Copyright © 1998
Cadman Enterprises Ltd
Welcome to the Graham Slee Audio Products Owners Forum Open to all owners plus those contemplating the purchase of a Graham Slee HiFi System Components audio product and wishing to use this forum's loaner program: join here (Rules on posting can be found here) This website along with trade marks Graham Slee and HiFi System Components are owned by Cadman Enterprises Ltd |
Achieving High Fidelity Sound |
Post Reply | Page <1 1011121314 28> |
Author | |
Graham Slee
Admin Group Retired Joined: 11 Jan 2008 Location: South Yorkshire Status: Offline Points: 16298 |
Post Options
Thanks(1)
|
Power supplies will always have noise as a component of their DC, but it is in the order of 3mV from a linear voltage regulator, and typical noise. Voltage regulators often have a capacitor on their output to ground, which attenuates high frequency noise. SMPS noise is approximately the same at up to say, 100kHz, after which their outputs must feature RF chokes to attenuate their high frequency noise (similar to filtering the sampling rate in digital audio, but using inductors to preserve the DC current). Whereas you can hold the voltage constant, the load draws current, which in anything from an amplifier to a processor, when active, will be a continual and continuously changing series of pulses on top of other continuously changing series of pulses, which because of their pseudo randomness, will place the current demands of those "noises" on the voltage regulator output. A linear regulator will combat such noise up to 100kHz to 1MHz depending on the regulator, after which the output capacitor takes over, in making the power source low impedance. Sometimes a linear regulator will oscillate slightly, due to it not being a perfect Thevenin source. It is the equivalent of a resistor in series with an inductor, and there is the capacitive load. The resistor can be seen as the regulation element, which there is no getting away from. The combination can have a Q which makes it resonant under some capacitive loads, and often, if the output filter capacitor is large, a small resistance has to be put in series with it. Such oscillations are so small that most linear regulator designs don't bother - after all, it would be negated by the load's own capacitance. A series resistor in the load's path might help, but then again, reduces the regulation. Generally output resistance and impedance is "brute force" (i.e. none to speak of - very little), and so the load must also have its own power supply "ripple" (should read "noise") rejection (PSRR). However, PCBs have inductance in their power lines (20nH per inch), which demands local decoupling of high speed device power supply terminals (pins, grid balls, etc). The bipolar 555 timer is often used to educate how this inductance combines with the device when it switches, and puts its "glitch" on the supply voltage trace. Provided the board designer and system designer have mitigated such behaviour by best practice decoupling, all should be fine.
|
|
That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
|
|
Ash
Senior Member Joined: 18 Mar 2013 Location: Dorset Status: Offline Points: 4334 |
Post Options
Thanks(0)
|
Thanks Graham. Got an Enigma upgrade on the way now then the DAK supply early next year. Still got MySphere 3.1 so I can scrutinize any changes in the digital transport.
EDIT- The Enigma has arrived today, very fast shipping, thanks John C. Another piece of the puzzle set in place. Edited by Ash - 23 Nov 2021 at 2:33pm |
|
Ash
Senior Member Joined: 18 Mar 2013 Location: Dorset Status: Offline Points: 4334 |
Post Options
Thanks(0)
|
As you will have read, I have been trying to get my head around the basics of computers and all the technical jargon, looking to build a foundation for further reading and investigation. A computer is a combination of hardware and software, and both aspects matter. Having a great piece of hardware is pointless without good software/driver/kernel/firmware/BIOS support. Similarly, having great software on poor performing hardware is a complete waste. Software compatibility for a particular type of hardware can be limited by the ISA (instruction set architecture) of the processor. The programming will be for one type of architecture only unless multiple releases are available. Consequently, to enjoy the freedom to experiment with all available software, I have to possess computers with the different ISAs. Hence my decision to finally try a Raspberry Pi 3B+; ARM64 architecture to complement the x86-64 architecture of my PC. Audio processing doesn't really require much processing power and networking allows one type of computer architecture to control another, whilst having separate hardware, separate power and whatever software is best for the job of the particular computer.
So with an x86-64 5V board and an ARM64 5V board, I can optimize both and compare the software used. So I will likely add an UP Squared into the mix. The main PC power supply does not have to be compromised for the sake of audio; I will give the audio its own isolated board to solve that issue. I may find that ARM64 audio software has better offerings available than x86-64, as streamers are often not x86-based.
|
|
CageyH
Senior Member Joined: 30 Apr 2012 Location: Toulouse, Franc Status: Offline Points: 1678 |
Post Options
Thanks(0)
|
But a lot of streamers are x86-64 based. https://Innuos.com/zen-mk3/
|
|
Kevin
European loan coordinator, based near Toulouse, France. |
|
BackinBlack
Senior Member Joined: 05 Feb 2012 Location: Hinton, N'hants Status: Offline Points: 2020 |
Post Options
Thanks(0)
|
Ash, have you played/streamed anything through your Pi yet? Is so, I'd be interested to learn your initial thoughts.
Ian
|
|
Just listen, if it sounds good to you, enjoy it.
|
|
discrete badger
Senior Member Joined: 14 Jul 2009 Status: Offline Points: 479 |
Post Options
Thanks(1)
|
There's another important moving part to consider - how the audio driver of the OS is set up. Bear in mind that default setup is usually convenience, and working with a wide range of hardware, not sound quality.
For instance, most x86 Linux distros default to resampling everything to 44.1/16, because that's the lowest common denominator of audio hardware. The resampling method used is often a low-cpu-drain one, not a high quality one. This does not mean that x86 or Linux is good or bad, just that it needs to be set up correctly for the desired use. Whereas the Raspbian Linux on ARM, which ships as part of Moode, is set up correctly for high audio quality, because Moode is a near-turnkey solution which has been tailored carefully for a specific application.
|
|
Ash
Senior Member Joined: 18 Mar 2013 Location: Dorset Status: Offline Points: 4334 |
Post Options
Thanks(1)
|
Just waiting for a USB-microSD adapter to come for flashing OS then I will commence with some experiments and comparisons. Getting some more Pi stuff for Christmas so I will do more around about that time too. Getting a Li-Po battery HAT for the 3B+ to see if computer power types makes any difference to sound quality. If there is no discernable advantage to a 5V computer over a 12V one, I won't bother getting a 5V x86 board like the UP Squared and will stick with the UP Xtreme i11 directly.
Some of this power stuff will be total nonsense but I have to rule it out by actually trying it.
|
|
Post Reply | Page <1 1011121314 28> |
Tweet
|
Forum Jump | Forum Permissions You cannot post new topics in this forum You cannot reply to topics in this forum You cannot delete your posts in this forum You cannot edit your posts in this forum You cannot create polls in this forum You cannot vote in polls in this forum |