x8 or 48k x8.
MQA could be even 16x.
But each track sould be docoded correctly.
Right, I see what you are saying, and I am sharing what I am reading and seeing with my own eyes - do you see that too? In your Status bar, 384k?

@FiiO - perhaps you can enlighten us as to what the 384k in the Status Bar is supposed to be indicating?

How does the 384k related to the full unfolding of 96k 24 bit files - as indicated by the UUAP streaming Tidal MQA files? Is it simply an indicator to differentiate full unfolding from 1x unfolding? Or is that the end PCM rate output of the unfolding?
 
Last edited:
Jun 7, 2020 at 4:36 AM Post #1,249 of 1,931
It should indicate right unfold as MQA file providing for some 96k for others 192k but not all is 352 or 384k. So I think Fiio need to clarify this with Tadal to fix it correctly.
 
Jun 7, 2020 at 4:40 AM Post #1,250 of 1,931
It should indicate right unfold as MQA file providing for some 96k for others 192k but not all is 352 or 384k. So I think Fiio need to clarify this with Tadal to fix it correctly.
Tidal is showing the correct sample rate in the status bar, it's when we use UUAP to stream Tidal bit perfect that the 384/352.8 shows up in the status bar.
 
Jun 7, 2020 at 2:34 PM Post #1,251 of 1,931
Tidal is showing the correct sample rate in the status bar, it's when we use UUAP to stream Tidal bit perfect that the 384/352.8 shows up in the status bar.

All apps affected, not just the Tidal, the MQA decoding process is:

1; (first) unfold, output stream sample rate is always 88.2 or 96kHz, depending on the container sample rate (44.1 or 48kHz)
2; rendering phase, reconstruction of the original sample rate, which can be 44.1, 48, 88.2, 96kHz.... or whatever which was the original sampling rate of the stream and the DAC is still able to play

The problem is, that it seems the sample rate of the stream is always 384KHz or 352.8kHz after the rendering stage, which is wrong, at least not bit perfect.
 
Last edited:
Jun 7, 2020 at 2:39 PM Post #1,252 of 1,931
All apps affected, not just the Tidal, the MQA decoding process is:

1; (first) unfold, output stream sample rate is always 88.2 or 96kHz, depending on the container sample rate (44.1 or 48kHz)
2; rendering phase, reconstruction of the original sample rate, which can be 44.1, 48, 88.2, 96kHz.... or whatever which was the original sampling rate of the stream

The problem is, that it seems the sample rate of the stream is always 384KHz or 352.8kHz after the rendering stage, which is wrong.
Finally some one
 
Jun 7, 2020 at 9:07 PM Post #1,253 of 1,931
All apps affected, not just the Tidal, the MQA decoding process is:

1; (first) unfold, output stream sample rate is always 88.2 or 96kHz, depending on the container sample rate (44.1 or 48kHz)
2; rendering phase, reconstruction of the original sample rate, which can be 44.1, 48, 88.2, 96kHz.... or whatever which was the original sampling rate of the stream and the DAC is still able to play

The problem is, that it seems the sample rate of the stream is always 384KHz or 352.8kHz after the rendering stage, which is wrong, at least not bit perfect.
Finally some one
Yes, it's all the same thing we have been trying to say, the question is why does FiiO display the 384k / 352.8k instead of the indicated sample rate of the track? My guess is FiiO needed something to indicate the unfolding software was engaged and working - if iFiiO displayed the same as Tidal 1st unfold indicates in the status bar, then how would we know the full unfolding is working?

I would have suggested keeping the same sample rates of the track but changing the font of the status display- or the font color - or added an 8x in the numeric display, or multiplied the actual rate by 8x => 44.1= 328.8, 48=384, 88.2=705.6, 96=768, so where is the 384k / 352.8k coming from? @FiiO

To me the 384k / 352.8 was the PCM output of the render to the DAC, but the numbers don't match the input sample rates, perhaps they indicate something else?
 
Last edited:
Jun 7, 2020 at 10:57 PM Post #1,255 of 1,931
The problem is, that it seems the sample rate of the stream is always 384KHz or 352.8kHz after the rendering stage, which is wrong, at least not bit perfect.

I just tried non-mqa in UAPP with latest firmware, it show 44.1khz in status bar.
If I play MQA files in UAPP with bit perfect mode, it will show 352khz in status bar.
 
Jun 8, 2020 at 4:49 AM Post #1,256 of 1,931
I just tried non-mqa in UAPP with latest firmware, it show 44.1khz in status bar.
If I play MQA files in UAPP with bit perfect mode, it will show 352khz in status bar.

I'm talking about just the MQA playback, player app output (MQA decoder input) must be bit perfect for the proper MQA decoding, but MQA decoder output also should be bit perfect, originating from the MQA stream. Otherwise the rate 352/384kHz is upsampling for the lower sample rates.
 
Jun 8, 2020 at 4:59 AM Post #1,257 of 1,931
Yes, it's all the same thing we have been trying to say, the question is why does FiiO display the 384k / 352.8k instead of the indicated sample rate of the track? My guess is FiiO needed something to indicate the unfolding software was engaged and working - if iFiiO displayed the same as Tidal 1st unfold indicates in the status bar, then how would we know the full unfolding is working?

I would have suggested keeping the same sample rates of the track but changing the font of the status display- or the font color - or added an 8x in the numeric display, or multiplied the actual rate by 8x => 44.1= 328.8, 48=384, 88.2=705.6, 96=768, so where is the 384k / 352.8k coming from? @FiiO

To me the 384k / 352.8 was the PCM output of the render to the DAC, but the numbers don't match the input sample rates, perhaps they indicate something else?


They could, and I think should sign this with an MQA/MQA-S label on the status bar too, or the MQA label could be blue or green to show if the MQA stream is Studio authenticated.

@FiiO: is there any design behind it?
 
Last edited:
Jun 8, 2020 at 4:07 PM Post #1,260 of 1,931

Users who are viewing this thread

Back
Top