|
|
Post by brubacca on Jul 7, 2016 17:00:47 GMT -5
|
|
KeithL
Administrator 
Posts: 10,273
|
Post by KeithL on Jul 7, 2016 17:07:10 GMT -5
A would tend to agree - wired Ethernet rarely has issues, unless you're trying to listen to music while doing something else that uses a really huge amount of bandwidth. (Anyone who uses that sort of bandwidth probably knows when they;re doing it  .) I would still say that wireless is somewhat more variable.... many people never have trouble.... but it still doesn't work terribly well in some locations (or situations). Now that Ethernet has become more reliable than it used to be a few years back, I don't think you will experience any dropouts with any of the options you listed, at least not if the hardware is compliant and built within reasonable quality limits. (Even Wireless Ethernet proves to be stable and reliable between my Asus RT-N66U wireless router and the built-in Wi-Fi adapter of my Astell & Kern AK240). |
|
|
|
Post by melm on Jul 7, 2016 17:08:47 GMT -5
Interesting.... although I'll admit to being somewhat perplexed..... The Aurender appears to have some very nice features. However, since it claims to deliver a bit-perfect output, and requires an external DAC, I'm wondering how it can make those bits sound better... (If you were using S/PDIF, then jitter might be an issue, but the USB inputs on everything you've mentioned are asynch, and so should be more or less immune to jitter.) I recently bought an Aurender N100H. It completely replaces a PC or a Mac as a streamer. It has an internal 2 TB drive, and an SSD drive for caching if you use a NAS for a source. It's pricey, but the sound is sublime. Even ripped CD's (ripped lossless) sound better than playing the actual disc. I do have an Oppo BD-103 I've used for playback of high Rez files, and there's no question the Aurender outperforms it sonically. The Aurender does require a DAC and then off it goes to my XMC-1. I use a PS Audio NuWave which handles both PCM and DSD which are supported by the Aurender. It does require an iPad for control, and their app is terrific and easy to use. Worth a look. Mel I'm not much of a believer in "audiophile magic" but the Aurender really sounds great. Maybe it's partly the different DAC from the Oppo, maybe something else. But with the usual caveat of my room, my equipment (magnepan 3.6 speakers, Emotiva electronics) it's an astonishingly fine musical presentation. |
|
|
|
Post by yves on Jul 7, 2016 17:21:47 GMT -5
Also, 5. It is a philosophy.  And, from the number of magazine articles published about it in the past few months, and how excited the "true believers" seem to be getting, I think it might be shooting for... 6. A religion (or maybe a cult).  7. Magazine Quibble Accumulated?  |
|
|
|
Post by yves on Jul 7, 2016 17:51:36 GMT -5
A would tend to agree - wired Ethernet rarely has issues, unless you're trying to listen to music while doing something else that uses a really huge amount of bandwidth. (Anyone who uses that sort of bandwidth probably knows when they;re doing it .) I would still say that wireless is somewhat more variable.... many people never have trouble.... but it still doesn't work terribly well in some locations (or situations). Now that Ethernet has become more reliable than it used to be a few years back, I don't think you will experience any dropouts with any of the options you listed, at least not if the hardware is compliant and built within reasonable quality limits. (Even Wireless Ethernet proves to be stable and reliable between my Asus RT-N66U wireless router and the built-in Wi-Fi adapter of my Astell & Kern AK240). Dropouts resulting from having a huge amount of ethernet bandwidth usage are actually not that hard to avoid. The socalled "VIP port" on an Asus ethernet switch can be used as a possible solution that works; IMO it is much more reliable than QoS, and another good alternative IMO is NetLimiter. Asus 10/100/1000 Mbps ethernet switches that feature a VIP port: 5-port www.asus.com/Networking/GXD1051_V38-port www.asus.com/Networking/GXD1081_V3 |
|
|
|
Post by Gary Cook on Jul 7, 2016 18:26:50 GMT -5
Every time I read a "streaming" thread I need a dictionary for the TLA's and it makes my brain ache. I have an Apple TV, a MacMini, a couple of Airport Extremes and iTunes handles the talking with the iPhones and iPads. Listening to the music I really don't want to know know about "Ethernet Switches" (I can work a light switch), "VIP Ports" (although the VIP lounge at the airport is nice), "S/PDIF" (what's the difference), "NAS" (there's an "A" missing I think) and the only AK I know is an AK47, is an KA240 a later model?
