DIY USB Cable

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
And who knows, I have been a victim to imagining audio improvements that turned out to be a flight of fantasy.

This is true for every one of us, and in areas well beyond audio. Your brain is wired to err on the side of false positives, and the easiest person in the world to fool is yourself.

Before you start in on setting up group tests (those are EXTREMELY difficult to do correctly, and the vast majority of engineers don't really have any competence in that area), try some testing yourself to see if you think it's worthwhile moving on to the next step. Some basics can be found here, in the "Testing, One, Two, Three..." article:

Online Articles
 
wlowes said:
I'll disengage from this thread until I have something credible to report.
You need to understand that the level of proof you require in order to be taken seriously is similar to the level of proof required from someone who is seeing fairies at the bottom of his garden. In both cases a claim is being made which violates known science. "My friend can see them too" might not be sufficient evidence for sceptics. This might surprise you, as it might surprise someone who has always believed in fairies.

To put it another way, you don't realise just how wrong your ideas are. They approach the category of 'not even wrong' used by some physicists to decribe ideas not worth serious thought, not worth debunking because they are so obviously wrong to anyone who knows anything about the subject.

I wish we didn't have to be so blunt, but so far you don't seem to be learning anything. It is sad that it is so relatively rare for someone to admit their mistake and begin learning, but it does still happen from time to time. There was a thread on here a few months ago where someone vehemently maintained his silly ideas at first but then backed down and started learning, thanking us for putting him straight.
 
Rojo, wloves etc.:

Thanks for posting your impressions as well as describing what you did to achieve them.

My use of the internet wrt. audio improvement is focused on and inspired by people like you who share means to alter sound reproduction which are (relatively) easy to check out for myself as well as being within financial range.

Your enthusiasm is enough for me; if I don´t hear the result as an improvement, I can revert or try something else (provided I don´t damage equipment but I trust myself enough to be aware of the risk, seeking further knowledge when in doubt and fully taking responsability for my own possible mistakes).

I haven´t close-read this thread yet but it seems there are 3 categories of posts:

- the practical (yours)
- the critical
- the sarcastic-stupid

I´ll ignore the last category (reminding me a bit of the "noise" I still need to remove from my digital audio chain before it matches a good analogue one)

- I´ll read through the second one to see if there´s any knowledge to be gained that I can use practically - and disregard the arguments for not trying out your tweaks, unless they point to a possible risk of damaging equipment

- and focus on the first one; after all it´s the only one where the human ear and perception has been involved in judging the actual alterations of usb-transfer.

I think me posting that I hear something will not be considered by current posters of credible evidence of something worth pursuing. And who knows, I have been a victim to imagining audio improvements that turned out to be a flight of fantasy.

see above; it´s fairly easy for anyone to explore if their own experience matches or differ from yours....

My plan is as follows. I live in Toronto and have hosted DIY meets at my business location. From that group there is a moderator of this site who appears to have credibility with the hard core designers at DIYAUDIO based on his work with shunt regs. I have invited him over to my home before to audition my system, and has offered to bring a scope and check out what is going on. He seemed impressed by the results I have achieved and fwiw I believe he has a well tuned ear. I'll see if he is interested in coming and at least be the listener in the A B test of a high quality commercial cable vs the one I claim bests it. Perhaps he can then suggest running a .WAV track with a clean test signal to see what is happening with the soup mix of impedance mismatches.

1. hand real-life, shared experiences are always welcome :)

This exercise will be quite helpful. I am on a path to take all of these findings and make a clean build of my final music server. My plan involves a 3d layout aimed at ultra short signal path for both usb and I2S while isolating the digital from the analogue. Hate to go to all that trouble based on an incorrect flight of fantasy.

Being on the somewhat same path (I think, using Mpdpup on an Alix board usb-connected to an usb-I2s QNKTC Dac) I´d be extremely interested in your findings.

Again, thanks for sharing your experiences.
 
