Stripped down ???
Nope.
These don't come with X. And that's about it.
DietPi is stripped down - a little more. Relevant ?? Not at all.
PicorePlayer @ "50MB" is what I'd call stripped down.
Relevant? Yep. If you want to store the entire OS in RAM you
better make sure the footprint is as small as possible.
The problem with TinyCore - the base OS of PcP - is, it is
awful to maintain.
BUT.
If an OS comes with 2Gig or 50MB doesn't really matter.
You won't be able to buy SD-cards below 16Gig anyhow.
What matters is the number and efficiency of processes
running on that machine.
Here again. PcP wins hands down.
Enjoy.
I think in a digital world everything is very deterministic. If software takes number 1011 from file and spits the same to I2S output then who cares does it use in memory OS does it use stripped OS or anything else. Behind I2S hardware comes to the picture - DACs etc.
If there is a way to make a real testing of different systems I could lend a couple of Pi boards. Without such testing everything is very subjective and biased.
If there is a way to make a real testing of different systems I could lend a couple of Pi boards. Without such testing everything is very subjective and biased.
I'd support that. Most inexperienced users demand a pretty much
plug'n play solution. Volumio, Moode, Rune, PcP and some others
were paving that road. That allowed many more users to enjoy a
RPI/ARM based audio solution.
Fact is. For many users burning an image to SD-card already feels like: "That's as far as I want (actually: able) to go."
Look. That PcP interface is meant to configure the OS and player. You do that once and then you forget about it!
Do you want fancy themes for that !?!? You gotta be kidding.
The GUI design is one of the most demanding exercises. If you look for nice GUIs checkout respective Android/iOS apps.
And all of you shouldn't forget. Most people involved in preparing these images/packages spent 100s of hours to get things in place and going.
Most of them not even have time to listen to music anymore.
Enjoy.
Last edited:
Here we're touching a hot topic. It's been discussed (at least) since I got myself involved in all this.
Just to mention it.
Tim never mentioned a word about soundquality or sound differences to similar OSes.
The people reporting it simply figured it out by experiencing it!
Here's my point of view:
"Bits are bits - it's all digital"
This is the main statement to begin with.
And this is absolutely correct, if we assume bitperfect transmission towards the DAC.
If "bit-perfection" or sometimes it's called "bit-transparency" is given, what else can make a sound difference on such a system!?!?
First of all we need to understand that bit-perfection refers to the "logical" state of bits. A 0 or a 1. And we can assume that a logical bit arrives at its target as intended.
However.
If you drive your car over a fresh paved freeway at 75mph with no wind, rain and other traffic or if you take an offroad track in bad weather conditions,
you'll get your car from A to B. The question is "how" it's done. Are there any side effects?
The computer with its peripherals is a pretty awful physical environment.
Basically an offroad track in the middle of a storm.
There's noise, all kind of it. There's EMI/RFI. There are power fluctuations.
There are different temperature conditions. There are vibrations. There
are reflections. And, and, and...
Many of these factors would influence an audio performance.
Noise, jitter (bit timing) , EMI/RFI, power fluctuations and ground loops are the typical sources for impacting your sound experience. You might want to add temperature and vibrations as factors.
Pretty much all DACs try to fight these factors. Some do with more some with less success. Obviously it's a matter of effort you'd be willing to put into it.
Meanwhile there are many gadgets out there fighting this physical mess - because the vast majority of DACs fail to do so properly.
You'll find USB or I2S isolators, reclockers, ethernet filters, high quality power supplies, shields, high quality USB and Ethernet cables, ground breaker asf asf.
All above factors are different for any audio system out there. Not a single noise floor will look the same. The noise floor is a mix of all kind of frequencies. It's like a fingerprint of a system. But that one can change and changes over time to make things worse.
Now. How can a OS or a OS setup change the fingerprint from a software perspective!?!?
Basically all you need to do is to start getting the rocks out of, leveling and paving your off-road track.
