beautox

Marce brought this up, calling it BS, so I was simply defending it. He claimed cable with any asymmetry would not work, and I went on to demonstrate this was itself BS (by a factor of 100). I think the useful information that I've provided is that marce and mansr don't know as much about USB as they think they do. The only industry affiliation that the original poster has is that he bought a cable, full price and is very happy with it. Do you think of yourself as being affiliated with every manufacturer you've ever bought a product from?? If you read what I said I was talking about prototype/test cables and also the 20cm-extra-dataline I made to show that in the real world, there's more tolerance than the spec-jockeys would have you believe. But just a minute : Why, if the data is being transmitted faithfully, does a bit of extra noise matter? This is a digital system. According to the bits-are-bits brigade, they should sound the same. And if they don't, then it seems that USB cables can sound different. Which is what I reckon. btw, the 20cm-extra cable sounded awful, worse than a regular printer cable. It had a very flat presentation with a vague stereo image. imho

Yeah I can understand that it's kind of upsetting to be shown up. Marce : confirmation bias works both ways. You will never hear a difference because you don't believe they exist.

..so USB cables do sound different? Or not?

Marce : I never said I would explain what I meant. In fact I said the opposite, that it was a trade secret. What I was doing was using your assumption about data line length, and showing that it was wildly incorrect. And positing from this that there might be other things, especially relating to USB, that you are also wildly incorrect about (eg USB cables surely all sound the same, LOL)

Marce: Blah blah blah you just can't accept you don't know as much as you think you do. You can't get around the simple fact it works, and you said it simply can't. 200mm is 100x what you guys reckon would work. Sure it's likely to have bad noise, horrible eye, but the point is that it still works. And works with more than one device I tried with. Here's an update : it almost works at 30cm - connects and plays a couple seconds. But 20cm is fine. Mansr: I can't help thinking you're changing your story, from "that can't work" to "your usb receiver is more tolerant". Yes, more tolerant that you expected, right? Thanks, but I know how specs work. But I am making a cable that sounds good and works with a wide range of DACs. It's an entirely valid compromise to make, to trade spec compliance against other factors. People can return if it doesn't work, but so far that has not happened. I can't keep track of every device that customers use, but the majority of devices use a fairly small number of chipsets. I do know of several dozen DACs that it works fine with (compared to none that it doesn't). The point I am trying to make is that your theoretical understanding (2mm) is at least two orders of magnitude out from the real world. Why do you think that is? Simply "It's not forbidden for an out of spec cable to work." - what, two orders of magnitude out of spec? Well guess what...that's not uncommon. For bonus points : how would you objectivists think that my "20cm extra" rigged USB cable sounds vs a regular printer cable (vs Black magic?). And why? Surely if the digital data is getting there (evidenced by no clicks, pops, hangups) it should be "perfect sound forever", right? I'll give my impressions later which you can disregard in due course.

Are you saying that I am not telling the truth with this cable I just knocked up? Try it, it works. Just proves to me that you don't know as much about USB in the real world as you think you do. That cable has 200mm of extra length. And I'm listening to music thru it right now. Honestly, your comprehension is terrible (or maybe just being disingenuous), You quote this line from which you seem to infer: But what I said was I did not say anything about production cables. But frankly, who cares if the cable is in spec, as long as it works. OK maybe folks like speccists/objectivists might but they use printer cables anyway. Seems to me that Marce and now you have been shown to have a theoretical understanding that doesn't map well onto reality when it comes to USB (and who knows what else??). And this is laughable: Right, so it's either length or new physics. LOL. https://en.wikipedia.org/wiki/False_dilemma

Is that right? First, let me state that length it not the asymmetry I use in Black Magic. But let's look at your claim: Let me refer you to this thing called "reality" My own BS detector went off strongly when I read your post - claiming that the two USB data lines have to be exactly the same length. So here is what I just did - cut the end off a 1 meter printer cable, soldered a plug on the other end so I could alter the length of one of the data lines. It was a 1-meter cable, so I first tried an extra 20cm, so that's 20% longer. You'll notice I added length to the D+ line (white wire). It works perfectly. Handshakes fine, as you'd expect. But also then plays DSD512 without a hitch, obviously running USB2 as it's about 50Mbit/s) . If you don't believe me, make your own and try. See the attached pic - it's not pretty but it works (and now it's got me curious how long you can go in practice..I might report back on that..) Why do you think I guessed this correctly? Well I think it's got something to do with the more-than-100 USB cables I made during development, to probe what you can and cannot get away with in practice. I think this nicely illustrates something I said earlier in this thread: Perhaps you should have a more open mind, you might find that we are not all BS-ing phools as you call us.. (both terms used in this thread)

I don't know what to say. I thought "asymmetrical" was a perfectly valid engineering term. While we at it, please explain why you think transmission line theory and the USB spec forbid whatever it is you think? Here's a problem I'm not sure how to solve: I spend a lot of time, effort and money developing innovative products and selling them. When it comes to selling, there's always pressure to reveal exactly how it works. Now I really don't want to do this as I see my designs as my intellectual property. However you can't just say nothing so I can understand the temptation to write misleading or just plain junk marketing descriptions. So the approach I decided to take was (a) to be upfront that I am not going to explain exactly how the designs work and (b) make true statements about the products, but leave out crucial parts. I make no bones about this. btw, in one sense, all USB cables are asymmetrical, unlike say RCA cables. If you can't figure this out, award yourself a thinking-inside-the-box award.

