Jump to content
IGNORED

Another new USB regenerator: Lightspeed Revive


Recommended Posts

That's an overly simplified and even erroneous view of how async USB audio in audiophile system works...

 

I suggest you take some time and read everything that John Swenson has written about the subject here and also in his interview over at Audiostream.

 

You're not getting it right at all. It isn't about getting the data through or not.

He already knows that as his subsequent reply to my posts shows. He just chooses to be troll-like in these posts

Link to comment
He already knows that as his subsequent reply to my posts shows. He just chooses to be troll-like in these posts

 

Ah well, why am I not surprised? :P

Dedicated Line DSD/DXD | Audirvana+ | iFi iDSD Nano | SET Tube Amp | Totem Mites

Surround: VLC | M-Audio FastTrack Pro | Mac Opt | Panasonic SA-HE100 | Logitech Z623

DIY: SET Tube Amp | Low-Noise Linear Regulated Power Supply | USB, Power, Speaker Cables | Speaker Stands | Acoustic Panels

Link to comment
Ah well, why am I not surprised? :P

 

Just because I have a different (and more experienced view on USB) why does that make me a troll...

I spend all day as said getting signals from a to b when doing layouts etc. quite often for critical situations, I have had to achieve 18GHz immunity on some designs so know how to stop noise getting through, instead of resorting to the usual schoolyard attitude try reading some stuff by Henry Ott, Ralph Morrison, Howard Johnson, Eric Bogatin etc. Spend a few days playing with some signal integrity and power integrity software and basically work on designs where extensive measurements are done so you get feedback of what is happening, then you'll learn what works and what doesn't...

Maybe you should question what you believe more often then you may find the truth... Audio is one hobby full of mis-direction and myths and really needs to get back to some reality, not just blindly following the herd.

Link to comment

Hey marce, do you have a decent audio system yourself ? I mean, is audio your hobby ?

I only ask because we audio crazies hear differences in everything. This could be because we like to hear those differences.

 

You ?

Lush^3-e      Lush^2      Blaxius^2.5      Ethernet^3     HDMI^2     XLR^2

XXHighEnd (developer)

Phasure NOS1 24/768 Async USB DAC (manufacturer)

Phasure Mach III Audio PC with Linear PSU (manufacturer)

Orelino & Orelo MKII Speakers (designer/supplier)

Link to comment
Just because I have a different (and more experienced view on USB) why does that make me a troll...

I spend all day as said getting signals from a to b when doing layouts etc. quite often for critical situations, I have had to achieve 18GHz immunity on some designs so know how to stop noise getting through, instead of resorting to the usual schoolyard attitude try reading some stuff by Henry Ott, Ralph Morrison, Howard Johnson, Eric Bogatin etc. Spend a few days playing with some signal integrity and power integrity software and basically work on designs where extensive measurements are done so you get feedback of what is happening, then you'll learn what works and what doesn't...

Maybe you should question what you believe more often then you may find the truth... Audio is one hobby full of mis-direction and myths and really needs to get back to some reality, not just blindly following the herd.

 

That is all very interesting Marce. Are you an engineer? If so, why don't you just build a better mouse trap and put it on the market? Or is there something already on the market that you think is better than the devices being discussed on this forum? Or do you just think these usb hubs unnecessary?

 

From your posts it seems you have no direct experience with the actual devices we are discussing. In which case all you have is theory. In real life theory does not always pan out.

Link to comment
Just because I have a different (and more experienced view on USB) why does that make me a troll...

I spend all day as said getting signals from a to b when doing layouts etc. quite often for critical situations, I have had to achieve 18GHz immunity on some designs so know how to stop noise getting through, instead of resorting to the usual schoolyard attitude try reading some stuff by Henry Ott, Ralph Morrison, Howard Johnson, Eric Bogatin etc. Spend a few days playing with some signal integrity and power integrity software and basically work on designs where extensive measurements are done so you get feedback of what is happening, then you'll learn what works and what doesn't...

Maybe you should question what you believe more often then you may find the truth... Audio is one hobby full of mis-direction and myths and really needs to get back to some reality, not just blindly following the herd.