In terms of SW. We just increase the efficiency. There are numerous
ways of doing that. It's a very complex environment.
And everybody can understand that e.g. lower CPU load and/or a more steady CPU load will influence several of above mentioned factors.
If mpd runs at 5% (jumping over different processors and showing unsteady load down to 2%) or running rather steady at 1% CPU load on a single CPU after increasing the efficiency for doing the same job. It'll make a difference to the "fingerprint".
Of course there are numerous other factors in play.
The way to achieve these kind of gains are usually
the sum of several factors. The revisions of SW (solving flaws and/or increasing efficiency of kernel/drivers/apps), even the compilation of SW can make a difference. Its configurations and numerous OS tweaks will also increase efficiency. If you play a track from
a RAM disk or if you continuously stream over network will have an impact
to your systems fingerprint. If you realtime decode a flac stream or if you resample to 384k PCM it'll have an impact.
Many of these factors are widely known and been discussed numerous times. Some show more some less impact on your systems fingerprint.
Many times you need to find a compromise between this or that approach.
E.g. "features vs. performance"
And again. How much impact your "UPSTREAM" fingerprint will have on your DAC depends on your DACs capabilities to fight them.
That means. The worse your DAC the more impact you'll see from different
fingerprints. It also means the worse your DAC the more impact you might
experience from SW modifications - as an indirect way to impact the fingerprint.
The main issue is, that most of us hobbyists doesn't have the tools to prove
above in a scientific way. We base our wisdom on experiences - we often share with many others out there.
The professionals and manufacturers obviously are not interested to discuss these flaws in public. They rather claim femto seconds jitter performance and you then realize that hooking up a different USB cable makes a hell of a difference.
And then there are rather ignorant software nerds out there who'd stick to the "bits = bits" theory, completely ignoring that there is a physical world out there.
I'm still waiting for a DAC that's immune to all above mentioned factors.
At that point these discussions hopefully will come to an end.
Hmmh. Probably not.
99.999% of all users out there will not be able to afford such a device.
Enjoy.
Last edited:
Ι have been involved with the development of Archphile since the beginning of 2014 and I have done countless tests towards the direction you just mentioned above.
My current transport is:
- Odroid C2 with a DIY Linear Power Supply
- Archphile 0.99.7-alpha
1. Archphile doesn't use any kind of web server (apart from the integrated to ympd mongoose code), databases, php etc. and as a result is very minimal comparing to the other solutions
2. irq interrupts (usb, eth) are configured and sent to different isolated cores
3. MPD runs in its own isolated core
4. the MPD package I currently use is compiled with much less dependencies:
playground/PKGBUILD at master * archphile/playground * GitHub
Why did I just write all of the above? I am not willing to advertise (I really don't care at all), but to do the opposite.
I am 100% sure that if you ask me to find the transport that I have been building for the last 4 years on a blind test with different OSes (even with different boards to be honest), I won't be able to identify it.
During these almost 4 years, I have read stuff online that make no sense. There are Linux Audio users claiming that they hear differences between different kernel profiles (term introduced by Volumio/Runeaudio at a time where they were struggling with the old RPI and before the FIQ USB driver - they made kernel changes to avoid pops/click etc back at that time), between different MPD versions, between different compilation flags etc....
I am sure that there are people out there that can hear a difference between different RPI plastic enclosures.
In my opinion, all we are talking about is HI END SNAKEOIL that destroys computer audio everyday transforming users to the traditional hi end guys that at some day will place a small stone on the RPI to make it sound better.
The computer audio community today is very lucky. There are plenty of different hardware and software solutions to choose. Just use what you feel comfortable with and ensure that you won't spend your time on the configuration of the transport.
Listening to music is way more important than all these sh1t.
Last edited:
Obviously there are quite some people as inmate tuxx out there. I mentioned that in my last post.
Working on "Archphile" and "selling" it for years -- I'm not sure what the "...phile" is actually suggesting after his summary -- just to realize over and over that he can not make any sense out of his doing and then fooling the stupid crowd by keeping his project under that name up'n running... Hmmh.
