There may be measurements of both at ASR. Also, IIRC ASR found one problem in Revolution that has since been fixed.
However, some of the differences in sound quality will not show up using standard measurements. In particular, some people have found that RF leaking out of dac chip analog outputs can cause non-stationary or quasi-non-stationary distortion in opamp output stages, particularly in the differential summing multi-feedback filter stage. So far as I can tell, the distortion does not stay exactly constant during audio FFT acquisition time. There may be some very small spurs visible, but fixing them can make bigger than expected sound quality improvement. One forum member here did some measurements and found that analog output RF noise increased greatly if opamps other than OPA1612 were used in the output stage of AK4499, a dac which uses a similar output stage topology to what is used for ESS dac chips. Of course, Allo Revolution uses the more expensive OPA1612 opamps because they sound better and are used on the ESS and AK4499 evaluation boards. KTB uses cheaper opamps, approximately similar to an upgraded 5532 type of thing. They had to do that sort of thing to keep the price so low. Therefore besides looking at measurements if there are any trusted reviewers you listen to, you might see if any have reviewed the sound of both dacs. Hans Beekhuyzen Channel on youtube migt have reviews for both.
However, some of the differences in sound quality will not show up using standard measurements. In particular, some people have found that RF leaking out of dac chip analog outputs can cause non-stationary or quasi-non-stationary distortion in opamp output stages, particularly in the differential summing multi-feedback filter stage. So far as I can tell, the distortion does not stay exactly constant during audio FFT acquisition time. There may be some very small spurs visible, but fixing them can make bigger than expected sound quality improvement. One forum member here did some measurements and found that analog output RF noise increased greatly if opamps other than OPA1612 were used in the output stage of AK4499, a dac which uses a similar output stage topology to what is used for ESS dac chips. Of course, Allo Revolution uses the more expensive OPA1612 opamps because they sound better and are used on the ESS and AK4499 evaluation boards. KTB uses cheaper opamps, approximately similar to an upgraded 5532 type of thing. They had to do that sort of thing to keep the price so low. Therefore besides looking at measurements if there are any trusted reviewers you listen to, you might see if any have reviewed the sound of both dacs. Hans Beekhuyzen Channel on youtube migt have reviews for both.
Last edited:
I used to have few different DACs included Bench Mark 2 which I sold and now have KTB. I am ok before moving to a next one.
Khadas review
Review and Measurements of WesionTEK Khadas Tone Board DAC | Audio Science Review (ASR) Forum
Khadas review
Review and Measurements of WesionTEK Khadas Tone Board DAC | Audio Science Review (ASR) Forum
Last edited:
Alashikata,
Thanks for sharing the link and your comments.
I've also went through the ASR review just after it came out, and saw that the KTB measures very well. I haven't been able to find a similar review with measurements for the Allo Revolution DAC yet.
Markw4,
Will be on the look out for more info/measurements of the Allo DAC.
Thanks for sharing the link and your comments.
I've also went through the ASR review just after it came out, and saw that the KTB measures very well. I haven't been able to find a similar review with measurements for the Allo Revolution DAC yet.
Markw4,
Will be on the look out for more info/measurements of the Allo DAC.
I stop spending on high end DAC since too many keep come out with new chipset. I like the KTB over Bench Mark 2 because it's support higher. However, the Bench Mark 2 is very decent.
I am not sure about Allo Revolution DAC but look like it's support DSD 512. I heard the KTB will come out new driver too which it will support DSD too but not sure if it works with DSD 512. Current driver works with DSD 64.
Khadas will come out with XLR connector in couple weeks with better specs so I will wait to see what they have.
I am not sure about Allo Revolution DAC but look like it's support DSD 512. I heard the KTB will come out new driver too which it will support DSD too but not sure if it works with DSD 512. Current driver works with DSD 64.
Khadas will come out with XLR connector in couple weeks with better specs so I will wait to see what they have.
Hello everyone,
I have my Khadas Tone Board (got the VIMs edition). I hadn't realized earlier (and had not read enough, or carefully enough), and now know that it is a USB only DAC, and the pins on the VIMs edition will only work with the Khadas products designed to work with it.
Now I was thinking about running this with a Raspberry Pi.
Came across these reviews:
Raspberry Pi 4B for music playback - YouTube
The Raspberry Pi 4 Model B: Game changer for audio? | Dimdim's Blog
Both say that the RPi 3 sounds better vs the RPi4?? Should it be?
My understanding is that the RPi 4 is a more powerful device and the USB has higher bandwidth also.
I haven't bought the RPi yet, so I am open to suggestions.
I have my Khadas Tone Board (got the VIMs edition). I hadn't realized earlier (and had not read enough, or carefully enough), and now know that it is a USB only DAC, and the pins on the VIMs edition will only work with the Khadas products designed to work with it.