 

I don't doubt for a moment that you work in this area so why not bring some of this/your experience to the topic & analyse the USB connection between devices for possible flaws in noise transmission? Why don't you relate some of your experiences where you measured & found a noise issue & what you did to fix this? Don't tell me this never happened. Maybe that would be adding something to the discussion rather than the cynical troll-like posts that you normally engage in?

 

PS.I see others have posted pretty much the same as me while I wrote this post. Does this tell you anything?

Link to comment

I have said that all the designs I have seen do not fully isolate the circuitry in terms of noise, something I find worrying in designs that are suppose to minimise noise. The main coupling mechanism for high frequency noise is capacitive, so clear moats in the layout crossed by pi filters for power and CM chokes for the signal are a big step forward.

I posted a layout for a USB isolator on another forum, don't have access to my works computer so will post the layout here and you can see the isolation barrier, also stack up details to get 90 ohm diff impedance for the USB lines.

Sorry if I seem picky but isolating noise is not easy and if claims are going to be made about a piece of kit doing that then in my view they do need backing up... and also looking at in detial.

Link to comment
I have said that all the designs I have seen do not fully isolate the circuitry in terms of noise, something I find worrying in designs that are suppose to minimise noise. The main coupling mechanism for high frequency noise is capacitive, so clear moats in the layout crossed by pi filters for power and CM chokes for the signal are a big step forward.

I posted a layout for a USB isolator on another forum, don't have access to my works computer so will post the layout here and you can see the isolation barrier, also stack up details to get 90 ohm diff impedance for the USB lines.

OK, look forward to it but are you sure it works at USB high speed rates 480MHz & up? For instance ADUM isolator chips will not work at 480Mbps - we are not talking about the 5V USB supply so pi filters have no real significance here - CM chokes are already well known in addressing CM noise on USB

 

Sorry if I seem picky but isolating noise is not easy and if claims are going to be made about a piece of kit doing that then in my view they do need backing up... and also looking at in detial.
Sure, it's not easy - that's why I asked about high speed above. It's also why measurements are lacking as it would seem this is affecting the DAC analogue outputs in ways that stock FFT measurements aren't suitable to expose or at levels below which we normally consider the threshold of audibility. Most of the people working in this area don't have the sophisticated (read expensive) equipment necessary to investigate this.

 

Maybe you could tell us what equipment you use on a regular basis in analysing these high-speed communication links?

Link to comment
Just because I have a different (and more experienced view on USB) why does that make me a troll...

I spend all day as said getting signals from a to b when doing layouts etc. quite often for critical situations, I have had to achieve 18GHz immunity on some designs so know how to stop noise getting through, instead of resorting to the usual schoolyard attitude try reading some stuff by Henry Ott, Ralph Morrison, Howard Johnson, Eric Bogatin etc. Spend a few days playing with some signal integrity and power integrity software and basically work on designs where extensive measurements are done so you get feedback of what is happening, then you'll learn what works and what doesn't...

Maybe you should question what you believe more often then you may find the truth... Audio is one hobby full of mis-direction and myths and really needs to get back to some reality, not just blindly following the herd.

 

The funny thing is despite name-dropping all that and all the work you do, you still don't get it.

 

The reason I'm not surprised is that we've already established the above from related conversations on diyaudio...

Dedicated Line DSD/DXD | Audirvana+ | iFi iDSD Nano | SET Tube Amp | Totem Mites

Surround: VLC | M-Audio FastTrack Pro | Mac Opt | Panasonic SA-HE100 | Logitech Z623

DIY: SET Tube Amp | Low-Noise Linear Regulated Power Supply | USB, Power, Speaker Cables | Speaker Stands | Acoustic Panels

Link to comment
I have said that all the designs I have seen do not fully isolate the circuitry in terms of noise, something I find worrying in designs that are suppose to minimise noise. The main coupling mechanism for high frequency noise is capacitive, so clear moats in the layout crossed by pi filters for power and CM chokes for the signal are a big step forward.

I posted a layout for a USB isolator on another forum, don't have access to my works computer so will post the layout here and you can see the isolation barrier, also stack up details to get 90 ohm diff impedance for the USB lines.

Sorry if I seem picky but isolating noise is not easy and if claims are going to be made about a piece of kit doing that then in my view they do need backing up... and also looking at in detial.

 

Hi marce, good point about how difficult noise isolation is, and looking forward to seeing some good information from you on designs for this purpose.