Cheers
Gary
|
|
|
|
Post by yves on Jul 7, 2016 18:50:02 GMT -5
Every time I read a "streaming" thread I need a dictionary for the TLA's and it makes my brain ache. I have an Apple TV, a MacMini, a couple of Airport Extremes and iTunes handles the talking with the iPhones and iPads. Listening to the music I really don't want to know know about "Ethernet Switches" (I can work a light switch), "VIP Ports" (although the VIP lounge at the airport is nice), "S/PDIF" (what's the difference), "NAS" (there's an "A" missing I think) and the only AK I know is an AK47, is an KA240 a later model? Cheers Gary Well, an Ethernet switch is not an especially sophisticated device. All it really does is move network data packets from one network cable on to the next; there aren't even any settings that you can apply.  |
|
DYohn
Emo VIPs 
Posts: 18,494
|
Post by DYohn on Jul 7, 2016 19:07:41 GMT -5
7. Like I said, flash in the pan.  |
|
|
|
Post by garbulky on Jul 7, 2016 19:32:21 GMT -5
The real problem for anything other than Coax, TOSLINK or optical, and UBS is that the other inputs (AES, BNC) are hard to find a proper transport for. The only things I know that have outputs for them are Emotiva CD players which have AES. On the PC side specailized pro-sound cards with hideous 32 channel break out cable snakes which would make even a man run for dear life on the aesthetic side of it. It's not WAF anymore it's PAF - people acceptance factor!
However I would put USB last. And I know yves will probably stick pins in my ears for that statement. Coax gets my vote .
As for the drop out question - unless you are currently experiencing audible drop outs, it's not a problem.
|
|
|
|
Post by novisnick on Jul 7, 2016 19:53:03 GMT -5
The real problem for anything other than Coax, TOSLINK or optical, and UBS is that the other inputs (AES, BNC) are hard to find a proper transport for. The only things I know that have outputs for them are Emotiva CD players which have AES. On the PC side specailized pro-sound cards with hideous 32 channel break out cable snakes which would make even a man run for dear life on the aesthetic side of it. It's not WAF anymore it's PAF - people acceptance factor! However I would put USB last. And I know yves will probably stick pins in my ears for that statement. Coax gets my vote . As for the drop out question - unless you are currently experiencing audible drop outs, it's not a problem. Your begging the question! And Im not all that,,,,,so,,,,,,why don't you care for USB? |
|
DYohn
Emo VIPs
Posts: 18,494
|
Post by DYohn on Jul 7, 2016 19:59:57 GMT -5
I use USB from streamers to DACS with no issues. I also use AES and Coax from transports. Which sounds better? Well it's apples and oranges comparing a transport to a streamed file, but I will say they all sound excellent to me.
|
|
|
|
Post by garbulky on Jul 7, 2016 20:26:10 GMT -5
So far the results have been lackluster with USB to my ears namely in the treble. I did try with my DC-1 which has asynchronous USB and what I assume is a decent reciever chip. Not ruling it out but so far I haven't heard any convincing USB presentations.
|
|
|
|
Post by Boomzilla on Jul 8, 2016 7:12:52 GMT -5
Thank you kindly for the digital audio primer & interface rundown, Keith. It's appreciated.
Now my Oppo BDP-105 has not just one but many of the interface options. Am I correct in assuming that provided I'm getting (what sounds to be) a dropout-less audio stream that one interface is as good as another? The most convenient option for me, currently is wired DLNA over Ethernet since my server computer is in the back of the house and the audio system in the front.
Also, should I opt to experiment with other (non-Oppo) DACs, few have Ethernet inputs. Therefore, I'd need to stream to the Oppo via DLNA over Ethernet, and then use the Oppo's digital coaxial output to feed the DAC. Would this conversion (Ethernet-DLNA to digital coaxial) cause any signal degradation?
And lastly, is there any current streamer on the market that meets ALL the criteria I listed in a previous post?
