View RSS Feed


How To Explore Computer Audiophilia.

Rate this Entry
This was actually a reply to a forum thread, but I thought it should have its own blog entry. These are a list of things to try to explore for better sound within the context of computer audiophilia and documents some findings along my journey.

1. Studying and researching computer audiophilia with the following viewpoints:

a. There must be a way to make digital sound at least as good as analogue

b. Just as we used to do with Hi-Fi analogue systems, we should optimise the digital playback chain

c. There's more to great music reproduction than just frequency response, which is a solved problem. Things like phase, attack transient smearing, filter ringing, etc...

d. If you can't afford to buy it, build it yourself

This costs nothing except time, and allows me to move forward and get great results while others are still debating 'bits-are-bits', which format is better, cables make no difference, and while trolls are thinking posting here is a kind of popularity contest and that being able to research is a kind of disease.

2. Room diagnosis with R.E.W.

This is free as well with just a forum registration, and you can't really hear your gear until you have tamed your room response, so it's a good first thing to do.

So if you think you're getting great sound but haven't yet done that, you may be surprised when you come to it and in the meantime, you aren't yet hearing your system.

3. Acoustic Room treatment with DIY acoustic panels and bass traps.

Not very expensive: looked for glass fiber/wool at the hardware store (they were already cut in useable panel sizes), and a suitable textile material from curtains. Also took some rolled up glass/fiber material that I placed in the corners of the room for bass standing wave trapping.

At this point, it made a huge change already.

I recommend doing the organic, acoustic treatment and solely this if you can. The goal isn't to make your room completely dead, but tame the biggest peaks and troughs of the room response.

It's only in case of very stubborn or difficult to tame peaks and troughs that you can consider DSD for room treatment as this also has its own sonic signature and will colour the sound to some extent.

People who head for the easy DSP EQ right away aren't getting as good results than those with a more natural/organic treatment process.

4. Building stands for my Studio Monitors.

Just the cost of wood and wood glue and wood screws, and some time figuring our how to do the geometry properly (I'm more of a Computer Science Engineering guy than a manual work guy).

This allowed me to move the monitors off the desktop (which was vibrating with music) to the stands, and hence dissipate any remaining cabinet vibration down to the stand.

Very, very large improvement in sound definition, more clarity in the high frequency range.

I since obtained Totem Mite bookshelf speakers which are now on the stands with fantastic results.

Bookshelf speakers on stands is one of the best things you can do, rather than get huge speakers with a lot of cabinet resonance (Vivid might be one of the rare exceptions, Dickie was working in sound reinforcement before doing speakers, so he brought a lot of his expertise from there to speakers).

In other words, don't think that 'bookshelf' means you have to put them in yours - that's the worst you could do. Instead, build or buy stands for them.

5. Trying DSD

That has made a tremendous different for me, with regards to 1 (a) above, even when I was merely listening with Audirvana -> PCM soundcard.

6. Researching and trying audiophile players, and getting Audirvana+ for playback instead of iTunes.

Damien is clever and on top of that, has a great ear for good sound.

Doing this is paradigm-shifting.

7. Researching about DSD vs PCM and getting a native DSD128 DAC, the iFi iDSD Nano

Finally, native DSD at home for around $200, and a huge, huge sound that is worth 10+ times more that it goes for! Total game-changer by Thorsten's team at AMR/iFi - it changed the industry.

8. Studying what these guys came up with: John Swenson, Thorsten Loesch, Bruno Putzeys, Dieter (Trinity DAC), Lukasz Fikus (Lampizator), Jon Risch, Miska, Hiroyuki Yokota, PeterSt, sbgk, Damien Plisson, Caelin Gabriel, Charles Hansen, Ted Smith, David Berning, Townshend, del Sordo, Tony Lauck, Kimura, Kusunoki, Lawrence Dickie, George at Tubelabs, Y.N. (cPlay), and many more.

Paying attention to credible people in this forum on top of those already mentioned, like Chris, Jud, Superbad Alex, Sandyk Alex (you may find some of his claims outlandish, but once you step in the realm of Yokota as well as the DRAM thread and do some tests, they aren't at all), EuroDriver, Geoffrey Armstrong, jabbr, tranz, etc... The people who know what they're talking about, have actually tested things and freely share information for the greater enjoyment of the community and this is info you can use to test things yourself. Just Superbad Alex and John S. have made tons of free contributions that are worth several times their new product price IMO.

