• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Review and Measurements of Paradigm PW-Link Streamer/DAC

invaderzim

Senior Member
Joined
Aug 30, 2018
Messages
320
Likes
502
Location
NorCal
I finally fired up my replacement PW-Link.

Occasionally at the start of a different album it will stutter, it sounds like it is playing a bit from forward in the song then within a fraction of a second starts it correctly. It does it less than the other one did so I'll run it for awhile and see what happens. Overall the sound is so much better with it in place that I hope it behaves most of the time.

Here's the graphs from my main system
arc.jpg
 

audimus

Active Member
Joined
Jul 4, 2019
Messages
154
Likes
109
Well, so much for the axiom in these forums, “if you can hear it, you can measure it”. :) Clearly, the pops were not measured.

I am being facetious, but it does have a lesson of the scope of the measurement methodology. You cannot measure what you have not set up to measure. In the lingo of formal systems, these measurements are accurate but not necessarily complete.

Anyway, I ordered one of these for experimentation as I have a setup to vary a lot of the input parameters to see if I can narrow down to what is causing these transitional pops. Will post if I find anything useful. Just getting a calibrated mic to test measure speaker set up with ARC is worth it to me at this price.

But I do have a suggestion for the measurement methodology in general if practically feasible based on the experience of this.

It might be useful to do future tests at two different level outputs - a ref 2v and a lower output because a device might be perfectly usable and even superb at low levels but may not measure well at ref levels. My receiver, for example, has a setting to adjust the input levels for each analog source. The options are Low (1v), Ref (2), High (4). If there is a DAC that is good or usable at a low level and performs well and/or has other redeeming features, do I really care if it does not measure well at ref levels? If this unit did not have ARC, it would have been panned as a badly engineered product as many other DACs that measure poorly. But would some of them be perfectly fine at lower output levels?
 

invaderzim

Senior Member
Joined
Aug 30, 2018
Messages
320
Likes
502
Location
NorCal
...Anyway, I ordered one of these for experimentation as I have a setup to vary a lot of the input parameters to see if I can narrow down to what is causing these transitional pops. Will post if I find anything useful. Just getting a calibrated mic to test measure speaker set up with ARC is worth it to me at this price....
Definitely keep us posted if you find anything that reduces the number or volume of the noises. I figured I'd start playing around with different settings and formats to see if anything helps.

I agree about the mic part of the value and I think I'm going to put a miniDSP OpenDRC-DI on my Christmas list this year. It looks like the least intrusive option by staying digital. Although I must say I'm not really impressed with their "you bought it, it is yours" return policy. I will also look into the Behringer DEQ2496

In the end it was a reasonably affordable way to test room correction.

Clearly, the pops were not measured.
I'd love to see a poll of who is getting noises and who isn't.
And what the noise is they are getting. Mine sound like a fraction of a second burst of the start of the song before the song starts. Sometimes it is so short that it just sounds like a click and others it sounds more like it skipped ahead a second and then back; but even at the longest it is about a quarter of a second at most. I don't tend to hear the noise on music that starts out really soft, I'm guessing if it is jumping ahead it is getting such a low volume section I'm not hearing it.
 
Last edited:

pjug

Active Member
Patreon Donor
Joined
Feb 2, 2019
Messages
137
Likes
82
So, if you are able to solve the issue with digital that's great! I hope someone here provides a solution. But if you are not, I think you should consider using the analog input. The improvement that sophisticated room compensation makes in most rooms is tremendous...and the difference between using digital or analog inputs is tiny (or perhaps non-existent) in comparison. The PW/Link will re-sample any digital signal fed to it. So you are not getting a bit-perfect transfer either way.
I have been sour on the idea of going D-A-D-A but you convinced me to try this. So far I think you are right; it sounds great and nice not to have to worry about the pops.

Still I hope the manufacturer offers a fix or someone figures out another way around the problem that allows this to work as a pure digital solution.

Edit: I should mention the other reason I went to the analog input. I was listening to music streamed with deezer through Bluesound node to the PW Link. Network traffic or something caused a long dropout, and then I got a very LOUD snap, much louder than anything I've had previously. Fortunately I am playing through an integrated amp and the volume was at a moderate level and no harm was done. But I think the PW Link can possibly do harm to your speakers. Not likely, but I thought it was worth mentioning.
 
Last edited:
Joined
Jul 7, 2019
Messages
7
Likes
1
I have been using mine optical in and out. As I reported earlier, I have the pops (which are relatively innocuous IMHO as they are very quiet), as well as the startup glitching. I mentioned that I thought there was a buffering issue, but I have more thoughts since then. Keep in mind that it's pure speculation on my part:

A couple of decades ago, I was interfacing/programming a DSP and I encountered a very similar sounding clicking. After discussing it with the chip designer, he recommended avoiding muting and unmuting unless absolutely necessary. So I added a delay, and voila, no clicks (unless it timed out of course).

