USB Cables and DirectStream

mikedc said
Paul McGowan said I think the problem is the Bryston is a Linux based computer and I don't know if there is a driver compatible with our DAC. Any time you connect a computer to a DAC there has to be an appropriate driver to connect the two. That driver is built in on a Mac, you can download the Windows driver from our website, but I don't know of a Linux based driver (Linux is an operating system like Windows).

I am pretty sure it won’t work.

That really doesn't explain why some USB cables work and some don't.
I'm going to be blunt - I'm not going to waste any of my time on cables that don't pass the USB compliance tests (e.g. that have a USB logo.) It's hard enough to deal with all possible combinations of people's systems with legitimate USB cables. Cables longer than 5 meters, cables without VBUS (+5V) connected, etc. WILL cause problems in some systems. There are reasons for those compliance tests and just because an illegal cable works in one system doesn't mean it will work in another, similarly if it works between two USB devices doesn't mean it will work between two other similar USB devices.

If you are having a USB problem - at least try a standard USB cable and try to run the bit perfect test. If you can’t get the bit perfect test to run, but you can hear music mostly correctly you probably have a set up problem. If the bit perfect test works with a standard cable but fails with a non-standard cable I don’t care.

I’m not saying there are no USB problems, but all of the problems I’ve been involved in that we ran into the ground were all set up problems in the OS or the playing software or bogus cables. We don’t know the outcome of Rob H’s USB issue yet - but he gets different problems depending on his cable choices and we haven’t seen any cable work correctly in it yet, so there’s still possibly a setup issue or a hardware problem. [Edit Rob H. just posted that a POS USB cable works, and that he’s happy he doesn’t think it’s a DS issue.]

mikedc said
Paul McGowan said I think the problem is the Bryston is a Linux based computer and I don't know if there is a driver compatible with our DAC. Any time you connect a computer to a DAC there has to be an appropriate driver to connect the two. That driver is built in on a Mac, you can download the Windows driver from our website, but I don't know of a Linux based driver (Linux is an operating system like Windows).

I am pretty sure it won’t work.

That really doesn't explain why some USB cables work and some don't.
Yes, there is a linux driver for the DirectStream/PWD. At least for ubuntu. I am running it at home.
Ted Smith said If the bit perfect test works with a standard cable but fails with a non-standard cable I don't care
Kudos!

Back to basic, fundamental engineering. It may not be as fun as pushing fairy dust around, but it works; fairy dust, not so much.

Dennis Kerrisk said

Tes there is a linux driver for the DirectStream/PWD. At least for ubuntu. I am running it at home.


Where did you find the driver?

Ted Smith said
If the bit perfect test works with a standard cable but fails with a non-standard cable I don't care.
LOL!!! Well don't look now guys but there's just been 100 posts on Audiogon from guys selling $1000 USB cables, 4 for $1.

In all seriousness, since you’re not gonna waste your time on this issue, Ted, perhaps it would be a service to those shelling out the thousands of dollars for PSAudio equipment if they released a statement to the effect that “audiophile” or “specialised” or “non-standard” USB cables should not be used with the Direct-Stream DAC and that customers should be sure to purchase ONLY USB cables that are stamped with the USB logo (who knew?!) instead of going on about how much difference a USB cable makes and not to skimp on purchasing the best cable one can afford.

But then again, I don’t care.

Well let’s not get carried away here. Ted might have been a bit abrupt in his response.

I think what he means is the few USB cables that are somehow doing something to the USB data that gives results such that DirectStream fails the bit perfect test we created, have something fundamentally flawed with them. They are rare.

Couple of points to note.

  1. No other manufacturer I am aware of provides a bit perfect test and the means to conduct it and verify the results
  2. By doing this you're able to see what possible damage the one or two aftermarket cables do
  3. If these suspect cables are not passing USB data properly with DS they are also not passing it properly on any other DAC
  4. Isn't it better to know about those few and cull them out?
So at least you're able to tell. USB cables make a significant sonic difference in DS as they do in most DACs. Ted is welcome to his own opinion on the subject and he and I openly disagree on any number of small topics like this. That's ok, it's how we like to roll. I don't demand people withhold their opinions on these forums, in fact I encourage them.

Ted’s one of the brightest engineers in our industry, period. That’s why he’s part of the team and why we collaborated with him. That he may not hear the same level of sonic changes in firmware, cables or anything else mentioned in these forums, that I do or Arnie does, is not relevant to these discussions. Nor, for that matter are my own observations. They are just that: my observations, Ted’s observations.