But. I expected this feedback. There's more to come.
There were times where asynch USB DACs were the holy grail.
As soon as people started questioning that, after e.g. adding a powered USB HUB between PC and DAC - building without knowing a reclocker, regenerator, repower device - soon similar commercial solutions popped up.
Similar discussions started when talking about USB isolation. I bought
a Rover 200 from a meteorologist about ten years ago. And that worked
very well on my DDDAC during those days.
Some years later isolation as topic was up. Why that late? As soon as USB 2.0 isolator chips hit the market you could find commercial Audio USB isolators.
Talking about RPI.
The PI power rails are all but stable. What happens to clocks on unstable power rails !?!?
The USB ports are of very primitive nature. The CPU cooling is all but reasonable implemented by most users. Then there is the external power supply...
Then pair e.g. a HifiBerry DAC (powered by the PI rails) with such a device.
Achieving audiophile spheres with such a HW setup is gonna be quite a challenge.
Now with the RPI at hand. Is anybody questioning the value of the Kali as I2S reclocker??
Is anybody questioning the impact of separate power supplies for HATs ?? Why the hell are inmates designing HAT isolators !?!?
Yep. It's all snakeoil. There's no noise. There's no jitter. There's no EMI/RFI....
Bottom line.
Just let the people decide. If numerous people out there hear a difference -
without being biased beforehand - just listen to it.
For sure if you've failed over years to achieve anything with your system doesn't mean that you are right dear tuxx. It also doesn't mean you're wrong. Your finding might apply to your setup only. Who am I to question that !?!?
Enjoy. I do.
Working on "Archphile" and "selling" it for years -- I'm not sure what the "...phile" is actually suggesting after his summary -- just to realize over and over that he can not make any sense out of his doing and then fooling the stupid crowd by keeping his project under that name up'n running... Hmmh.
But. I expected this feedback. There's more to come.
There were times where asynch USB DACs were the holy grail.
As soon as people started questioning that, after e.g. adding a powered USB HUB between PC and DAC - building without knowing a reclocker, regenerator, repower device - soon similar commercial solutions popped up.
Similar discussions started when talking about USB isolation. I bought
a Rover 200 from a meteorologist about ten years ago. And that worked
very well on my DDDAC during those days.
Some years later isolation as topic was up. Why that late? As soon as USB 2.0 isolator chips hit the market you could find commercial Audio USB isolators.
Talking about RPI.
The PI power rails are all but stable. What happens to clocks on unstable power rails !?!?
The USB ports are of very primitive nature. The CPU cooling is all but reasonable implemented by most users. Then there is the external power supply...
Then pair e.g. a HifiBerry DAC (powered by the PI rails) with such a device.
Achieving audiophile spheres with such a HW setup is gonna be quite a challenge.
Now with the RPI at hand. Is anybody questioning the value of the Kali as I2S reclocker??
Is anybody questioning the impact of separate power supplies for HATs ?? Why the hell are inmates designing HAT isolators !?!?
Yep. It's all snakeoil. There's no noise. There's no jitter. There's no EMI/RFI....
Bottom line.
Just let the people decide. If numerous people out there hear a difference -
without being biased beforehand - just listen to it.
For sure if you've failed over years to achieve anything with your system doesn't mean that you are right dear tuxx. It also doesn't mean you're wrong. Your finding might apply to your setup only. Who am I to question that !?!?
Enjoy. I do.
Working on "Archphile" and "selling" it for years -- I'm not sure what the "...phile" is actually suggesting after his summary -- just to realize over and over that he can not make any sense out of his doing and then fooling the stupid crowd by keeping his project under that name up'n running... Hmmh.
Yes, it's true, after my statement I will lose my customers
It's the last time Ι mention Archphile in this topic, but as you are not sure about what the "phile" is suggesting, let me help you clarify:
Archphile - A Linux Audiophile Distribution - FAQ
I did not state that there's no jitter or EMI/RFI. I just believe that claiming that you hear a difference in sound because of a minor kernel setting (or other similar stuff) is a PLACEBO.