I'm guessing the PW-Link has a similar issue, and in addition has a sizable buffer that it needs to process the convolutions. The mute behavior and the buffering behavior are probably interacting to cause the startup glitch. If that's true, the software implementer has the choice between unmuting the playback asap (hoping for the buffer to be full or filled with start-of-track silence) or delaying the playback causing a lack of responsiveness.

What's also interesting is that Tidal played through my Squeezebox has in-between clicks, and Tidal played through the Play-Fi app only has startup clicks. And I don't recall hearing the startup glitch via Play-Fi. My guess is that the Play-Fi app plays better with the DSP's API, and that the Squeezebox implementation does not.

Also, one other interesting datapoint is that I believe a Roon user reported updating his/her player software, and the clicks disappeared.

In any case, I am guessing that a player-side software update is where it's likely going to fix the issue, and that a firmware fix might do things like mess with gapless playback (again a buffering implementation issue). All that being said, I am living with the clicks, as they don't bother me that much.
 

pjug

Active Member
Patreon Donor
Joined
Feb 2, 2019
Messages
137
Likes
82
And I don't recall hearing the startup glitch via Play-Fi. My guess is that the Play-Fi app plays better with the DSP's API, and that the Squeezebox implementation does not.
I'm not sure whether you mean the pops or the other startup glitch, but I do remember getting pops with Play-Fi, even though I did not spend much time using it.
 
Joined
Jul 7, 2019
Messages
7
Likes
1
I was mainly referring to the pops, and to be clear: on Play-Fi, the pops will occur on start and stop, but not between tracks. If you are interested, I would get the latest Play-Fi app (there has been at least one recent update in the last month or so), and try it.
 

pjug

Active Member
Patreon Donor
Joined
Feb 2, 2019
Messages
137
Likes
82
I was mainly referring to the pops, and to be clear: on Play-Fi, the pops will occur on start and stop, but not between tracks. If you are interested, I would get the latest Play-Fi app (there has been at least one recent update in the last month or so), and try it.
Thanks. I'll give it a shot but unless the Pay-Fi app is very much improved I don't think I would want to use that. Also, if there are any pops at all I think I'd rather stay with the analog input so that I don't get a repeat of what happened as a result of the streaming service dropout.
 

invaderzim

Senior Member
Joined
Aug 30, 2018
Messages
320
Likes
502
Location
NorCal
Just got mine
Setup is a breeze
But...

The volume issue is strange.

Sometimes the sound is really compressed and bad.

I’m doing digital in from cca, digital out to a dac.

Please help!
Have you tried lowering the volume on it? Mine had a bad pulsing volume level like someone had speakers on the end of a rope and were swinging them around. I turned the volume down a half dozen clicks and it fixed that.
 

invaderzim

Senior Member
Joined
Aug 30, 2018
Messages
320
Likes
502
Location
NorCal
I've found specific songs that it does the noise on and others that it doesn't. So far the ones it doesn't make the noise on that I've found were purchased as hi-res Flac files and the ones it does tend to be ones I've ripped from CDs. When I had the original one I hooked it up to a disk player and it did it with CDs too; I haven't tried it with this one yet.

Any ideas on what would be causing it or other details to look at on the different files?
 

audimus

Active Member
Joined
Jul 4, 2019
Messages
154
Likes
109
Check the sampling rate and bit depth on the files and see if you notice any correlation with the noise at the beginning.
 

invaderzim

Senior Member
Joined
Aug 30, 2018
Messages
320
Likes
502
Location
NorCal
What do I use to see those details for the file.

File explorer will show bit rate but I don't see anything else there. The ones that weren't doing it yesterday were all higher bit rate ones but that might just be a coincidence.

*** Update *** it looks like VLC will show me file details so I'll try different ones tonight and make notes on which ones make the noise and which don't.

I need to compare some other purchased but lower quality files to see if they do it.

The ripped files that do it vary from wav to mp3 to flac.


On a side note, even with the noise I'm listening to music a lot more with the PW-Link in the chain.
 
Last edited:

invaderzim

Senior Member
Joined
Aug 30, 2018
Messages
320
Likes
502
Location
NorCal
I couldn't find any patterns to the sampling rate and bit depth, the only difference I found is the tracks that were downloaded were much less likely to have any noise compared to tracks ripped from disks.

I also tried the analog in/out connection and I got a distorted sound like the volume was over cranked but it did it at any volume setting on the PW-Link.

I'm listening without it right now and I do miss the sound improvement from when it was connected.