Costs nothing except dogged determination and intellectual effort, and gives me a thorough understanding of how the field clicks and where it's going, while the idiots clamor you're here to increase your post count and say you know how to use Google as if it's something to be ashamed of. Eff' em: their mediocre thinking will only bring them mediocre sound unless it's handed to them as a product. Conclusion: waste no time with the idiots and trolls and instead read what the above-mentioned guys say and try to understand why and/or apply in a DIY build.

9. Critical listening sessions for comparisons between:

- lossy and lossless
- lossless compressed vs uncompressed
- PCM and DSD
- lower-rate DSD vs higher rate DSD
- different HQ Player setups (computer to DAC direct vs client-server mode, different filter and modulators)
- realtime PCM-to-DSD vs offline up-conversion
- different DAC filter settings
- different Audirvana+ filter settings (esp. regarding phase)
- different Audirvana up-sampling settings (Redbook to DXD, no over-sampling, power of 2, etc...)
- Solid State amplification vs Tube amplification
- Generic USB cable vs DIY USB cable
- Default power cable vs DIY power cable
- Default equipment support vs DIY equipment vibration isolation
- Default room response vs DIY room treatment response
- Default AC power vs DIY AC filtered power
- Default equipment system connection vs DIY System Chassis Grounding
- Default equipment system connection vs DIY Signal cleanup (some say 'Grounding' here, but it isn't that in reality)

This costs just time, but is crucial in honing my listening skills as well as filtering out what works from what doesn't (while the others are wasting time debating double blind, blind, measurements, ABX, etc... which only matter if the effects are too close to discern convincingly and if you're trying to prove something for science), give me a good reference for tweaks/changes, makes me know my system inside out and finally allowed to find the sweet spot of my system. The sweet spot is DSD128 and above. Hence:

10. Re-ripping and/or converting all my library in uncompressed (AIFF)

Best foundation for high-quality reproduction (you could choose WAV on a non Apple system too).

11. Up-conversion of my tracks to DSD128/DSD256 for critical listening and musicophilia

This is sublime. If you think about it, I am doing a smaller version (with less features and a few different steps missing) of what the PS Audio DirectStream does with PCM.

Last night's example: my gf asks to listen to Billy Ocean, I play 'Loverboy' first in PCM, then in offline up-converted to DSD128. At the beginning of the DSD version she says she finds no difference (while I already hear much clearer sounds, what I call 'texture', i.e. the individual constituents of the sounds), then when the vocals hit, she agrees there is a large difference, and then when a percussion hit, we both simultaneously point to the speaker where we heard the sound, then look at each other quizzically and smile.

12. Taming Equipment vibration:

This, with Room Acoustic and Room treatment is a fundamental issue you need to take care of, and here, it's not just vibration from outside into your equipment, but also vibration internal to the equipment, and from equipment to another as well.

This can make a large difference, especially when it comes to bass response, DIY and materials do not cost a lot.

I designed a way to tame vibrations for large and small speakers, the description of which is in the related thread.

13. AC filtering, isolation, conditioning, linear power, balanced power

[NB: safety first if you tinker with this yourself]

This is a third (not necessarily in that order) fundamental issue you need to take care of. Not only can noise enter your system from outside, but it can also be from inside in two ways: other non-audio equipment, but indeed, shockingly (ah ah ah) from your own audio equipment.

With my DIY AC filter box this brought a lot of presence, clarity and detection of 'new' sounds and performances from albums I know very well.

I have more experiments along the way with this, including implementing a design I invented but which is the subject of quite recent scientific research, so getting the proper material for DIY may be difficult for now.

Some designs are described in the thread.

Read about great products from manufacturers and what they use in them. Many use Plitron transformers and there are reasons for that.


14. Grounding, eliminating Ground Loops

This is also fundamental and you won't know what you're missing until you do this properly. Here again, do not listen to nay-sayers who tell you it shouldn't work or how this is supposed to work: the reality is that in heterogeneous systems you are likely to end up with (many different manufactures and cable constructions and different grounding schemes, coupled with dirty AC power), there are a lot of issues.

Moreover, and you should find this flabbergasting as I did: the subject is quite complex, and many manufacturers have seriously compromised the engineering of their equipment by cutting corners. Some of them make the issues worse!