Anyways, I am just a hobbyist (of course I don't "sell" anything contrary to what you mentioned) and just I expressed my opinion on all these non-sense computer audio details that people is worrying about.
Last edited:
You call everything that's been discussed snakeoil, placebo, nonsense.
Without coming up with any general argument supporting your "opinion".
Nope. It's not enough that you couldn't find any difference and derive from
that the rest of the world is nuts if they do experience (or even measure) differences!
At the same time selling an "audiophile" piece of SW build around these
snakeoil myths. How do you actually develop and test your audiophile package?
Or the tunings you've been applying over the years?? You don't hear any differences!??!
Or do you just blindly copy/paste stuff you catch somewhere else
and sell it under your audiophile brand !??
You know what. Sorry. Your attitude sucks!
And then you even admit you havn't understood a tiny bit what's been said.
Knowing that jitter and other disturbing factors exist won't get you anywhere!
You need to connect the logical and the physical world and combine that with your setup.
That's what I've been talking about.
DIY-Audio is a place different from e.g. Computer Audiophile or Audio Asylum.
Usually people working hand in hand in a pretty constructive and openminded manner. There are many great examples.
Obviously this is not the case with people like you Tuxx.
I don't give up hope that this might change one day.
Enjoy.
Without coming up with any general argument supporting your "opinion".
Nope. It's not enough that you couldn't find any difference and derive from
that the rest of the world is nuts if they do experience (or even measure) differences!
At the same time selling an "audiophile" piece of SW build around these
snakeoil myths. How do you actually develop and test your audiophile package?
Or the tunings you've been applying over the years?? You don't hear any differences!??!
Or do you just blindly copy/paste stuff you catch somewhere else
and sell it under your audiophile brand !??
You know what. Sorry. Your attitude sucks!
And then you even admit you havn't understood a tiny bit what's been said.
Knowing that jitter and other disturbing factors exist won't get you anywhere!
You need to connect the logical and the physical world and combine that with your setup.
That's what I've been talking about.
DIY-Audio is a place different from e.g. Computer Audiophile or Audio Asylum.
Usually people working hand in hand in a pretty constructive and openminded manner. There are many great examples.
Obviously this is not the case with people like you Tuxx.
I don't give up hope that this might change one day.
Enjoy.
Last edited:
You call everything that's been discussed snakeoil, placebo, nonsense.
Without coming up with any general argument supporting your "opinion".
Nope. It's not enough that you couldn't find any difference and derive from
that the rest of the world is nuts if they do experience (or even measure) differences!
At the same time selling an "audiophile" piece of SW build around these
snakeoil myths. How do you actually develop and test your audiophile package?
Or the tunings you've been applying over the years?? You don't hear any differences!??!
Or do you just blindly copy/paste stuff you catch somewhere else
and sell it under your audiophile brand !??
You know what. Sorry. Your attitude sucks!
And then you even admit you havn't understood a tiny bit what's been said.
Knowing that jitter and other disturbing factors exist won't get you anywhere!
You need to connect the logical and the physical world and combine that with your setup.
That's what I've been talking about.
DIY-Audio is a place different from e.g. Computer Audiophile or Audio Asylum.
Usually people working hand in hand in a pretty constructive and openminded manner. There are many great examples.
Obviously this is not the case with people like you Tuxx.
I don't give up hope that this might change one day.
Enjoy.
@Soundcheck it's obvious that you really don't like it when others disagree with you.
- To begin with, you keep mentioning the words "selling" and "brand". It's clear that you have tried to establish false impressions about me. I am not selling anything and Archphile is not a brand. I am a diyer and as a free software enthousiast, I keep sharing EVERYTHING with the community since day one. I am very curious if you have ever published any analytical guide on your own linux transport tweaks or configuration.
