 How to get good sound quality audio from your computer? This is literally the question that was asked to me several times. In this last video of 2018 I will try to give you answers. The hard disk is the ideal storage medium for digital information. That is why it is used in almost all computers. Ok, today often SSDs are used as system drives and sometimes even as storage medium. Therefore in this video let's consider hard disk and SSDs being functionally identical. For they are that for music. Far more important is that both are far superior to the optical media like CD, DVD and Blu-ray that are also used for audio storage. Unfortunately, normal computers, especially the powerful ones, generate a lot of electronic noise. You can't hear but can and normally will have great impact on the sound quality. The focus of this video is to see what the impact is and how to minimize it. So the goal is not to prove it would work, for any computer can reproduce music. NIDO will uncover what software to use or how to prevent the operating system from ruining the sound by DSP actions. Those will have to wait for other videos. This video is about hardware choices. Computers basically work the same as a cuckoo clock. A mechanism called escapement that is driven by a mechanical force like a cornspring or weight lets a cock step forward one tooth at a time. See escapement in Wikipedia for details. The mechanism is designed so it lets the second hand go around 360 degrees in 60 steps. The cock that moves the big hand is programmed so it moves 360 degrees in one hour and the cock that moves the small hand is programmed so that it goes around in 12 hours. The programming is done by choosing relative ratios of a number of cocks. Another set of cocks moves the cuckoo every 15 minutes and operates the bellow that blows the whistle for the cuckoo sound. The ticking of the clock caused by the escapement getting caught between teeth is a by-product and was no design goal. If the clock is not carefully designed it might be slowed down by the mechanism that moves the cuckoo and operates the bellow. A computer works basically the same. Only at lightning speed and it contains numerous clocks, for instance for the processor, for the video card, for the USB ports, for the Thunderbolt port, for the audio card and so on. And all these clocks usually influence each other to some degree, like the cuckoo mechanism influences the time mechanism in the cuckoo clock. Where the ticking sound of the clock might be experienced as relaxing, the electronic noises the computer generates can't even be heard by us. We do hear the mechanical noise of the hard disk and the wind noise of the fan but not the noise that is passed over the cables to our precious audio equipment. And we are not aware that the audio clock that sends out the audio data, supposedly at very precise intervals, is not as precise as good sound quality requires. Please do understand that you still hear uninterrupted audio in most cases and this makes people say that it's ok. But if audio quality is your goal, the noise should be as low as possible as where the clock should have extreme precision. Placing the sound card inside the computer, close to all the other clocks that radiate high frequency noise, will never lead to good sound. So the least you can do is use an external sound card. In the audio technique an external sound card is called a digital-to-analog converter, abbreviated to DAC. There are several ways to connect a DAC to the computer. SPDIF, the RCA digital connection, Toslink, the optical connector or USB. SPDIF and Toslink are in fact both the same signal but one is transported using voltage while the other uses light. In both cases the signal is sent at the speed the computer dictates, which is the nominal speed including the variations due to the noises inside the computer. Therefore connecting a DAC to a computer over SPDIF or Toslink isn't preferred. Using USB 2 is far better, especially if both the operating system of the computer and the DAC support audio class too. This is the protocol that supports sampling rates higher than 96 kHz. Macintosh and Linux computers support this already for years. For Windows a special driver needs to be installed. Often the manufacturer of the DAC offers a driver for download or on a CD-ROM. The advantage of USB 2 audio class too is that the signal is sent as packets to the DAC and are then clocked inside the DAC. Still there are things that can go wrong. If the electrical noise from the computer travels over the USB cable to the DAC that can distort the digital audio signal to some degree, again leading to timing variations by audio technicians called Jitter. See my video connecting your DAC number 2 how digital can go wrong, link in the video description. Now if you do have your computer connected to your DAC and are not completely happy with the result, there are all kinds of boxes that can or say they can improve the digital signal. The ones I have tried varied in result from sometimes a slight improvement to a steady and decent improvement. The