 

A couple of things on the audio side:

 

- For at least some of these pieces, part or all of the theory of operation is not isolation but avoidance of self-noise in the USB receiver circuitry.

 

- Noise to the point of bit errors isn't a problem in any DAC I'm aware of these days, so what's being worked on isn't what I presume you're after in your job (avoiding such errors), but rather noise getting into the clock circuitry and causing jitter, and/or coupling into the analog side.

 

Neither of these points takes anything away from the importance of isolation, which as you've said is difficult to do well.

One never knows, do one? - Fats Waller

The fairest thing we can experience is the mysterious. It is the fundamental emotion which stands at the cradle of true art and true science. - Einstein

Computer, Audirvana -> optical Ethernet to Fitlet3 -> Fibbr Alpha Optical USB -> iFi NEO iDSD DAC -> Apollon Audio 1ET400A Mini (Purifi based) -> Vandersteen 3A Signature.

Link to comment
Hi marce, good point about how difficult noise isolation is, and looking forward to seeing some good information from you on designs for this purpose.

 

A couple of things on the audio side:

 

- For at least some of these pieces, part or all of the theory of operation is not isolation but avoidance of self-noise in the USB receiver circuitry.

 

- Noise to the point of bit errors isn't a problem in any DAC I'm aware of these days, so what's being worked on isn't what I presume you're after in your job (avoiding such errors), but rather noise getting into the clock circuitry and causing jitter, and/or coupling into the analog side.

 

Neither of these points takes anything away from the importance of isolation, which as you've said is difficult to do well.

Yes, so it would be interesting to see noise measurements that Marce has encountered in his work, what the causes were & what he has done to alleviate this.

 

The problem I see is that marce's schematic on DIYAudio is based on an ADUM isolation chip none of which handle high speed USB 2.0 - which most current USB audio devices operate at.

 

So, if this is the schematic, it's a non-runner.

Link to comment
@marce & @YashN - How about we keep it civil and figure out what each other doesn't understand or help educate the CA community on this issue. I'm serious. It would be cool to look at each other's point of view and try to question specific items etc... No need to get personal. Just looking to educate.

 

How is recommending some reading to do so as to get more insight about the finer details as explained by John Swenson be not civil?

 

Anyone who is really willing to learn has access to the information for free.

Dedicated Line DSD/DXD | Audirvana+ | iFi iDSD Nano | SET Tube Amp | Totem Mites

Surround: VLC | M-Audio FastTrack Pro | Mac Opt | Panasonic SA-HE100 | Logitech Z623

DIY: SET Tube Amp | Low-Noise Linear Regulated Power Supply | USB, Power, Speaker Cables | Speaker Stands | Acoustic Panels

Link to comment

The problem I see is that marce's schematic on DIYAudio is based on an ADUM isolation chip none of which handle high speed USB 2.0 - which most current USB audio devices operate at.

 

So, if this is the schematic, it's a non-runner.

 

I believe Superbad Alex as well as John S. have some views on the disadvantages of using the ADUMs as well (jitter-related as far as I can remember).

Dedicated Line DSD/DXD | Audirvana+ | iFi iDSD Nano | SET Tube Amp | Totem Mites

Surround: VLC | M-Audio FastTrack Pro | Mac Opt | Panasonic SA-HE100 | Logitech Z623

DIY: SET Tube Amp | Low-Noise Linear Regulated Power Supply | USB, Power, Speaker Cables | Speaker Stands | Acoustic Panels

Link to comment

@Chris

"figure out what each other doesn't understand or help educate the CA community on this issue."

The problem is that marce does the opposite of what you are asking for in your quote - in this matter, he doesn't educate the CA community (he misinforms), he doesn't try to understand (he repeats the same bits are bits point) & yet he presents himself as an expert in this field.

 

We have asked him a number of questions & would be interested in hearing his answers.

Link to comment
How is recommending some reading to do so as to get more insight about the finer details as explained by John Swenson be not civil?

 

Anyone who is really willing to learn has access to the information for free.

Wow, don't take it so personal. I asked you guys to do something. I didn't scold you for previous behavior.