And I do care.

It seems to me that Ted has been quite consistent on his views of USB cables and his own use of modest cables. Paul has said that he uses an expensive cable (that Ted has said he did’t care for as much), but I don’t recall Paul telling people they really need to get the most expensive cable they can afford, which is different from saying that cables do matter (more to Paul than Ted, apparently). Members may have chosen to do so and that’s their right. Ted’s point is that he shouldn’t be expected to take the time to figure out why every particular high-end cable that ignores USB standards doesn’t work in a particular system. That seems perfectly reasonable to me.

mikedc said

In all seriousness, since you’re not gonna waste your time on this issue, Ted, perhaps it would be a service to those shelling out the thousands of dollars for PSAudio equipment if they released a statement to the effect that “audiophile” or “specialised” or “non-standard” USB cables should not be used with the Direct-Stream DAC and that customers should be sure to purchase ONLY USB cables that are stamped with the USB logo (who knew?!) instead of going on about how much difference a USB cable makes and not to skimp on purchasing the best cable one can afford.

As other's have said you are slightly misquoting me and PS Audio. I never said don't use whatever USB cable you like. I said if you use a non-standard USB cable and it doesn't work but a standard one does, that I feel no obligation to spend much time helping. If you buy a power cord that doesn't provide enough current capacity, or use a power conditioner than doesn't provide (roughly) 120 V at 60 Hz I don't feel an obligation to help either.

Paul and I have consistently said which USB cables we prefer and we do stand by our recommendations (i.e. we’ll definitely try to help people that can’t get the DS to work with them.)

I’ve also always expressed skepticism about cables that don’t hook up 5V and long USB cables. I’ve answered questions about them over and over. I also don’t feel bad about asking people to try legitimate USB cables, how are we supposed to know how every USB like cable out there cheats?

I doubt anyone would question whether I care about helping users of the DS - I care enough that I don’t want to take time away from development or helping customers who will try standard cables, at least as a test. (BTW some of the development time is spent trying to broaden the types of systems my products work in, including trying weird things to make USB sound better and what “cheats” I could stand by.)

Ted Smith said

I doubt anyone would question whether I care about helping users of the DS . . .


It is apparent you care deeply. You spend a great deal of volunteer time on this forum (and others) helping people with a myriad of issues.

mikedc said

. . . it would be a service to those shelling out the thousands of dollars for PSAudio equipment if they released a statement to the effect that “audiophile” or “specialised” or “non-standard” USB cables should not be used with the Direct-Stream DAC . . .


This is not what anyone stated. Rather, those “audiophile” USB cables which are unable to work as basic USB cables are not something Ted, PS Audio or anyone else can, or should, be expected troubleshoot. Frankly, the buyer that elects to buy into such silliness deserve what they get. Sadly, there are “audiophile” cables that have trouble adequately functioning as a slow speed printer interface.

OTOH, if you buy a specialty cable that is able to pass basic muster as a real live functioning USB cable (most do), Ted and everyone else here is happy to give you a hand.

Undies unbunched?

Certainly there are “levels” of support and reasonable limits on each level.

3rd party parts are made to written standards and assembled as such with the expectation that all will function as expected when adhering to these standards.

We have a somewhat unique mixed blessing here at PSA world in that the talented 77_gif Engineer as well as the handsome proprietor are active in the general forum where it is certainly appropriate, helpful and fun, to discuss the experiences we have when using accessories that may enhance our enjoyment.

i am sure the USB feedback and comments are helpful to the engineers over time, but just not on the hot list today.

Cables DO make a difference.

some work and some don’t. See… We figured that out together.respect-010_gif

Happy Friday Y’all.

ozzymilton said

When I connect the USB from the Bryston to the Direct Stream on the Bryston settings page it shows the devices connected but the USB is not listed.


Have you contacted Bryston? Issue resolved?

Bryston is supposed to have a conference call with me this evening. We’ll see how that goes…

Two questions for Engineer Ted:

Let’s say I have two cables; one from an audio cable manufacturer of some repute, one a Gigaware from RadioShack. The Gigaware packages says “USB certified.” The one from the cable manufacturer proudly displays its name, but I can’t discern its lineage.

  1. Is there any way to tell if the “boutique” USB cable is USB Certified, other than asking the manufacturer, who will most likely say, “But, of course.”

