Hi Vince,
I am also interested in possible software tweaks. I am more skilled in hardware and rather not in software, so I am very interested in what you have found. So please share what you have found (and withstand the non-believers that immediately will jump on you, unless you have scientifically proved it, you have evidence for it and have double blind tested it all 🙂
).
I develop and build all my audio gear (amplifiers, DAC, power supplies ect.) completely on the basis of listening tests. From scratch to end result I do hundreds of small listening tests to decide what is better/optimal in a given situation. After having squeezed out the most of a design I start to design PCB boards and do the actual build of it. Take a good design (or perhaps design it yourself) and see where it can bring you…! For me that whole process is big fun and is what I like most of High End Audio!
If you have tweaks that you can share, I am willing to test it in my system and report back what I have found.
Peter
I am also interested in possible software tweaks. I am more skilled in hardware and rather not in software, so I am very interested in what you have found. So please share what you have found (and withstand the non-believers that immediately will jump on you, unless you have scientifically proved it, you have evidence for it and have double blind tested it all 🙂
).I develop and build all my audio gear (amplifiers, DAC, power supplies ect.) completely on the basis of listening tests. From scratch to end result I do hundreds of small listening tests to decide what is better/optimal in a given situation. After having squeezed out the most of a design I start to design PCB boards and do the actual build of it. Take a good design (or perhaps design it yourself) and see where it can bring you…! For me that whole process is big fun and is what I like most of High End Audio!
If you have tweaks that you can share, I am willing to test it in my system and report back what I have found.
Peter
Daphile is a standard linux distribution, using regular linux kernel, 3.18 as of the latest version. The kernel provides drivers for all the soundcards https://git.kernel.org/cgit/linux/k...d?id=b2776bf7149bddd1f4161f14f79520f17fc1d71d
For playback it uses standard squeezeplay https://github.com/Logitech/squeezeplay, playing through alsa-lib library git.alsa-project.org Git - alsa-lib.git/summary . It uses a proprietary web interface for control.
Why should daphile sound different to other bit-perfect bare-bones linux installations?
How can it sound different between versions?
Voodo rules the audiophile world.
For playback it uses standard squeezeplay https://github.com/Logitech/squeezeplay, playing through alsa-lib library git.alsa-project.org Git - alsa-lib.git/summary . It uses a proprietary web interface for control.
Why should daphile sound different to other bit-perfect bare-bones linux installations?
How can it sound different between versions?
Voodo rules the audiophile world.
If you want to tweak the software, use any of the many no-gui linux distributions which provide access to command line, either directly or via ssh.
But won't a circuit done using a PCB sound completely different than one done on a prototyping board?
No provoking for the sake of provoking - simply clarifying and questioning some of the more controversial claims.
To me, DIY is not just about assembling something yourself, but about learning and understanding. Audio systems are based on electronics, acoustics and increasingly information theory - all fields of science and technology. Thus what we do is apply theory from those fields in practical designs.
I think we do the community disservice if we propagate unverified anecdotal claims, hearsay and purely subjective opinions that contradict established knowledge in those fields.
Yes indeed, layout is critical. It will often sound better when everything is tightly placed on a PCB. But sometimes I am only comparing values of components or different types of components that fit in the same space.
What improved the sound of my Daphile computer was the use of an isolation transformer. This transformer isolates the computer from the mains voltage and blocks a lot of noise coming out of it. This noise can otherwise affect the rest of the audio equipment.
Also placement of both Daphile computer and transformer on a dedicated audio rack gave a clear and easy to distinguish improvement in SQ.
Peter
But might it not be that a component value that was best in the prototype circuit isn't optimal any more on a tight PCB?
Wouldn't a LC filter accomplish the same thing?
Have you tried to subject the gear to bumps or vibration to see if any of the components are microphonic?
The Loki sounds OK. I don't have many DSD tracks and none to compare to my 24-bit 192K records. I'm working on getting more records in DSD. Not cheap!
The sound is smooth and there's a very quiet background. It takes a bit of getting accustomed to it. Schitt Audio sells a USB power bypass to power the Loki instead of using the power from the PC's USB. I know someone who uses the Wyrd USB interface and likes the results. He does use it with the Gungnir, but I think the Loki would benefit from it since it only has USB power.
I really like the way Daphile can switch between DAC so easily. It does run slower than the RPi/Volumio setup. Switching between tracks is slow, maybe due to reading the database. I don't know. Away, it's a consistent speed.
Aside from Daphile software/OS itself, there are several places where tweaks can be made. Starting with the case and the power supply. For example, if using solid state drive (SSD), try using an external SATA power supply and not using the main board power supply. As for the case, use aluminum instead of steel. I've seen where people have glue vinyl to the inside of a case to dampen it. Of course there are other materials to help dampen vibrations.
There's also the BIOS of the motherboard. Making adjustments and shutting things off that are not necessary, for example legacy USB support. Caution must be used when adjusting the BIOS, or things can get worse rather than better.