|
|
KeithL
Administrator
Posts: 10,273
|
Post by KeithL on Jul 8, 2016 9:15:14 GMT -5
That's what I do...... There are plenty of small, low cost streamer devices like Roku, and Apple TV, and even Chromecast... which do one simple thing pretty well. Then there are a bunch of expensive streaming servers and clients that include hard drives, and nice little OLED screens, and a bunch of features. While many of the latter look cool, they tend to be expensive, sometimes VERY expensive, and I'm not at all clear on what justifies the prices on many of them. If the goal is to provide a digital data stream, then "sound quality" isn't really an issue - because that depends almost entirely on the DAC. And, to be blunt, I can get a pretty powerful laptop computer for $300 or so, with a nice big high-res color screen.... (I like Windows; MacBooks cost a bit more, but a Mac Mini is also pretty cheap.) And a computer will run whatever player program I choose, connect to a wide variety of storage devices, and pretty well do everything I could possibly want.... I simply can't imagine why I would spend twice - or ten times - as much for a device with less capability and a smaller screen. I use either an Apple TV 4 or a Roku 3 for video streaming but for audio, I'm intentionally using a laptop so I can utilize products like Amarra, Amarra SQ+, Amarra for TIDAL, PureMusic, Audirvana and more. I find these products enhance my listening experience greatly. |
|
KeithL
Administrator
Posts: 10,273
|
Post by KeithL on Jul 8, 2016 9:42:58 GMT -5
In very basic THEORY, all digital audio should sound exactly the same (as long as it's bit-perfect) - because "numbers is numbers - and they're either right or wrong". The only place where there should be any room for differentiation should be the DAC itself. However, at "the next level", it is true that the timing at which the numbers are converted is also critical.... If the correct numbers are converted, but the clock timing is flawed, then the analog output will be flawed as well (with some interesting and difficult to quantify distortions). Therefore, there's room for the "quality" of the digital signal arriving at the DAC to make a difference - both the quality of the source and of the cabling. However (again), while this was an issue in the past, most modern DACs use one of various methods to "fix" the clock on the signal, and so should be immune to this as well. Computer USB outputs are notorious for having a very high level of jitter (clock "errors"), and old style USB DACs were notorious for being very fussy. But almost all modern USB inputs on DACs are asynchronous - which means that the DAC controls the data clock - and those should be almost perfectly immune to this. However, until recently, the Coax and Toslink inputs on many DACs were still sensitive to it; and those outputs on source devices vary in terms of quality (and cabling can also add jitter). However (yet again), many modern DACs use an ASRC to replace the clock on ALL of their inputs, including Toslink and Coax, with a new and undamaged one - which should eliminate this as well. (And some higher-quality older DACs use other methods of "repairing" the clock - like PLLs - which, while somewhat less effective, can still work pretty well if done properly.) "At that level of theory", for example, a USB DAC with an asynchronous input, or a Coax or Toslink input whose signal is passed through an ASRC, should be totally immune to ANY differences in the source or cabling. Incidentally, the USB inputs on all of our current DACs, including the USB Stream input on the XMC-1 are asynchronous; and the XDA-2 and DC-1 have ASRCs (which can be enabled or disabled). The reality is that none of these mechanisms is absolutely 100% perfect - and there are totally unrelated issues (like the possibility of analog noise creeping in over the ground or power connections), which also leaves a tiny bit of room for there to be slight differences. However, I'm inclined to go with your final suggestion..... The DAC itself is what has the most significant effect on the sound, and different DACs do indeed sound quite different. And, while some people obsess over this or that DAC CHIP; the associated analog circuitry, and the overall design of the device, are at least as important as what chip you use. (Arguably, most good DACs are awfully close to perfect; the differences are small and quite subtle... but they are there.) So, yeah, in order to fairly compare different sources, you really need to do so using the exact same DAC... with the same type of connection. And, as I said, the Aurender offers some very useful options - including a few unique ones. For one thing, they seem to offer the ability to connect an external UAC2 DAC (which is something any computer can do, but most "media devices" cannot). They also claim to have paid specific attention to clean ground and power - which are quite important to many devices - and which tend to be largely ignored by computer designers. (In other words, they do seem to have thoroughly optimized their design to work well with a wide variety of external DACs - which is good.) Interesting.... although I'll admit to being somewhat perplexed..... The Aurender appears to have some very nice features. However, since it claims to deliver a bit-perfect output, and requires an external DAC, I'm wondering how it can make those bits sound better... (If you were using S/PDIF, then jitter might be an issue, but the USB inputs on everything you've mentioned are asynch, and so should be more or less immune to jitter.) I'm not much of a believer in "audiophile magic" but the Aurender really sounds great. Maybe it's partly the different DAC from the Oppo, maybe something else. But with the usual caveat of my room, my equipment (magnepan 3.6 speakers, Emotiva electronics) it's an astonishingly fine musical presentation. |
|
KeithL
Administrator
Posts: 10,273
|
Post by KeithL on Jul 8, 2016 11:00:50 GMT -5