  2. If I’m checking the “boutique” cable, is there ANY reason at all why the DS DAC would encounter a problem, but the PWDII would not take issue with that cable? *

  • I have further testing to do to confirm this, but it seems that this might be the case.

Thanks, Ted. We’re all fortunate to have you helping us.

Rob

ozzymilton said Bryston is supposed to have a conference call with me this evening. We'll see how that goes...
Excellent! Please let us know how it goes.

The answer should help others with related issues.

Rob H. said Two questions for Engineer Ted:

Let’s say I have two cables; one from an audio cable manufacturer of some repute, one a Gigaware from RadioShack. The Gigaware packages says “USB certified.” The one from the cable manufacturer proudly displays its name, but I can’t discern its lineage.

  1. Is there any way to tell if the “boutique” USB cable is USB Certified, other than asking the manufacturer, who will most likely say, “But, of course.”

  2. If I’m checking the “boutique” cable, is there ANY reason at all why the DS DAC would encounter a problem, but the PWDII would not take issue with that cable? *

  • I have further testing to do to confirm this, but it seems that this might be the case.

Thanks, Ted. We’re all fortunate to have you helping us.

Rob

It costs money to get your product (in this case a cable) certified. You can't use the USB logo without being certified (http://www.usb.org/developers/compliance/).

Tho the DS and the PWD both use parts (and code) from XMOS, those parts are different and have different levels of software. The DS uses a part with an integrated USB PHY and the PWD uses a separate PHY, I suspect that any differences are due to the differences in the PHYs or the level of XMOS software. I can’t guess which things might be different, but I know that there’s apparently more trouble with connecting to the DS than the PWD with some levels of OSs. Besides those differences we have looked to see if there’s something obviously different in the implementations but haven’t found anything in particular yet.

Ted Smith said:

It costs money to get your product (in this case a cable) certified. You can’t use the USB logo without being certified (http://www.usb.org/developers/compliance/).

Tho the DS and the PWD both use parts (and code) from XMOS, those parts are different and have different levels of software. The DS uses a part with an integrated USB PHY and the PWD uses a separate PHY, I suspect that any differences are due to the differences in the PHYs or the level of XMOS software. I can’t guess which things might be different, but I know that there’s apparently more trouble with connecting to the DS than the PWD with some levels of OSs. Besides those differences we have looked to see if there’s something obviously different in the implementations but haven’t found anything in particular yet.


Perfect!

Just knowing there’s a difference in implementation strategies is enough to convince me that there can be a difference. IOW, if you had said that the two units are exactly the same, I’d be scratching my head.

I’m about ready to throw in the towel on that cable. I could test it further on the DS, but I don’t want to. Between having a USB module on the Off Ramp go bad, and the Mini port that it was attached to exhibiting unsociable behavior, I really don’t even want to reintroduce it into the system, lest I bust something else.

If you have a moment: In a sense, isn’t wire, wire? Even if said manufacturer didn’t pay for certification, could s/he have done something that frakked things up?

One thing I’ll say is that the results from my testing seem remarkably consistent. The only variation is Port 1 on the Mini, which may have been damaged by the cable.

Rob H. said If you have a moment: In a sense, isn't wire, wire? Even if said manufacturer didn't pay for certification, could s/he have done something that frakked things up?
Oh yes. USB isn't like RS-232. It's high speed. Capacitance (and inductance) will slow a signal down and roll off the edges. Rolled off edges is jitter. A too slow signal will cause timeouts in the USB protocol with certain situations (mostly with long cables and multiple hubs, but you get the idea.) These time outs can affect both connecting and disconnecting (and hence also cause hibernation and sleep issues) and it can cause retries which might just make a disk drive connection slower, but for audio it can cause dropped packets...

Wire gauge is important - too small and the losses (especially in the 5V line) will cause the level to be too low to drive the device at the end of the cable. Or worse (in some sense) it can cause random false connects and disconnects: this isn’t good with disk drives or audio.

Shielding is important, …

If you read the specs you’ll see the kinds of things that aren’t obvious and many ways to screw them up.

Just got off the phone with Bryston. No luck getting the Bryston BDP2 USB to work with the Direct Stream. They are going to get more great minds involved next week.

Nuts, although good to hear they are on it.

It must be a challenge to support all USB DACs.

Gotta give 'em credit for customer service.