result can also depend on the combination of equipment. But in general it's like fixing a dent in your fender as a result of failing brakes without fixing the brakes. Yes, the fender looks great again but I would have gone for the brakes first. If you systematically want to tackle the problem, you start by using asynchronous data transport. So USB 2 audio class too or Ethernet. These send data in packets that are clocked in the DAC. SPDIV and Toslink and AES-EBU are so called isogranous data transport, meaning that the sending device dictates the clock and bit after bit is sent regardless the conditions at the receiving end. Since in all three formats the clock signal is encoded in the data stream, all things can go wrong and you need really good designed equipment to achieve a very good sound quality. The only interfacing protocol I can advise is I2S, written like this. Here the data and the clock are sent separate from each other. If all is done well, this is a safe and good sounding way too. But it is found only on a limited number of devices and there is no standard connector. It could be over two B and C cables, a UTP cable or an HDMI cable. And the pin layout of the HDMI connector varies between brands. There are all kinds of accessories for computers to make them audio file. Like special low noise power supplies and output boards. SOTM offers a card that outputs cleaner USB signals, especially if an external audio file power supply is used for that card. And there will be others. Since cables for digital audio can't be too long, I would stay within 1.5 meters or 6 feet. The acoustical noise of the computer should be tackled too. If you do all these things you often end up spending a few thousand euros on an audio file computer. That just might be a better way. I said earlier that the more powerful a computer is, the more electrical and also, often, acoustical noise it potentially will make. Therefore a good solution would be to use a very simple light computer for then there is no need for forced cooling while the electrical noise will be far lower too. An audio file great power supply that can power such a light device will also be cheaper. And just passing on audio bits is a very light application for computers. The precision required is a matter of choosing the right audio file add-ons. The disadvantage is that the cataloging will be rather limited. So you end up with a slow and or simple browsing experience. Why don't we use a decent quick computer to do the cataloging and storing if you like? And let a very simple computer that is powered from a small audio file power supply do the play out to the DAC. On the full blown computer you install bit perfect audio playback software that is able to send the audio over the home network to the light computer that only functions as a play out station. The latter is depending on the software called Network Bridge, Renderer, Endpoint and so on. You can start with the Raspberry Pi with 5 volt DC, iFi iPower power supply and a simple SPDIF or USB output card. Don't use the USB port of the Raspberry Pi for that performs extremely bad. The audio output on the Pi is even worse. If you use an audio grade add-on board, they start from 25 euros. From there you can go upward as you can see in my Raspberry Pi for audio playlist. Links in the video description and at the end of this video. If your demands are or become higher, you could go for a network bridge like those by Aurelic, Sonora, SOtM, Metrum, Acoustics and others. See my playlist Network Bridges. And the most expensive network bridge plus a normal desktop or laptop computer is cheaper than an audio file computer that is silenced both acoustically and electric. If you already have a functioning setup and don't want to change that but still want to improve the sound, check out my playlist USB Conditioners. If you got interested in using a network bridge or DNA renderer, you could start with the Raspberry Pi if you are handy with computers or go for the SOtM-SMS-200, the base model. Just try it and if you want more, search more upmarket. My playlist Network Bridges might help you there. Of course the software you have to use does matter too. But I leave that for a later video. In the meantime, watch my series of Audio Hygiene for all these things matter to a certain degree depending on the other equipment you use and how critical you are as a listener. And keep following this channel by subscribing or follow me on the social media. I will let you know when new videos are released. If you liked this video, give it a thumbs up and please consider supporting the channel through Patreon or Paypal. Many thanks to all that already support this channel financially. It keeps me independent and thus trustworthy. See the links in the comments below this video on YouTube if you want to join in. Help me to help even more people enjoy music at home by telling your friends on the web about this channel. I am Hans Beekhuyzen, thank you for watching and see you in the next show or on theHBproject.com. And whatever you do, enjoy the music.