Founder of Audiophile Style | My Audio Systems AudiophileStyleStickerWhite2.0.png AudiophileStyleStickerWhite7.1.4.png

Link to comment

Please keep in mind that threads and comments are about much more than the people involved at the moment. Telling someone to lookup information is far less helpful than simply providing a link to what you're talking about. Plus, when someone seeking this information finds this thread through Google in 1,2,5 years from now, it's much more beneficial to actually have the discussion right here or provide links.

 

I'm not worried that people did the opposite of what I'm asking, before I actually asked it. That's why I asked it. So the future posts will help educate.

Founder of Audiophile Style | My Audio Systems AudiophileStyleStickerWhite2.0.png AudiophileStyleStickerWhite7.1.4.png

Link to comment
Wow, don't take it so personal. I asked you guys to do something. I didn't scold you for previous behavior.

 

Not taking anything personally: your recommendation is worthwhile, but I figured you were possibly referring to mmerill99's words instead of mine (Marce quotes me but is in fact referring to mmerill99's characterisation of him as a 'troll' - I haven't used the word but I think no less). In case you were in fact referring to me, please re-check Post #50 where I suggest additional info for him to dig deeper.

 

In any case, if Marce wants to learn deeper, he now knows who to ask to where to look further.

 

Thing is, one would need to 'empty one's cup' to be able to do that. Knowing the person from another forum for years, I'd tend to agree with mmerill89's assessment though.

 

About emptying his cup, I think I have already mentioned where he gets it wrong, and that's his assumption that as long as USB transfers the data correctly, then everything is fine and dandy.

 

This, as we know, isn't the case (although ideally we'd like it to be so).

 

Now, maybe John himself or PeterSt can provide him directly with more info, and this coming from people who have actually done much more work and measurements and experimentations and building than Marce has in this specific domain.

 

The info is already there (on Audiostream, here on CA, and also on a not so old thread over at diyaudio). People who are serious about getting it will get to it.

 

In this particular case, I do not think that he is disposing himself to any new understanding, but that's my own opinion, informed from what I've seen over at diyaudio for years.

 

You'd think that his account here would be to learn something new, but it doesn't appear to be so, at least to me.

 

As for providing links, I have often provided them in the past, Marce can search for them if he's really interested, or he can find my started thread about "Beyond Bit-Perfect". Others can do the same.

Dedicated Line DSD/DXD | Audirvana+ | iFi iDSD Nano | SET Tube Amp | Totem Mites

Surround: VLC | M-Audio FastTrack Pro | Mac Opt | Panasonic SA-HE100 | Logitech Z623

DIY: SET Tube Amp | Low-Noise Linear Regulated Power Supply | USB, Power, Speaker Cables | Speaker Stands | Acoustic Panels

Link to comment
Wow, don't take it so personal. I asked you guys to do something. I didn't scold you for previous behavior.

 

I must say also that things can easily be misinterpreted as in this case with the new 'Notification' system. It makes things a bit formal and ominous sometimes.

 

But thanks for taking the time to explain a little more of your post, as it didn't actually come across like what you describe at all.

Dedicated Line DSD/DXD | Audirvana+ | iFi iDSD Nano | SET Tube Amp | Totem Mites

Surround: VLC | M-Audio FastTrack Pro | Mac Opt | Panasonic SA-HE100 | Logitech Z623

DIY: SET Tube Amp | Low-Noise Linear Regulated Power Supply | USB, Power, Speaker Cables | Speaker Stands | Acoustic Panels

Link to comment
Yes, so it would be interesting to see noise measurements that Marce has encountered in his work, what the causes were & what he has done to alleviate this.

 

The problem I see is that marce's schematic on DIYAudio is based on an ADUM isolation chip none of which handle high speed USB 2.0 - which most current USB audio devices operate at.

 

So, if this is the schematic, it's a non-runner.

 

 

The whole point of the design posted was the isolation moats to restrict capacitive coupling of the noise, the design was from another bit of kit where galvanic isolation was required for safety... Yes the Adums can increase jitter (how critical is that to USB audio? many differing views, but that's another discussion). I do believe I did explain this in quite some detail on that thread. The idea has been copied I believe on another commercial design, though this is not an idea I have come up with, it is pretty standard practice to filter noise where a cable or input enters a board, I just spent a few weeks working with an RF engineer to fine tune the idea to minimise noise and interference coupling. RF engineers tend to make the best EMC guys I have found. But the basics are there, just replace the Adum with say a SMSC USB2412 and associated circuitry and you have a USB hub fully isolated.

