I would like to keep this discussion alive with some info about the USB board.
Hope it is useful for someone.
I found this one works at once and fine, with all the connections in place, the spdif one as well.
As always, placing the USB board as close as possibile to the main board is essential.
The seller emails the drivers on request.
AIYIMA XMOS Scheda Figlia XU208 Scheda di Decodifica Supporto 32BIT 384 K ES9018 Decodifica Compatibile Amanero USB IIS Interfaccia Digitale|Amplificatore| - AliExpress
Also I am not sure how to get rid of the second opamps pair.
Is it still possible to go unbalanced without them?
Hope it is useful for someone.
I found this one works at once and fine, with all the connections in place, the spdif one as well.
As always, placing the USB board as close as possibile to the main board is essential.
The seller emails the drivers on request.
AIYIMA XMOS Scheda Figlia XU208 Scheda di Decodifica Supporto 32BIT 384 K ES9018 Decodifica Compatibile Amanero USB IIS Interfaccia Digitale|Amplificatore| - AliExpress
Also I am not sure how to get rid of the second opamps pair.
Is it still possible to go unbalanced without them?
Last edited:
Do they claim or they really are? 
Apart that, reading the discussion many seem to have met many problems with the usb boards, apparently leading to even main board malfunctions.
I don't know if the one I suggested is the best, but at least works, plug and play for about 25$ a pop.
For the knowledge of other readers would you mention the ones supposed to be better and the reason they are such?
Thanks
Apart that, reading the discussion many seem to have met many problems with the usb boards, apparently leading to even main board malfunctions.
I don't know if the one I suggested is the best, but at least works, plug and play for about 25$ a pop.
For the knowledge of other readers would you mention the ones supposed to be better and the reason they are such?
Thanks
Some have better clocks than others. Some feature galvanic isolation, which can help reduce ground loop and ground noise issues from USB sources such as computers. Some can run from clean power rather than USB power, others can be modified to similar effect.
IME if a dac is good enough, power amp is good enough, and speakers are good enough, then using good clocks and generally doing things right does audibly help SQ in multiple ways. However, while typical AP dac measurements provide some useful figures of merit, they do not completely predict SQ. Clocking is one of those issues that is not easily visualized in a typical audio FFT, so some people appear to me to be in denial about how SQ can be affected.
Bottom line, when people talk about I2SoverUSB being well worth the cost, there are good reasons for it.
The Chinese USB to I2S boards that have Accusilicon AS318-B clocks can also be quite good if modified to run on clean power, although there remains a problem with clock warm-up times when switching between 44kHz and 48kHz sample rate families. IME good clocks need to be left running continuously for best performance. IME 3-days of continuous power-on time is required for clocks to fully stabilize at the lowest jitter/phase-noise. Some clocks can take even longer. The Chinese USB board with good clocks do not leave both clocks running at all times, so they are best if one sticks to playback of sample rates from a single clock family.
IME if a dac is good enough, power amp is good enough, and speakers are good enough, then using good clocks and generally doing things right does audibly help SQ in multiple ways. However, while typical AP dac measurements provide some useful figures of merit, they do not completely predict SQ. Clocking is one of those issues that is not easily visualized in a typical audio FFT, so some people appear to me to be in denial about how SQ can be affected.
Bottom line, when people talk about I2SoverUSB being well worth the cost, there are good reasons for it.
The Chinese USB to I2S boards that have Accusilicon AS318-B clocks can also be quite good if modified to run on clean power, although there remains a problem with clock warm-up times when switching between 44kHz and 48kHz sample rate families. IME good clocks need to be left running continuously for best performance. IME 3-days of continuous power-on time is required for clocks to fully stabilize at the lowest jitter/phase-noise. Some clocks can take even longer. The Chinese USB board with good clocks do not leave both clocks running at all times, so they are best if one sticks to playback of sample rates from a single clock family.
Last edited:
Thank you for the explanation.
But, is there at least one available for purchase and plug and play compatible with this main board, after all we are talking about it, that can do so?
If it does not cost too much I would purchase it at once.
I don't have any interest in making Aiyima USB's advertising, but anyway the one I linked has been able to clearly show the improvement in sound quality that came out with switching from a Seagate Barracuda hard disk to a Western digital Blue SSD in my HP laptop.
So, I don't know if it has clock/jitter issues so big.
In the end, I wonder how many listeners could discern between a just switched on "normal" usb board and one three days under continuous power without knowing which is which.
But, is there at least one available for purchase and plug and play compatible with this main board, after all we are talking about it, that can do so?
If it does not cost too much I would purchase it at once.
I don't have any interest in making Aiyima USB's advertising, but anyway the one I linked has been able to clearly show the improvement in sound quality that came out with switching from a Seagate Barracuda hard disk to a Western digital Blue SSD in my HP laptop.
So, I don't know if it has clock/jitter issues so big.
In the end, I wonder how many listeners could discern between a just switched on "normal" usb board and one three days under continuous power without knowing which is which.
There shouldn't be any difference in sound from changing disk drives. If there is, it means you probably should have galvanic isolation and clean power rather than USB power for your USB board.
I2SoverUSB is configurable to work with many different types of dacs, it has good clocks, galvanic isolation, etc. The only thing is that you have to supply a clean 5v power supply for it (same as you would have to do with any other USB board running on clean power). Best to use a linear supply rather than a switcher if at all possible.
I2SoverUSB - I2S over USB Audio
If you want to check out the manual first:
http://jlsounds.com/uploads/I2SoverUSB v.III.pdf
I2SoverUSB is configurable to work with many different types of dacs, it has good clocks, galvanic isolation, etc. The only thing is that you have to supply a clean 5v power supply for it (same as you would have to do with any other USB board running on clean power). Best to use a linear supply rather than a switcher if at all possible.
I2SoverUSB - I2S over USB Audio
If you want to check out the manual first:
http://jlsounds.com/uploads/I2SoverUSB v.III.pdf
I have a setup that’s required to use a usb source, and have vastly improved the power supply of the input board using an old wall wart, and an lm7805 with some capacitors.
The improvement was shocking, especially considering how the parts were just surplus stuff from under the workbench I used to test the theory that the laptop power was gritty. It may improve again if I were to use a better transformer, and regulator, or even a battery.
The improvement was shocking, especially considering how the parts were just surplus stuff from under the workbench I used to test the theory that the laptop power was gritty. It may improve again if I were to use a better transformer, and regulator, or even a battery.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.