- You blamed me once again that I called everything a snakeoil/placebo/nonsense. This is NOT true. What I called a placebo (sound-wise) and I insist on it, are sophisticated kernel modifications, sonic differences between minor software releases etc, OS tweaks like CPU isolation etc.
This is VERY different from an I2S recklocker (which I OWN one) or a USB isolator (which again I OWN one), or with general terms like jitter (LOL).
I also stated that most possibly I would not be able to identify my distro on a blind test. I have no problem to state it again and again, and you know why? Because I don't have the hi-end illusion that my ears are "golden" and because I believe in a scientific field known as psychoacoustics.
- You blamed me for "selling" a piece of software based on snakeoil myths, which is again not true. If I wanted to "sell snakeoil", I would not give the recipe,the or the configuration files.
- You also wondered how I actually develop and my distro. The answer is very easy: with thousands of tests and measurements. And no, I don't copy/paste configuration tips I cought somewhere else. In my previous post, I mentioned that my odroid image has extra tweaking with regards to irq interrupts, cpu isolation etc. Below you will find all the explanations along with measurements that will give you an idea why I've done all this tweaking:
http://www.diyaudio.com/forums/pc-b...ion-raspberry-pi-udoo-quad-7.html#post4863821
http://www.diyaudio.com/forums/pc-b...ion-raspberry-pi-udoo-quad-7.html#post4864057
As you see, apart from tweaks that affect sound, there are others that affect the overall performance and stability of crappy devices like the Odroid C2 or the Raspberry Pi.
I assume that based on the post above, I really do have understood in which forum I participate, right?
To be honest, I am not used to be dragged into non-productive forum discussions, especially with people that have the illusion of being superior to others and this will be my last post to you.
Regards,
Michael
Last edited:
The discussion becomes too hot. I see that some testing and measurements were mentioned several times. Is it possible to come up with well-defined tests in this area at all? Does that need any hardware (oscilloscope, logical analyzer...) or it should be possible to use just software tests? Is it possible to eliminate DAC from this equation or not?
Again if any knowledgeable (not biased) person could volunteer to conduct such tests I could provide couple Pi boards with pre-installed images in question. Otherwise this will be everlasting debate like who is stronger whale or elephant.
Again if any knowledgeable (not biased) person could volunteer to conduct such tests I could provide couple Pi boards with pre-installed images in question. Otherwise this will be everlasting debate like who is stronger whale or elephant.
That device you picked would be difficult to achieve that.
Get the IQ Audio Pi DAC+ - it will blow you away and be much easier to implement. IQ Audio do a great case, too.
Archphile was the only mpd based distro that actually managed to handle my 160000+ songs in the library.
Rune, Volumio and especially Moode was slow, unstable, laggish and did not meet up to my expectations.
I actually felt it sounded better too, but that may perfectly be my imagination.
The only distro I actually advise using for MPD based playback is archphile. (using rpi2)
Thanks tuxx for your great support and software.
Rune, Volumio and especially Moode was slow, unstable, laggish and did not meet up to my expectations.
I actually felt it sounded better too, but that may perfectly be my imagination.
The only distro I actually advise using for MPD based playback is archphile. (using rpi2)
Thanks tuxx for your great support and software.
Look. Keeping on this kind of discussion is more then needed.
For many people the soundquality factor has major influence
on the decision process.
Soundquality is the ultimate subject underlying our hobby. At least for most of us.
There are 10-thousands of people out there, endless sites
and businesses focusing on this subject only. And this is going on for decades.
I do know that people like Tim (Moode) are afraid of talking
about sound quality. They simply feel better not being put into the snakeoil fraction by folks like Tuxx.
And this discussion is not about Archphile vs. anything.
It's the guy called Tuxx and his attitude which is at stake!
I do agree that such an attitude is not very beneficial for such a thread.
But. These guys will kill and are killing and have been killing any thread when it comes to discussions about sound quality.
I didn't accept it in 2007 and I don't accept this attitude today.
Enjoy.