I bought a used Behringer Ultracurve Pro DEQ2496 but skimped on the research and I don't have a microphone that will work with it, I don't know if it is something that amirm would want to test or not but I won't likely be buying a mic for it for some time.
 

audimus

Active Member
Joined
Jul 4, 2019
Messages
154
Likes
109
The other difference could be the presence or absence of silence or very low levels at the beginning of tracks. Will be evident if you load the file in something like Audacity. This can be tested using samples with pre-gapping with silence, with low level initial noise and snipping out any initial silence or low level using the same track and audio editing software like Audacity. Some players also have settings to do some of the above.

There are a number of variables here. If playing through a PC, entire chain from the first component that reads the file to the audio drivers need to be looked at because sample rates and bit depths can change anywhere in the chain. Don’t know if the PW link interface displays what is arriving and what is output. Will have more when my unit arrives and I do some testing.

From my experience, the startup and between tracks pops/noises can be from one or more of the following:

1. Resampling problems at the device for what it receives to what it needs to do to output. If, for example, it does its processing at 48khz 24 bits, can output at that setting and receives the same, a resampling may not be needed. Testing with controlled upstream and downstream rates and depth could eliminate this possibility. Change in sampling rates between tracks with or without gap between tracks could also trigger this if there is a problem with the device.
2. Buffering issues on the device. The ability of the device to fill its buffers properly before processing could trigger noise or pops at the start before things get in sync. This could be tested by using playlists with gaps between tracks as well as playlists with no gaps between (it should manifest itself only at the start in this case). Also, can be tested using players that can output inaudible noise in silent portions or between tracks, so it is seen as one continuous track.
3. Delays in DSP circuitry waking up in response to signal arriving. This could happen even between tracks of where is aggressive power management within the device or some muting mechanism switched on/off in the device to ensure total silence with no signal. This could also be tested by introducing low level inaudible noise always at the beginning and between tracks. Tracks that have a relatively low level start periods may mask this problem over tracks that start with a bang from total silence.

In short, it requires careful, controlled experiments and a full understanding of what is happening end to end to test for only one variable at a time. As long as the problem is deterministic, it should be possible to narrow it down. Hopefully, it is deterministic even if it seems otherwise without nailing down all parameters.
 

invaderzim

Senior Member
Joined
Aug 30, 2018
Messages
320
Likes
502
Location
NorCal
The other difference could be the presence or absence of silence or very low levels at the beginning of tracks. .....
That is the first thing that made it more difficult to tell which tracks caused an issue. When I first connected the new PW-Link I thought it was all better because I wasn't hearing anything but I was listening to styles of music that start out very quiet.

At this point I just used the microphone from it and REW to help move the subwoofer and my speakers a bit and make adjustments on the subwoofer. While it isn't as big of an improvement without the added room correction it is enough that I took the PW-Link out of the chain.

The problem can't be unknown to Paradigm and there is no sign of a fix from them which is still quite disappointing.
 

audimus

Active Member
Joined
Jul 4, 2019
Messages
154
Likes
109
Ok, I finally got to play around with it. What a mess of a product.

I would not recommend it for any of its intended uses because while it kind of works like a undergraduate class project, the amount of glitches, the sorry state of the PlayFi software from DTS (who should be ashamed of putting out such an app, it is beyond bad) but necessary for its intended use is a showstopper. This was a failed Sonos competitor and why they are clearing it out. There are also significant security gaps in the way this is initially configured and it has an unprotected web interface where the firmware can be updated. So, make sure that this unit is not ever used in any public, shared or compromised networks. This is a common problem with many of these IoT units.

However, I would recommend this product as a S/PDIF inline DRC unit with pretty good Anthem Room Correction to play around with. The results are pretty good and it is the cheapest one at this clearance price with a decent DRC and an included calibrated mic and is far more user friendly than REW to get setup quick.

As to the noise between tracks, I narrowed it down to non-muting or perhaps delayed muting of the S/PDIF output while syncing up to the beginning of the input stream.

This is a problem that has plagued many early and/or cheap DACs and AVRs but this particular configuration is not encountered frequently because S/PDIF in and S/PDIF out is not a common use case. From what I can see, this unit does mute the analog out for and S/PDIF in and analog in should not have this particular problem of a pop/noise (but may have other issues like chopping off the beginning if the DSP is slow to wake up, I have not tested these configurations). There is some evidence that muting and unmuting is very slow on this machine. For example, if you are using a PC as a source, the short system clicks and other short UI sounds are dropped (it will only play one out of 3 or 4 coming in quick succession).

Of course, use of the analog introduces the use of the DAC (to do DRC) which may introduce noise except at lower output levels as the measurements show.