Now I was thinking about running this with a Raspberry Pi.
Came across these reviews:
Raspberry Pi 4B for music playback - YouTube
The Raspberry Pi 4 Model B: Game changer for audio? | Dimdim's Blog
Both say that the RPi 3 sounds better vs the RPi4?? Should it be?
My understanding is that the RPi 4 is a more powerful device and the USB has higher bandwidth also.
I haven't bought the RPi yet, so I am open to suggestions.
Yes, Tone Board is not a HAT device that attaches to the Pi. As far as the Pi is concerned, i would go with the latest version. On the other hand, if your desire is to run the USB out and into the K Tone Board, I would suggest getting an Asus Tinker Board. It has a much better USB out and you can also set it up for SPDIF out and it has a headphone out. I have been running an Asus Tinker Board for several years with Volumio free software and running it via USB out into a Emotiva DAC. It works great. Almost all the streaming software that folks uss on the Pi have versions for Tinker Board.
RPi4 generates and radiates much more EMI/RFI than RPi3. That's what people are reporting the audible effects of. It helps to physically separate the RPi and the dac hat as much as possible. Putting shielding between them may help some too. Some products have been designed are in the process of being designed to help minimize the problem. Iancanada offers groupbuys of such things from time to time, for example.
@Markw4 and others:
I have the Khadas Tone1 DAC and would like to know if it is an oversampling DAC by default.
So I am curious how for example 320kbps Ogg Vorbis from Spotify or 16/44.1 FLAC from Qobuz is processed in the Khadas.
Is there a certain oversampling rate for all incoming data or does it work as NOS DAC?
Any info is very much appreciated.
Thank you
Matt
I have the Khadas Tone1 DAC and would like to know if it is an oversampling DAC by default.
So I am curious how for example 320kbps Ogg Vorbis from Spotify or 16/44.1 FLAC from Qobuz is processed in the Khadas.
Is there a certain oversampling rate for all incoming data or does it work as NOS DAC?
Any info is very much appreciated.
Thank you
Matt
Internally, the 9038Q2M in the DAC uses a 8x oversampling filter, like almost all delta-sigma DACs. Primary function of the filter is to shift the Nyquist frequency well above the human hearing range, and thus improve performance of the digital filter. This also has the effect of increasing the S/N ratio and apparent resolution (depending on analog filter).
This has nothing to do with the incoming bitrate, samplerate or the compression method used. Lossy compression algorithms use padding to fill the data back to 1411kbps rate required for a 16/44 stream before sending it outboard for conversion. This is done by the playback system, not the DAC. Any upsampling/bit padding is done at this stage. The DAC, for all intents and purposes, sees a 16/44 stream (or whatever you converted it into).
The 9038Q2M cannot be operated in the 1X mode like the 9038PRO can, and thus does not qualify as a NOS DAC. Some TI DACs can be operated with the oversampling filter disabled. Benchmark does this in their DAC2 (and I assume the later models). There is a nice explanation of why they chose to do it, I forgot where; John explained it in some detail.
This has nothing to do with the incoming bitrate, samplerate or the compression method used. Lossy compression algorithms use padding to fill the data back to 1411kbps rate required for a 16/44 stream before sending it outboard for conversion. This is done by the playback system, not the DAC. Any upsampling/bit padding is done at this stage. The DAC, for all intents and purposes, sees a 16/44 stream (or whatever you converted it into).
The 9038Q2M cannot be operated in the 1X mode like the 9038PRO can, and thus does not qualify as a NOS DAC. Some TI DACs can be operated with the oversampling filter disabled. Benchmark does this in their DAC2 (and I assume the later models). There is a nice explanation of why they chose to do it, I forgot where; John explained it in some detail.
So the output (after padding) from Spotify to the DAC is always 16/44 (and not 320kbps Ogg Vorbis for example)?
Thx
Matt
Last edited:
Yes, it is - if you're not resampling the native data, or have the decoder set up differently. DACs do not understand OGG or MP3. They just look at what the data header says and try to sync their clocks to that frequency. The way to think about is that the stream is decompressed before reaching the DAC, by a system that understands how to decompress it. Of course, lacking the original information it is dependent on software interpolation (which is what the OGG/MP3 decoders in your playback software do).
Hello KTB users. Has someone experienced some issue with usb connector? After 2 weeks of usage connected to a RPi 2B, ktb stopped working Maybe some short circuit (log On rpi says “over current change). I found an alternative to feed signal+power to Khadas using its gpio strip but i would understand where to check possibile short in the board (i.e. Test points and relative voltage values).
Thank you!
Thank you!