This is why I asked originally if anyone has tried anything. 😀
The sound is smooth and there's a very quiet background. It takes a bit of getting accustomed to it. Schitt Audio sells a USB power bypass to power the Loki instead of using the power from the PC's USB. I know someone who uses the Wyrd USB interface and likes the results. He does use it with the Gungnir, but I think the Loki would benefit from it since it only has USB power.
I really like the way Daphile can switch between DAC so easily. It does run slower than the RPi/Volumio setup. Switching between tracks is slow, maybe due to reading the database. I don't know. Away, it's a consistent speed.
Aside from Daphile software/OS itself, there are several places where tweaks can be made. Starting with the case and the power supply. For example, if using solid state drive (SSD), try using an external SATA power supply and not using the main board power supply. As for the case, use aluminum instead of steel. I've seen where people have glue vinyl to the inside of a case to dampen it. Of course there are other materials to help dampen vibrations.
There's also the BIOS of the motherboard. Making adjustments and shutting things off that are not necessary, for example legacy USB support. Caution must be used when adjusting the BIOS, or things can get worse rather than better.
This is why I asked originally if anyone has tried anything. 😀
Last edited:
Shutdown/restart Problems
Hello
I have a new musicserver (Zotac ZBOX nano CI320/Daphile 64-bit x86) .
I tested everything with my laptop and booting from usb-stick with sucess before. I have now the problem , that after doing a shutdown or restart the server hangs .I got last Message (alt + F2): "remounting remaining filesystems readonly [ok]" . It is the same if i do that over the webui or the powerbutton. The only way to stopp the server is to push the powerbutton for several seconds . Only after doing this the server/PC shutdown . Is there a solution for this issue ?
Greetings from Germany
peter
Hello
I have a new musicserver (Zotac ZBOX nano CI320/Daphile 64-bit x86) .
I tested everything with my laptop and booting from usb-stick with sucess before. I have now the problem , that after doing a shutdown or restart the server hangs .I got last Message (alt + F2): "remounting remaining filesystems readonly [ok]" . It is the same if i do that over the webui or the powerbutton. The only way to stopp the server is to push the powerbutton for several seconds . Only after doing this the server/PC shutdown . Is there a solution for this issue ?
Greetings from Germany
peter
I have experimented with different case materials for my DAC recently and preamp in the past and the worst is steel, then aluminum, then copper and the best is….. simply wood. So my DAC has a wooden case and I avoid the use of too much metal. Additionally I damped the wooden case with this self-adhesive stuff:
http://www.amazon.de/Presto-Antidrhnplatten-Farbe-schwarz-Stck/dp/B00BSOND9W/?tag=tivital0f-21
I buy it at a local car parts shop. You can heat it a little with a heat gun and it will easily stick on all sorts of curved surfaces. I also use it directly on electrolytic capacitors with great effect!!
Probably wood is no option for a PC housing because it has no shielding effects at all for all the high frequency garbage inside. But I will try it nevertheless when I am finished with my DAC and the Daphile PC will be the next project.
Peter
How does the case material affect the sound...
What are the mechanisms?
What is the effect on the signal?
Have you any empirical data?
Of course using wood provides no shielding whatsoever, but even using metal to shield, unless you know what you are doing using a metal case incorrectly can increase EMC problems.
There are two mechanisms that possibly have an effect here:
1. Wood, especially in combination with self-adhesive damping material has favorable anti resonant properties.
2. Magnetic fields induce electrical currents in metal and the metal itself will radiate a magnetic field. When you use wood (or another non conducting material) you avoid that mechanism.
That is why a prototype on a wooden plank suddenly has lost its magic when everything is neatly built in a beautiful metal case. Sometimes people write, perhaps it needs burn in time, but that’s not the real reason.
Peter
Have you tried to subject the gear to bumps or vibration to see if any of the components are microphonic?
I have to very stringent levels, and also ballistic tests...for the vibration tests the equipment HAD to work with no chance of failure... all in metal cases, this is why I wonder sometimes regarding vibrational problems in a domestic set up, which I consider a VERY benign environment for electronics, even when you may get some resonances caused by the speaker outputs, even then they are not that bad and basic damping will solve them.
Again with metal cases how bad is the induced eddy currents, minor I would suspect and the benefits of proper screening in the RF rich world we live in today far outweighs any interaction between the electronics and the case.....
I do believe the wooden enclosure thing to be a perceptual improvement and would suspect under certain test conditions where the enclosure can not be seen that the difference would not be noticeable...
Again with metal cases how bad is the induced eddy currents, minor I would suspect and the benefits of proper screening in the RF rich world we live in today far outweighs any interaction between the electronics and the case.....
I do believe the wooden enclosure thing to be a perceptual improvement and would suspect under certain test conditions where the enclosure can not be seen that the difference would not be noticeable...