The solution to this is to use player software that pre-gaps any track (unless it is present as a non-gap playlist which will not have this problem) with a caveat. What triggers this seems to be stopping the input bits for silence and starting again for the next one which requires renewed syncing processing by the unit. However, silence can also be “played”, in other words the input frames are coming in but the content input levels are zero. Even if a new stream starts with this played silence, the pop will happen but it is a silent one. This is why music that starts naturally with silence or low volume seems immune to this problem.

To be fair, I don’t particularly find even a pop that problematic but tracks that start with a bang can be startling because it plays the bits without syncing and restarts after syncing. So YMMV as to its acceptability. Great if you can have player software that ensures silent playing at the beginning with a setting. Different software does this differently.

But once you are past that, it works very well as an ARC room correction box for two channel PCM sound (I will have a post later on how you can do 2.1 correction or SW integration with this unit if you have an AVR or powered speaker that has crossover settings). The PW Amp has SW out and so can do 2.1 and bass management on its own using ARC genesis. Not PW link unless you can do a workaround since ARC genesis sees only two speakers to configure this unit.

Whether they will have an update to fix the muting problem, I doubt it. This looks like an abandoned child at Paradigm for the core firmware but this appears to be built for modular updates.

The PlayFi subsystem can be updated independently by DTS software without requiring a core firmware update and this is done via the app. Similarly, ARC does its own update to ensure compatibility with ARC protocol/api and so is able to support the latest ARC genesis to be used on this.

As an experimental DRC box that is relatively cheap, it has a lot of potential. The calibration files for the included mic for use with REW can be done if one has access to another calibrated mic like the minidsp one that works with REW. But this mic can be used with REW for a number of things that does not require calibration like speaker delay settings, etc. Also works reasonably well in REW for level balancing using pink noise without calibration.

The ultimate experimental goal would be to reverse engineer the protocol between the ARC Genesis and the PW link and spoof an Anthem unit with multiple channels to get the filter definitions for use with any parametric equalizer like a miniDSP or software like Equalizer APO. Purely for educational purposes, of course.

Will have some posts later on how to get the most out of ARC with 2.0 or 2.1 setups using the professional mode to get the tonal balance you like and smoothing out the response across the spectrum.
 

invaderzim

Senior Member
Joined
Aug 30, 2018
Messages
320
Likes
502
Location
NorCal
That is a very good summary of it.

I thought for awhile I could live with the between track noises on mine and did listen to it for hours like that but there got to be too many times where it was startling and that just reduced the enjoyment of listening to the music.
 

audimus

Active Member
Joined
Jul 4, 2019
Messages
154
Likes
109
You must be really sensitive. For those of us that grew up with the vinyl pops and crackles and an occasional shaky hand in placing the cartridge, this is nuttin’. :)

I wouldn’t let this discourage people from trying it out for themselves.
 

invaderzim

Senior Member
Joined
Aug 30, 2018
Messages
320
Likes
502
Location
NorCal
it seems pretty odd to compare what we had available in the 70s for music reproduction to a current add-on component, that is designed to improve the sound, to defend it in an age of nearly perfect music reproduction. Especially on a site that is so down on companies that don't chase perfection. Why not buy all older Schiit gear and say "well it is better than the transistor radio I had in 75". Sometimes the things that get ripped apart for being 'noisy' on here and the things that get a pass surprise me.
I would say everyone should make their up their own mind on it and for that they need to know both sides And the noise level does vary a lot based on what music you listen to. It all depends on how loud the song starts. It is possible to listen to dozens of songs and hear nothing or next to nothing if they start softly but it is also possible to listen to dozens of songs and hear a loud burst before every song if they start at full volume. The first evening with mine I thought it wasn't making noise but the second with different music it was quite loud.
 

audimus

Active Member
Joined
Jul 4, 2019
Messages
154
Likes
109
Chill out. some humor to set the context of how intrusive/loud (or not) this non-muting is. Just pointing out it is possible to overstate the situation with frequent repetition. That is not defending imperfection, it is setting up a YMMV with context. Remember the host of this site recommended this unit in the review because of DRC despite its DAC performance test issues (the testing did not test for the initial sync issues granted). So trade-offs are not unknown on this site for less than stellar test results while “chasing perfection”.

While I have noticed this non-muting issue once in a while, from a couple of hours of listening every day since I set it up with everything from rock to blues to jazz to classical, this has never been an annoyance to me. More of an occasional surprise as you don’t expect it in modern equipment and it was nowhere close to deceasing the enjoyment of the music for me. But this isn’t some $1500 unit nor is there anything that is an alternative at this price. It is $150 or so and the cheapest way to get competent 2-channel DRC and enjoy it. I wouldn’t tolerate this from the former but from the latter...? Perspective and context matters.
 

Similar threads

Top Bottom