In principle they should all sound the same. In reality, I've never heard any difference in which input you use on an Oppo, but it is theoretically possible that their implementation of one might work slightly better or worse than another, and they may not offer identical options on each. (For example, our XMC-1 isn't going to sound different because you use the USB Stream input or an HDMI input to play your digital audio, but our HDMI input supports DSD, while the USB Stream input doesn't.) The actual conversion process should be lossless.... and, because anything sent over a network is packetized, by the very nature of network traffic, ANY audio signal sent over a network MUST be re-clocked, so things like jitter are pretty much under the control of the receiving device. From what I know about DLNA, it is capable of transmitting a lossless signal, so it should work fine. (I would check the documentation of the specific DLNA server you're using to make sure that they don't alter the signal before transmitting it.) Just to be clear, to me there are two "grey areas" about DLNA. The first is that the signal that gets sent is under the control of the server that's sending it. Many DLNA servers accept a variety of types of input signals and files, and some of them may transcode (convert) everything they take in to some "standard" format before transmitting it. In other words, just because "DLNA" supports 24/192k doesn't mean that the specific DLNA server you're using does, or that it wasn't written to convert everything it receives to 16/44k before sending it. I would also check that there isn't some sort of configuration setting that configures it to "send everything as received" or not. (There are various reasons why a server might do this. For example, early versions of Squeezeserver converted everything they sent to 44k by default.... The reason was that the client it was designed to work with only supported 44k, so, to ensure that it would play whatever you told it to play, their server software was configured by default to down-sample everything it received at higher sample rates to 44k before sending it. Others may offer this option to conserve network bandwidth, or because, on a slow network, a slower sample rate would be LESS likely to experience data dropouts.) The second area in which I'm not fully informed (and I haven't found the information so far) is in what DLNA does when a data loss occurs. In other words, if I'm copying a file to a network server hard drive, and the network stops for a split second, the server will simply wait for the next packet; and, eventually, it will give up and issue an error - but it will never write a corrupted file. In contrast, with many streaming services, or your cable TV, if there's a "stoppage", the system will do the best to avoid a dead stop.... the video may become pixellated, or the audio may "fall back" to lower quality. Now, personally, with a TV show ,I'd rather the show keep playing, even with a few flickers; but, with audio, I'd rather a dead stop, or a dropout. (I haven't seen this spelled out anywhere, and I haven't played with DLNA enough to have experienced it either way.) However, all that said, a lot of people use it, and it seems to work pretty well, so it's almost certainly your best bet considering your requirements. As for streamers, I'm really not that familiar with all the current options...... All of the cheap ones I've researched have had limitations that reduce their appeal to me.... If I were doing a streamer I'd probably go with a Raspberry Pi running XBMC.... which DOES do Ethernet. It's quite cheap, and very powerful, but there is "some assembly required". Thank you kindly for the digital audio primer & interface rundown, Keith. It's appreciated. Now my Oppo BDP-105 has not just one but many of the interface options. Am I correct in assuming that provided I'm getting (what sounds to be) a dropout-less audio stream that one interface is as good as another? The most convenient option for me, currently is wired DLNA over Ethernet since my server computer is in the back of the house and the audio system in the front. Also, should I opt to experiment with other (non-Oppo) DACs, few have Ethernet inputs. Therefore, I'd need to stream to the Oppo via DLNA over Ethernet, and then use the Oppo's digital coaxial output to feed the DAC. Would this conversion (Ethernet-DLNA to digital coaxial) cause any signal degradation? And lastly, is there any current streamer on the market that meets ALL the criteria I listed in a previous post? |
|
|
|
Post by qdtjni on Jul 8, 2016 11:02:15 GMT -5
Now, I also noticed that it supports several "modes", like Squeezeserver, which most definitely are NOT bit perfect. Where did you get that idea because it is simply false. |
|
|
|
Post by Boomzilla on Jul 8, 2016 11:05:06 GMT -5
...(I would check the documentation of the specific DLNA server you're using to make sure that they don't alter the signal before transmitting it.) Just to be clear, to me there are two "grey areas" about DLNA. The first is that the signal that gets sent is under the control of the server that's sending it. Many DLNA servers accept a variety of types of input signals and files, and some of them may transcode (convert) everything they take in to some "standard" format before transmitting it. In other words, just because "DLNA" supports 24/192k doesn't mean that the specific DLNA server you're using does, or that it wasn't written to convert everything it receives to 16/44k before sending it. I would also check that there isn't some sort of configuration setting that configures it to "send everything as received" or not. (There are various reasons why a server might do this. For example, early versions of Squeezeserver converted everything they sent to 44k by default.... The reason was that the client it was designed to work with only supported 44k, so, to ensure that it would play whatever you