I will digest the other posts and reply in more detail later today hopefully when I have time.

 

I do see that YashN has had a good go at slagging me off and my abilities and what I know and work on (just read his reply). I have well over 30 years experience with electronics and PCB design so I would appreciate it if before my abilities are demeaned the person doing it actually finds out what experience I have... I tend to work mainly on cutting edge projects these days including many years on this project responsible for the PCBs...

http://www.northropgrumman.com/capabilities/anvic5/documents/vic5.pdf

I am also working on a few more projects of a similar nature, so surprise surprise we actually do a dam site more measurements that many others due to the products reliability requirements and the fact that in many cases any problems with noise will result in a lot more problems than distortion... So I probably have a lot more practical experience on this than many.....

Just wanted to put this straight since I am being made out to be a bit of a Muppet and I'm not really happy about it...

Yes my views do differ from many and just because I have some valid comments on the layouts and noise reduction I get crucified, not for me I have bigger things to worry about at the moment...

Link to comment
Hey marce, do you have a decent audio system yourself ? I mean, is audio your hobby ?

I only ask because we audio crazies hear differences in everything. This could be because we like to hear those differences.

 

You ?

I tend to focus mainly on building speakers as these have the most influence of the resultant sound at my ears... My current ones are OB's everything above about 60Hz through the Fostexes with the 2x15 inch per side driven by a SS amp active crossover, also did a pair of curvy changs, a few back loaded horns some standard 3 way designs over the years... Now I just keep things simple, basic PC running squeezebox server and nowt' else and squeezebox touches for streaming... Sold or gave away a lot of gear such as Kef 102/4s, micromega CD palyer, marantz CD63 audiolab stuff etc. Funny when I got more into blind testing the less I am concerned now about the minutiae, but we all choose our own path, mine now is I realise how easily our perceptions can be fooled so dont take anything for granted.

:)

Bye

Marc

258380d1325684241-full-range-speaker-photo-gallery-ob1-small.jpg

Link to comment
The whole point of the design posted was the isolation moats to restrict capacitive coupling of the noise, the design was from another bit of kit where galvanic isolation was required for safety... Yes the Adums can increase jitter (how critical is that to USB audio? many differing views, but that's another discussion). I do believe I did explain this in quite some detail on that thread. The idea has been copied I believe on another commercial design, though this is not an idea I have come up with, it is pretty standard practice to filter noise where a cable or input enters a board, I just spent a few weeks working with an RF engineer to fine tune the idea to minimise noise and interference coupling. RF engineers tend to make the best EMC guys I have found. But the basics are there, just replace the Adum with say a SMSC USB2412 and associated circuitry and you have a USB hub fully isolated.
You seem to be confused about what isolation is which is somewhat surprising considering the image of expertise you portray on this forum.

 

So let me give you the basics (which I assumed you would know) the ADUM is an ISOLATOR IC - it electrically isolates the USB communication but only works at USB full speed & not at USB high speed so it is useless for most current USB audio devices. The concept of isolation moats are only applicable when an isolation chip is used

 

The USB2412 is a USB hub IC - it is a central messaging hub which routes USB signals through it's USB ports - NO ISOLATION is involved, no moats will change this

 

The 5V USB Vbus is not material to what is being discussed here so your pi filter has no use here. BTW, the ground plane moats are electronic basics & not the sort of exciting news you seem to be suggesting!

 

I'm afraid that you are displaying a complete lack of knowledge in this area. Despite the fact that I already told you an ADUM was not applicable you continue to state what you did & compound the misinformation by suggesting that a USB2412 chip could be substituted for the ADUM chip.

 

I'm sorry but I have to ask this - do you really work in the field you are purporting to work in? As your displayed knowledge is so lacking in the basics that I'm beginning to doubt this claim.

 

 

I will digest the other posts and reply in more detail later today hopefully when I have time.

