I have a oscilloscope. How can I use it for my problem?I remember that the oscilloscope can only measure the frequency, can't measure the bit depth?
To measure actual used bit depth, it depends if it is a digital scope with long memory. Some expensive scopes even have I2S decoders, but they may be impractical for most hobbyists.
However, bit depth can be implied from BCLK frequency. Doesn't mean all the bits are really used, some may be zeros and never change.
The way it works is like this: If you have two channels and 16-bits at 44.1kHz, the calculation would be: 2-channels * 16-bits * 44.1kHz = 1.4112MHz = BCLK
Last edited:
I have a music file is 32bits 192kHz. When the 16bits into 32bits function of FifoPi is turned off, the output frequency of the BCLK of the FifoPi is 6.1MHz. In other words, when the 16bits into 32bits function is turned off, the bit depth is reduced to 16bits, even if the music itself is 32bits. Interestingly, when I turn resmapling of volumio on, whether it is 24bit or 32bits, the output frequency of BCLK is 12.2MHz, that is, the bit depth becomes 32bits. However, only the choice of resmapling into 24bit, the sound actually heard is no noise. In other words, resmapling into 24bits on Volumio, the bit depth is actually still 32bits and will not become 24bits.
Then, when the 16bits into 32bits function is enabled, the bit depth is 32bits and will not be reduced to 16bits. Even if the music file itself is 16bits, the output bit depth of FifoPi is 32bits. In other words, my FifoPi's 16bits into 32bits feature is normal.
However, Spotify without mpd's resmapling will not be able to play normally.
The conclusion is that it is inexplicable...
Then, when the 16bits into 32bits function is enabled, the bit depth is 32bits and will not be reduced to 16bits. Even if the music file itself is 16bits, the output bit depth of FifoPi is 32bits. In other words, my FifoPi's 16bits into 32bits feature is normal.
However, Spotify without mpd's resmapling will not be able to play normally.
The conclusion is that it is inexplicable...
When playing music with Moode Audio, I must also resampling into 24bits to play music without noise. So this problem should not be about the software. The most problematic would be buffalo IIIse?
Depends on service. Premium is 256 k AAC or 320 k vorbis depending on platform. That of course has no effect on 16 vs 24 vs 32 bits.
Hi ofswithed,
I would suggest you running your BIII se directly from RPi, without anything else. If it still works with 24bit only, and problem with 32bit, then you have to ask Twisted pear for support.
Or, if it you want, you can send your BIII se to me. I will find out the reason of the problem for you. Though I'm not responsible for debugging user's system, but I would be glad if I can help.
Regards,
Ian
It appears that all Spotify music file formats are lossy compressed. At some point a lossy compressed file will have to be decoded into some fixed format to send to the dac, which may be where ofswitched is having a issue.
It’s really grateful to hear that. In fact, I bought a total of two BUFFALO-IIISE PRO, one of which was completely unusable, I guess the firmware is damaged. At that time, I chose the most convenient way that buy another BUFFALO-IIISE without debug. However, I don’t want to spend any more money to buy a BIII se again, so I’m really happy to hear that you are willing to help me.
If you can, both will check it for me? Many thanks! What is your address?
When FifoPi is connected directly to BIII se, it is not surprising that I still need to set resampling into 24bit in Volumio in order to play normally.
The problem I am facing now may be that the DAC can only accept certain special information.
I have specifically studied Spotify, and I have tried to use the resampling feature that is specific to Spotify, but I can't complete the installation smoothly. The resampling feature maybe just a semi-finished product.
Spotify can easy to listen to current popular songs, then I will pay for physical CDs and do CD ripping myself.
CD ripping doesn't really achieve the best quality by EAC alone, because it is very likely that it will be read incorrectly every time ａｎｄ EAC mistakenly believes that this is correct. In the whole process, the reading performance of CD drive has a great influence, and the power quality of CD drive is important. According to my own actual experience, the sound quality can be improved by about 30%.
The CD drive is recommended to use at least Pioneer BDR-S09XLT, the top is Pioneer BDR-S12J-X, and use SATA to USB 3.0 Adapter to provide a high quality power supply for the CD drive.
Last edited:
Demagnetizer is also helpful, but many people don't want to believe it. I bought it and tried it directly. I found that it really helped. If you think that the Japanese brand Demagnetizer is expensive, you can buy it from Taoboa that the performance is better.
The HifiBerry DAC+ only supports up to 24bits, I wonder if something in the driver or Volumio is causing a problem because of this. When you play the 32bit 192kHz file, what does the alsa hw_params and sw_params show?
I may be mistaken, but I think EAC uses a library of check sums to ensure it has bit perfect rip, and keeps working until it gets the correct check sum.
When playing music using Moode Audio with IanFIFO I2S audio device, I Similarly must also resampling into 24bits to play music without noise. So this problem should not be about the HifiBerry DAC+.
When I enable the resampling:
the logs:
http://logs.volumio.org/volumio/PliiVkd.html
When I disable the resampling:
the logs
http://logs.volumio.org/volumio/gJlbsqg.html
Code:
access: RW_INTERLEAVED
format: S24_LE
subformat: STD
channels: 2
rate: 44100 (44100/1)
period_size: 4410
buffer_size: 22050
tstamp_mode: NONE
period_step: 1
avail_min: 4410
start_threshold: 17640
stop_threshold: 22050
silence_threshold: 0
silence_size: 0
boundary: 1445068800http://logs.volumio.org/volumio/PliiVkd.html
When I disable the resampling:
Code:
access: RW_INTERLEAVED
format: S16_LE
subformat: STD
channels: 2
rate: 44100 (44100/1)
period_size: 4410
buffer_size: 22050
tstamp_mode: NONE
period_step: 1
avail_min: 4410
start_threshold: 17640
stop_threshold: 22050
silence_threshold: 0
silence_size: 0
boundary: 1445068800http://logs.volumio.org/volumio/gJlbsqg.html
I am not very familiar with the theory, but I actually use Pioneer BDR-S12J-X and SATA to USB 3.0 Adapter with independent input power for CD ripping to get more detailed music files. If you are interested, I can upload the files to you to listen it.
I guess that even though EAC uses accuraterip for bit perfect rip, it is possible that the information of accuraterip is provided by 99% of ordinary CD drive, and EAC mistakenly believes that the information provided by 1% high-level CD drive is wrong.
The CD drive offsets list in Accuraterip are only the entry-level BDR-S09, but there is no expensive BDR-S12J-X. I am alone in the real bit perfect rip. I can't contribute information from accuraterip because I use BDR-S12J-X. I am particularly slow at CD rips because there is no database from accuraterips.
Digital Audio Extraction
When I do CD rip, I have to turn off Accuraterip. In the beginning, in order to start Accuraterip, I also repeatedly tested a large number of CDs in order to find the accuraterip key disc, even in the past can be used as the acceleratorrip key disc, EAC also told me that it can not be used. In the end, I only thought about viewing the CD Drive Offsets list. As a result, there is no Offsets for BDR-S12J-X. Perhaps the use of Accuraterip's database would reduce the quality for BDR-S12J-X.
So you believe that the 99% of users who all get exactly the same result all get the wrong result, but the different result you get is somehow magically get the right one? 🙂