By all means, check the related thread (it's the same as the AC one), and read about the Tripoint Troy devices, as well as the Entreq Tellus ones and what they do and the results people are getting with them then go and do the same.

Many learned people will give you an incomplete or outdated theoretical model of how things should work here. They are WRONG, as they forget to include the modern situation of being surrounded by cellular EMI/RFI.

In fact, many just ignore the issues surrounding EMI/RFI completely when it comes to computer audiophilia (and cables...).

Read everything you can get about Electromagnetic Compatibility. In particular, read everything from Ott, Morrison, Keith Armstrong, Jim Brown, Neil Muncy, Bill Whitlock, Philip Giddings. Apart from Ott, Morrison, and Giddings, many of these references came from Speedskater, but who gets the basics wrong when he forgets to add that currents want to loop back to their source distributed preferentially through lowest impedance paths. Additionally, IMO he also gets it wrong when he talks about 'at these low frequencies it doesn't matter' where I specifically talked of high frequencies and make no assumption on their source.

Read, then go and do it, or if you can, just go do it. The results are immense: tall soundstage, clarity, presence, much of what you get with cleaner AC power.

15. DIY USB cable

This was mind-blowing: I didn't expect any results at all here, and in fact expected the worse: drop-outs or total disconnection or no detection of the DAC.

My cable sounds stellar. This proved to me that a 'digital' cable can have a bearing on SQ. The difference is big especially when it comes to attack transients, soundstage, reverb tails, dynamics. This cable makes you want to get up and dance.

It isn't the kind of difference that is close so as to make you even think 'it's close I should perhaps do a DBT'. It is a 'night-and-day' difference.

This is all about shielding and geometry. I have posted the design on a thread somewhere here. Doesn't cost more than some time, some shielding and a normal USB cable as donor.

16. DIY Power Cable

If there was a second thing expected to make no difference just like 'digital' cables, it sure was power cords. So, I jumped in and built my own power cable for the amplifier. Right, right, here's another myth busted from the get-go: I have a much larger detailed audio range now with my DIY power cable.

Here, it's about shielding again, but also using solid-core wire. Very low-cost but high return!

17. DIY Single-Ended Triode Tube Amps

While researching Tube amps for my build, I settled for the Single-Ended Triode architecture which isn't perfect technically but is revered for SQ. Further research and study brought me to the Tubelab section on diyaudio, and to George's website. I liked what he wrote and his methodology, ordered the PCB (looks stunning), researched transformers and ordered 3 large Edcor ones, researched Tubes and got a whole Russian set (Electro-Harmonix KT-88 and others).

It sounds sublime, totally beats my SS amp when it comes to fast attack transients. Needless to say what this does to: soundstage, presence, rhythm/timing, etc...

The cost here was mainly in the transformers, the other parts do not cost a lot, even the tubes aren't that expensive. In all, it probably cost me in the range of $300-375.

18. Learning about the inner workings of computer audiophile setups and the preponderance of gray areas and EMI/RMI

The DRAM bit-flipping thread and the papers I researched with as well as the great ensuing conversations with John Swenson, Miska, PeterSt, and sbgk had a lot to do with furthering my knowledge of what really happening in digital land.

Digital land has a lot of gray areas which have a lot to do in reality with analogue phenomena and the issues with them.

The gray areas are related to A/D and D/A and the EMI/RFI of digital equipment along the processing chain. It's near the little magnets in HDDs, around the caps in your DRAM, the analogue signals in your USB cable, in the memory pathway you use in your audio software...

This brought me to start work on a prototype DIY audiophile player which sounds great out of the box, but whose next iteration should be mind-blowing if I get it right.

In the meantime, the clueless are still stuck in the 'bits-are-bits' paradigm, so far removed from the challenges we face in reality.

Bit-perfection is necessary but not sufficient for great SQ.

Reading that thread costs time but that's it.

19. Continuing to read, research, learn and build and optimise along the whole digital playback chain

This costs just time, and allows for fruitful exchanges with other knowledgeable people intent on furthering the field, and also allows you to filter out the trolls (like Daudio, 4est, kumakuma, etc... who haven't done the tenth of my efforts to date) to spend more time with the former.

I have already designed my ultimate setup, and one day I will build it. While it may take me years to get there, I already know what is going to be in it and it will require me to build my own speakers among a few other things.
Tags: audiophile