- I´ll read through the second one to see if there´s any knowledge to be gained that I can use practically - and disregard the arguments for not trying out your tweaks, unless they point to a possible risk of damaging equipment

These tweaks will probably cause signal integrity issues, and unless you have the equipment to look at the interface (a network analyser or scope at least) then you have no idea if anything has improved.
How is someone with no electronics knowledge and no understanding of digital signal transmission (or how USB audio is transmitted) going to improve the interface that has been developed by so many engineers.
And despite numerous request to what the mechanisms may be that are causing the sound to change we have just had silly replies, you cant improve anything without at least a bit of an idea as to the causes and effects, that how development works.
Audiophoolery is becoming like alchemy, you just need to mix four basic ingredients to improve sound:
Silver cables.
Teflon dielectric.
A total lack on understanding regarding what you are doing.
A wife in the kitchen (not an audiophile) who will instantly confirm the improvement your changes have made.


Just for your interest do a poll of how many improvements lift a veil etc (ie great, night and day) and how many tweaks don't improve of make things worse.

This madness does not help Audio reproduction improvements.
 
The future of the snake oil business is in serious jeopardy with all you 'non believers' around.

If you need proof, just look at what happened to the whale oil trade when some dam fool came up with the electric light!

There could be a serious demand for 1/2" long cables out there........somewhere......cant think where, but there just might be!
 
Last edited:
Hi marce, thanks for your reply

These tweaks will probably cause signal integrity issues, and unless you have the equipment to look at the interface (a network analyser or scope at least) then you have no idea if anything has improved.

Sorry, I have to disagree here (I think you are forgetting my ears :)

All changes in my music reproduction chain that are staying are there because they are improving my listening experience - mostly by removing obvious (to me) perceived errors in said music reproduction (voice-reproduction, especially "s" and "t" sounds, is for me where the chain most clearly reveals itself as sounding "natural" or artificial).

("natural" in quotation-marks because it´s not there yet (and I have my doubts any reproduction chain ever will be able to fully reach up to the actual musical invent - to the point where I won´t be able to tell the difference)).

Hearing and perception, listening ability - probably differs from person to person as well as may be variable factors for the individual, prone to improvement or degration.

Other area are the "scientifically", possible measurable causes of impact on music/sound reproduction.

I searched a bit further and found a thread which pointed to which elements in usb-cabling could impact the sound. Not because causing data-errors but because causing timing-errors (jitter). Could maybe add a dimension to your term "signal integrity issues"?

USB cable MATTERS! - Page 2

How is someone with no electronics knowledge and no understanding of digital signal transmission (or how USB audio is transmitted) going to improve the interface that has been developed by so many engineers.
Could it be (no offence) engineers are having different priorities and may not (yet) know everything there is to know about data-transfer and sonic impact?

And despite numerous request to what the mechanisms may be that are causing the sound to change we have just had silly replies, you cant improve anything without at least a bit of an idea as to the causes and effects, that how development works.
Not silly replies, just admitting not knowing the causes for the experienced differences/"improvements" (that´s not a crime, is it)?

Finding/exploring "mechanisms that are causing the sound to change" requires effort on several levels; you and others effort amounts to requesting knowledge from a very limited source about mechanisms, causes and effects that you deny excist....

Of course you can improve (as well as the opposite) by chance/accident not knowing causes and effects. Knowing causes and effects comes after experiment in human history.

This madness does not help Audio reproduction improvements.
Though madness is a hard word I totally agree (but as you know my perspective on what does or doesn´t help Audio reproduction improvements differs somewhat from yours :)

Regards,
 
Last edited:
Finnogangela said:
Not because causing data-errors but because causing timing-errors (jitter).
USB was designed to transfer data. Anyone using it to transfer timing too has already decided to put up with audio problems. Transferring data and data alone, as it was intended to do, USB will be fine using any USB-compliant cable. Any non-compliant cable can only make things worse, never better - especially if the cable is being unreasonably expected to handle timing too. These are simply facts. Unpalatable facts, maybe, for DIY tinkerers and cable merchants but facts they remain.
 