It's about $15 to ship to the USA and I'd guess it would be about the same (or maybe a little less - NZ post is quite expensive) to send back. I think it's funny that when I defend myself in a polite, non-name-calling way against other's claims of me talking BS then it's branded as "spice". Particularly choice was from "supreme member"(!!) mansr who waved his hands, declared "it is absolute BS" (without it being clear what he was referring to exactly..) and then proceeded to cite both transmission line theory and the USB spec as somehow forbidding whatever it was he was talking about. I wonder if it was my statement that the cable has asymmetrical data path? This is definitely true.

Apologies for rules infraction. #1 fixed. #2 well I was.. Bonus : Thanks for the tip. I've found offering sugar to disbelievers is pointless. But I take your point..

..and that's pretty much exactly what I'd expect you to say. I don't claim to have discovered magic (where did I say that??) The magic in the product name is exactly that. A trademark. Like, say, magic marker. btw, I have a degree in electronic engineering, and although that was obtained almost 40 years ago, yes, we did know about transmission lines in those days. I can see how it's easy to think inside the box and claim BS, but that's the exact problem - you guys think theory maps exactly onto reality, when in fact it doesn't always. And in audio there are some interesting anomalies that defy ready explanation. I know you refuse to believe that USB cables can even sound different - I didn't use to think they could, but the fact is they do, as I have demonstrated to myself and others with blind tests. So in such cases, when the theory says BS but experiments say otherwise, I choose to side with experimental data, assuming the theory is incomplete.

Why would I have an independent proctor? Or a written protocol. Haha. I'm not trying to prove anything to cynical bystanders; the purpose of such tests are mostly to convince me (and a few others) that I have something that's not just noise. And I know how to conduct such tests (cue picking more holes, this never ends so I basically don't care). BTW in some cases blind test show that there's nothing there, but when they work often you get 100% success. My own view is that many folks who criticize haven't spent much time either training their ears to hear small differences, or actually running such tests.

It's not BS. Just because you can't understand how it's done doesn't mean that it's not so. But no, I'm not going to explain it. It's a trade secret. I'd rather the cable stands on it's merits, and this is why we offer money-back if not satisfied. So far nobody has returned one, far from it. In fact half a dozen users have reported that they have now removed various dongles preferring the cable on it's own. But of course I realize there are folks who are simply going to write off USB cables as making a difference. But these cables have been blind tested and picked by testers in 100% of cases (and of course I realize that these same folks will write off my claims just as easily). But frankly, I wouldn't sell something I personally could not pick in a blind test.

I have recently built a system to upsample everything to DSD512. Some details. I use two main DACs, a Audio GD master 11 (uses PCM1704UK R2R) and an LKS DA-004 (ES9038pro). WIth PCM, both sound more similar than different. The AudioGD is perhaps a touch more musical/mellow and the LKS more dynamic and detailed. (That's both feeding them 44/16 and upsampled PCM at up to 384/24) However I tried using HQplayer to convert to DSD128, which was as fast as my music PC would go. Wow, that is sure some difference. The DSD sounds much less digital; going back to PCM with either DAC makes the 'digititus' very obvious. With DSD it was sounding more like a good analog system for musicality, but had the detail, dynamics, etc, that digital does well. So, I decided to build a system to do DSD512. I found a small PC that used a fastish i7-4750HQ CPU along with a decent nVidia graphics card (for CUDA offload). GTX-750Ti This worked great for upsampling PCM to DSD512. However, due to the heat produced, it was noisy with it's fan running fast. The solution to this was to move this PC to another room. The original music PC now sits in the music room, running Roon Remote Control. It also runs HQplayer's Network Audio Daemon. The fast PC runs Roon core, which I'm using also to do room correction via filters I built with Accourate. It then sends output to HQplayer to convert to DSD512. This is them routed to the network audio daemon on the music PC, and then via USB to the LKS DAC. Performance When playing music, the core PC (i7 cpu) runs at around 40-45% CPU use. The nVidia GPU is also running at around 45%. It's also using between 45-50Mbps network bandwidth. This is the PC https://www.aliexpress.com/item/Eglobal-Mini-PC-Intel-Core-i7-4750HQ-Quad-Core-8-Threads-Max-3-2GHz-Nvidia-GTX/32814789183.html This PC is just about fast enough to work perfectly - ideally you'd have a PC twice as fast as this, but I was on a budget! The music room PC is a fanless NUC style thing; just about anything will do here. Even with a lowly Celeron j1900 cpu, it only runs a few percent CPU