 

I do see that YashN has had a good go at slagging me off and my abilities and what I know and work on (just read his reply). I have well over 30 years experience with electronics and PCB design so I would appreciate it if before my abilities are demeaned the person doing it actually finds out what experience I have... I tend to work mainly on cutting edge projects these days including many years on this project responsible for the PCBs...

http://www.northropgrumman.com/capabilities/anvic5/documents/vic5.pdf

I am also working on a few more projects of a similar nature, so surprise surprise we actually do a dam site more measurements that many others due to the products reliability requirements and the fact that in many cases any problems with noise will result in a lot more problems than distortion... So I probably have a lot more practical experience on this than many.....

Just wanted to put this straight since I am being made out to be a bit of a Muppet and I'm not really happy about it...

Yes my views do differ from many and just because I have some valid comments on the layouts and noise reduction I get crucified, not for me I have bigger things to worry about at the moment...

OK, then show us some examples of this expertise that you claim to have in this area!

In what capacity/role are you working on these products?

What measuring equipment are you using?

As I asked before - give us examples of a USB product that didn't pass the noise testing & how you went about fixing this?

Link to comment
I tend to focus mainly on building speakers as these have the most influence of the resultant sound at my ears... My current ones are OB's everything above about 60Hz through the Fostexes with the 2x15 inch per side driven by a SS amp active crossover, also did a pair of curvy changs, a few back loaded horns some standard 3 way designs over the years... Now I just keep things simple, basic PC running squeezebox server and nowt' else and squeezebox touches for streaming... Sold or gave away a lot of gear such as Kef 102/4s, micromega CD palyer, marantz CD63 audiolab stuff etc. Funny when I got more into blind testing the less I am concerned now about the minutiae, but we all choose our own path, mine now is I realise how easily our perceptions can be fooled so dont take anything for granted.

:)

Bye

Marc

 

Funny that you post a picture of a speaker that shows a Fostex drive which has been ENabled (those painted spots around the drive's baffle & whizzer). Painting these spots on the drive is claimed to improve the sound. Have you experienced this?

Link to comment
You seem to be confused about what isolation is which is somewhat surprising considering the image of expertise you portray on this forum.

 

So let me give you the basics (which I assumed you would know) the ADUM is an ISOLATOR IC - it electrically isolates the USB communication but only works at USB full speed & not at USB high speed so it is useless for most current USB audio devices. The concept of isolation moats are only applicable when an isolation chip is used

 

The USB2412 is a USB hub IC - it is a central messaging hub which routes USB signals through it's USB ports - NO ISOLATION is involved, no moats will change this

 

The 5V USB Vbus is not material to what is being discussed here so your pi filter has no use here. BTW, the ground plane moats are electronic basics & not the sort of exciting news you seem to be suggesting!

 

I'm afraid that you are displaying a complete lack of knowledge in this area. Despite the fact that I already told you an ADUM was not applicable you continue to state what you did & compound the misinformation by suggesting that a USB2412 chip could be substituted for the ADUM chip.

 

I'm sorry but I have to ask this - do you really work in the field you are purporting to work in? As your displayed knowledge is so lacking in the basics that I'm beginning to doubt this claim.

 

 

 

OK, then show us some examples of this expertise that you claim to have in this area!

In what capacity/role are you working on these products?

What measuring equipment are you using?

As I asked before - give us examples of a USB product that didn't pass the noise testing & how you went about fixing this?

 

Why don't you read what I said instead of following this path you have decided on and stop trying to belittle me and call me a liar and patronising me in your reply.

As you don,t seem to be able to read without twisting what I said I shall reiterate, the design was to show the isolation moats at the input and output of the board crossed by pi filters, the circuitry in the middle can be what you want it does not matter, it is the EMC (noise) isolation I was trying to show...

As explained earlier everything I work on is covered by NDAs or other resirtictions I am just giving some ideas from the world of electronics I work in, you can check me out on line I am sure currently work at Quadra Solutions and I am on Linkedin, much better than calling someone a liar, just because I don't follow your viewpoint. Funny many clients use me because of my experience and knowledge in layout for signal integrity, low noise etc maybe there fools, they must be because they keep coming back.....

 

Anyway more things to worry about in life than being patronised... my true thoughts would get me banned...

The Fostex drivers I got at a very good price from Planet 10 as I only have one pair of the 208s I cannot compare them so cannot give an opinion on the effects of enable, I enjoy the speakers as a whole and they look funky so I am happy with my lot.

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...