I searched a bit further and found a thread which pointed to which elements in usb-cabling could impact the sound. Not because causing data-errors but because causing timing-errors (jitter). Could maybe add a dimension to your term "signal integrity issues"?

Do you know what signal integrity is and what can effect it! Do you think when we are doing DDR3 interfaces and other high speed designs we do not know what jitter is and the causes as well as all the other gremlins that want to disrupt the signal transmission, that's why we have numerous engineering guidelines and reference books such as High Speed Digital Design, by Howard Johnson.
 
@Finnogangela : the problem with your line of reasoning is that you assume that the "problems" associated with usb cables can be solved by shooting blindly in the dark.

Differences in between usb cables can indeed be measured ; there's something about the influence of usb cables on jitter measurements in the hifi news from january 2011, in which one "cheap" usb cables does better than some audiophile ones.

BUT:

- those differences are arguably well under audibility and certainly way less significant than other factors in the usb chain (the USB protocol used, the quality of your usb ports, of the receiver, etc).
- even if you insist on a "perfect" solution, you won't get there by building cables with worse geometry than the common usb cables.
- you need the tools to measure what you did if you want to claim improvment.

There are "engineers approved" ways to a "perfect" usb: decent asynchronous operation, data buffering , galvanic isolation, low noise PS for the receiver are a few examples. Plenty of threads about that around here.
 
I find USB audio quite decent but I use an external clock between PC and DAC. This proved good enough for me to stop using a CD player. I doubt i'll go back to CD now.

I bought a decent up-grade USB cable from our local TESCO supermarket while food shopping. It sounds every bit as good as the one I swiped from the printer but the missis sometimes needed the printer while I was listening to music.

An upgrade USB cable has proven benefits, it improved domestic relations for less than £10.

Remember, you heard it here first folks :D
 
Some info for those interested 1 of 2

I have exchanged some email with potential ABX testers. We will likely do it after the holidays at one of our regular DIY events. Possible there is no interest and we'll never do it.

I have absolutely no interest in fame or fortune from this testing. I simply like to get the most out of my personal rig. When I see something that works for me I share it in case it leads to someone else gaining a similar pleasure real or imagined.

I did read the info on good testing provided by Sy. Good article, and thank you. I will use it.

I received a few PM asking for details for people interested but not wanting to enter the fray of this thread so I felt it important that I at least go back and do some comparisons of cables myself to see if I still believe this. I have been using a short cable for some years now, so who knows?

This thread also captured my interest to know if the warnings about having the PC board close to the DAC would seriously affect the sound.

The four configurations being compared are:
1. A standard usb cable. Pulled it off the printer
2. The short cable I have been using for the past 3 years. 2 twisted pair using solid copper core teflon dielectric. Reasonable care taken to get the D lines the same length. Separated a bit from the +5 Gnd using packing tape.
3. The silver cable I reported earlier which kicked off all the "excitement". The D lines are 3 9s 24guage silver. Identical length. The +5 & Gnd are just 24g copper in teflon. This thing was thrown together as a rough test. No twisting of the +5 Gnd and while I planned to have the D runs carefully placed parallel and well away from the power, I failed and I agree it's a mess.
4. The end goal is to try very short. So less than one inch. Silver on the D and copper on the +5 Gnd. All parallel. All identical length. No sharp bends with bend radius of about 1/3".

What ever effect I hear, real or imagined may be totally restricted to my system so I will describe it for the record. It is a minimalist headless PC (ALIX router board) running stripped down linux (mpdpup) from a memory card. nothing on the board but rj45 and usb. NAS holds the musiclib and samsun phone is the wifi client. It feeds via asynd usb2 to a WAVEIO usb board. This then feeds a 1541A dac with a 6n2p buffer ouput. Lighspeed attenuator, silver interconnects to 110W OTL tube amps. Full range speakers with minimalist xover. It is pretty well resolved. Even a high quality resistor in the signal path anywhere is obvious and able to impart a unique sonic signature.
 

Attachments

  • Good better great.jpg
    Good better great.jpg
    334.9 KB · Views: 302
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.