told it to play, their server software was configured by default to down-sample everything it received at higher sample rates to 44k before sending it. Others may offer this option to conserve network bandwidth, or because, on a slow network, a slower sample rate would be LESS likely to experience data dropouts.) The second area in which I'm not fully informed (and I haven't found the information so far) is in what DLNA does when a data loss occurs. In other words, if I'm copying a file to a network server hard drive, and the network stops for a split second, the server will simply wait for the next packet; and, eventually, it will give up and issue an error - but it will never write a corrupted file. In contrast, with many streaming services, or your cable TV, if there's a "stoppage", the system will do the best to avoid a dead stop.... the video may become pixellated, or the audio may "fall back" to lower quality. Now, personally, with a TV show ,I'd rather the show keep playing, even with a few flickers; but, with audio, I'd rather a dead stop, or a dropout. (I haven't seen this spelled out anywhere, and I haven't played with DLNA enough to have experienced it either way.) However, all that said, a lot of people use it, and it seems to work pretty well, so it's almost certainly your best bet considering your requirements... Thank you kindly AGAIN, Keith - I'm slooooowly edumocating myself about the technology that I use. Cordially - Boomzilla |
|
KeithL
Administrator
Posts: 10,273
|
Post by KeithL on Jul 8, 2016 11:10:55 GMT -5
Ignoring the others, an Ethernet switch is basically a junction box for network connections (like a USB hub for Ethernet). Excluding the finer points.... like the differences between hubs, switches, and routers.... and managed switches.... You use one when you want to connect several devices to a single output, and you can also position extra ones "strategically" to reduce network traffic in specific spots. They also act like a repeater, so you can use a switch to connect two really long cables to go even further. Switches, even little ones, used to be uber expensive (as in $500 up)... but now a cheap one, which actually works very well, can be had for under $10. And, in case you were wondering... what most people call a cable modem is actually (all in one little box): 1 wired router 1 switch 1 WiFi router (usually) 1 cable modem 1 DHCP server Every time I read a "streaming" thread I need a dictionary for the TLA's and it makes my brain ache. I have an Apple TV, a MacMini, a couple of Airport Extremes and iTunes handles the talking with the iPhones and iPads. Listening to the music I really don't want to know know about "Ethernet Switches" (I can work a light switch), "VIP Ports" (although the VIP lounge at the airport is nice), "S/PDIF" (what's the difference), "NAS" (there's an "A" missing I think) and the only AK I know is an AK47, is an KA240 a later model? Cheers Gary |
|
KeithL
Administrator
Posts: 10,273
|
Post by KeithL on Jul 8, 2016 11:42:27 GMT -5
Now, I also noticed that it supports several "modes", like Squeezeserver, which most definitely are NOT bit perfect. Where did you get that idea because it is simply false. Really..... It supports Squeezeserver - which may or may not be bit-perfect - depending on how it's configured. It also allows connectivity with "streaming services" (via other sources) - MOST of which are lossy. And they did mention "Internet radio" - ALL of which is lossy. And it accepts streams from Roon; and a stream from Roon may or may not be bit-perfect - depending on where Roon got it from. And I'm pretty sure the same holds true for Logitech's media server (still). (And, oh yeah, HQPlayer also resamples your audio data, depending on how it's configured - which is a major part of its purpose.....) Therefore, it does indeed support a wide mix of sources - some of which are lossy. (And, since it's a CLIENT, it has no control over what happens to the content at the server, before it gets it.) However, please note that none of this is an indictment of the Sonic Orbiter per-se.
In fact, flexibility and the ability to play audio from a huge variety of sources is a good thing.My point, in the context it was stated, was that, when I play a FLAC or WAV file off my hard disk, I know where it came from and whether it is bit-perfect or not. It came from my hard drive, where I put it after ripping it from an original CD, and I KNOW that a file played on FooBar2000 via a WASAPI connection from my computer is bit-perfect. (And my ripping software uses AccurateRip to confirm that the rip itself is bit-perfect.) But, when I play a song through something like the Sonic Orbiter, it can be rather complicated to figure out the actual provenance. (Unless you very carefully cripple all that flexibility to exclude everything except files on your own hard drive it's more like playing a file you downloaded....) For example, if the Sonic Orbiter is playing something that originated on Spotify, my DAC connected to it is going to show it as PCM (because it was already decoded upstream). This is the same as if I were playing it on a computer; EXCEPT that, if I have Spotify running on my laptop, connected to my DAC, I can see that it's the Spotify client running on the computer. As a "source consolidator" the Orbiter cannot avoid obscuring some of this sort of information. It's simply the price of convenience... And, to be fair, I suspect that the Sonic Orbiter isn't going to alter any bits by itself (it doesn't have an ASRC, right?) However, I would be much more comfortable if they actually SAID that as if they considered it to be something important.... rather than leave me to assume it. As in: "Our box will always send the bits it receives directly to your external player - and never alter them". (Since that's probably the single most important feature to me, it seems odd that they didn't actually mention it - in screaming headlines.) |
|
.)