Audio-GD master 7. Can switch between nos, 2x, 4x and 8x oversampling

Viruses won't attack a computer behind a firewall - they need help from a human. Nealy all viruses are transmitted by human actions like 'doing something in a browser' or opening email attachments, etc. If you don't do that on your audio pc then you are hardly at risk from them. Anyway, there's this thing called backups that everyone should do anyway, that can protect you from this. Of course, it's entirely a matter for the individual, but my audio pc is in test mode and I don't have any issues, (and I speak as someone who's been in the IT industry, mostly as software designer, for 40 years.)

A useful thing to do on windows 10 setups used for audio is to prevent the pesky automatic updates, by putting windows 10 into test mode. Here's how: To enable Test Mode on Windows, open an elevated command prompt (Press Start->Search->cmd then right-click on it and click Run as administrator) Enter the following command: bcdedit /set testsigning on and reboot after running the command above To disable Test mode, open an elevated command prompt as above and enter the following command: bcdedit /set testsigning off and also reboot after running the above command.

I note that like most measure-heads he measured the amp doing nothing (playing silence), playing a 1khz sine wave and the old 19khz, 20khz intermod test. (Also note the amp is playing into a resistor, not a speaker. This is important for 2 reasons - speakers are complex loads, and amps may well be affected by the same sound being played audibly - resonances, etc) This to me is like reviewing a car by just measuring it's part-throttle power and torque outputs, rather than actually driving it. Actually more like measuring it on a roling road simulating driving on a straight road at 30mph. Most amps will reproduce a sine wave very easily, which is why you see these vanishingly low distortion figures even for pretty ordinary amps. Making a 1khz tone most amps sound the same.. Similarly, I note that most amps can reproduce very simple music, eg one simple instrument, quite well, espeically at lower volumes. Amps tend to have problems playing very complex music at louder volumes, eg orchestra at full bore. (I don't mean pusking amp into clipping, just playing loud within it;s limits) If you look at how this differs to 1khz test tones you see that you have many many signals all mixed together. The trick is being able to reproduce the lower level signals while also reproducing the larger level signals, at the same time as being bathed in air that's vibrating almost in sympathy with the signals the amp is trying to reproduce. When measure-heads can find a way to accurately measure an amp's performance under these conditions then I will be interested in their results (yes I have heard of diffmaker, and it is not a suitable tool to do this kind of thing). I think that then they will be able to measure what the objectivists hear. On a related point : I've noticed that the term "noise floor" is used very differently by the obs and subs. Objectivists think that 'signal to noise ratio' and 'noise floor' are the same sort of thing. Subjectivists talk about noise floor as they experience it, ie "apparent noise floor" - it's a measure of how far you can hear into the music, how well you can hear low level signals in the presecnce of loud signals. As I mentioned, when amps are asked to play loud complex music they generate 'noise' that obscures the finer details. Actually it's not really noise at all, it's distortion (the distinction being that distortion is related to the music while noise is random), but this effect has become known as 'noise floor' in audiophile-speak. Yet another reason obs and subj talk past each other.

Cool, I remember doing that - my fave track was Don't Leave Me Now from The Wall..has the most amazing 3d shapes in there..

SMPS are quite tolerant of power supply voltage spikes/surges. For instance, many SMPS will be rated "100V - 240V" input, so you can use the same smps if your mains is 120V or 240V. And they won't run any hotter on 240V than 120V Agree with you re: smps potentially polluting the power supply. But you don't need to wory about spikes and surges.

Note that putting a capacitor in series with the tweeter will protect it from DC coming out of the amp, but would not stop the hash you had before killing the tweeter..for that you need a fuse (or polyswitch...not sure how good they sound)

OK new product category : Break-in software that will break in your software for you..

Someone kill this thread

Yes after I discovered this I looked into why the GND weas needed, and I agree with you. But my point remains that the presence of the GND is one thing (maybe the main thing) that causes SQ degradation on USB. (But of course if you don't believe in any such degradation, the point is moot.) I didn't use the correct term - by 'properly balanced' I meant 'does not need the ground to function'

There are 4 wires on a USB connection : Data+, Data-, GND and +5v When I first tried to make a 'data only' cable I tried just connecting Data+ and Data- wires, but this does not work, at least on the few USB connections that I tried. I found (to my disgust) that it also needs the GND connection to work. This implies that the USB data connection is not properly balanced (as if it were, you would not need the GND - it would be optional). I think that this is one of the root causes of USB problems - the involvement of the ground. That is why devices like the intona work so well - they break the ground link.