|
|
Post by foggy1956 on Mar 17, 2019 17:34:25 GMT -5
Boom, using terms like gobbledy-gook, disingenuous, and downright fraudulent shows me both that you just don't get it or you don't understand what is being said and as in most of your destined-to-be-epic posts you begin with your mind made up and your posting is masturbating in public to satisfy your own ego. Done. I'm done.I'm sensing anger issues |
|
|
|
Post by garbulky on Mar 17, 2019 19:21:58 GMT -5
Oppo is discontinued. No way they are going to keep up. I was impressed they included the Dolby vision update.
|
|
klinemj
Emo VIPs 
Honorary Emofest Scribe
Posts: 14,740
|
Post by klinemj on Mar 17, 2019 19:29:40 GMT -5
Well, THIS gives pause: "RAAT is built to evolve over time. We continue to improve the network protocol. We might decide to change the buffer size requirements on the device to increase stability. We might decide to build a second network protocol optimized for streaming over WAN, or something else like that. We give the same advice for users of Roon as we do to manufacturers building RAAT-based products: under-specced systems lead to bad user experiences; hardware is cheaper than ever and getting cheaper all the time; don't over-economize if you want the best result." This is inherently contradictory gobbledy-gook from Roon. To claim that "hardware is cheaper than ever and getting cheaper all the time" immediately followed by "don't over-economize if you want the best results" is disingenuous at best, and downright fraudulent at worst. If one buys a "cheaper than ever" streamer now for somewhere between $700 and $5,000 - there's absolutely NO guarantee that the manufacturer will periodically update their device to keep up with Roon's changes to RAAT. I've found NO devices (other than DIY kludges) for less than $700 that claim to be Roon-Ready. Does Roon REALLY consider that "cheaper than ever?" I don't. Under the circumstances, Oppo DOES have a track record of updating their firmware (a trend that they've promised to continue) AND despite being primarily a disc player, they already ARE Roon-certified. I can think of no other streamer that would offer any better performance for the cost of my Oppo (already paid for). So my decision is made. Keep the Oppo & use it as a streamer with a higher-quality external DAC. Once I get the external DAC, I will also try the Sonore Microrendu to see if it offers any improvement - but I'm not really expecting much (if any). Considering I can build an R-Pi streamer for about $60 in parts and it's as easy a snap-together model kit, yeah...they are cheaper than ever. You just keep focused on the pricey ones. Mark |
|
|
|
Post by garbulky on Mar 17, 2019 21:13:19 GMT -5
Well, THIS gives pause: "RAAT is built to evolve over time. We continue to improve the network protocol. We might decide to change the buffer size requirements on the device to increase stability. We might decide to build a second network protocol optimized for streaming over WAN, or something else like that. We give the same advice for users of Roon as we do to manufacturers building RAAT-based products: under-specced systems lead to bad user experiences; hardware is cheaper than ever and getting cheaper all the time; don't over-economize if you want the best result." This is inherently contradictory gobbledy-gook from Roon. To claim that "hardware is cheaper than ever and getting cheaper all the time" immediately followed by "don't over-economize if you want the best results" is disingenuous at best, and downright fraudulent at worst. If one buys a "cheaper than ever" streamer now for somewhere between $700 and $5,000 - there's absolutely NO guarantee that the manufacturer will periodically update their device to keep up with Roon's changes to RAAT. I've found NO devices (other than DIY kludges) for less than $700 that claim to be Roon-Ready. Does Roon REALLY consider that "cheaper than ever?" I don't. Under the circumstances, Oppo DOES have a track record of updating their firmware (a trend that they've promised to continue) AND despite being primarily a disc player, they already ARE Roon-certified. I can think of no other streamer that would offer any better performance for the cost of my Oppo (already paid for). So my decision is made. Keep the Oppo & use it as a streamer with a higher-quality external DAC. Once I get the external DAC, I will also try the Sonore Microrendu to see if it offers any improvement - but I'm not really expecting much (if any). Considering I can build an R-Pi streamer for about $60 in parts and it's as easy a snap-together model kit, yeah...they are cheaper than ever. You just keep focused on the pricey ones. Mark I don't think B'zilla wants to tinker around. I know that's what turns me off, figuring out the different parts, chassis, software etc. I don''t want to speak for him but maybe he would be interested in a ready made one by you at a modest price? |
|
|
|
Post by boomzilla on Mar 17, 2019 21:16:43 GMT -5
If / When my Oppo no longer functions as a Roon destination, I'll consider replacing it. For now, however, it is certified "Roon-Ready" by both Roon and Oppo. That's sufficient for me.
|
|
|
|
Post by garbulky on Mar 17, 2019 21:19:28 GMT -5
If / When my Oppo no longer functions as a Roon destination, I'll consider replacing it. For now, however, it is certified "Roon-Ready" by both Roon and Oppo. That's sufficient for me. That's what I'm telling ya. You've always had a perfectly fine solution for ethernet based audio. No need to limit your options to ethernet dacs when you've got an Oppo as a solid transport. |
|
|
|
Post by Casey Leedom on Mar 17, 2019 22:24:28 GMT -5
A couple of items:
Today, RAAT (Roon Advanced Audio Transport) is an Application Layer protocol on top of TCP (the standard InterNet Transmission Control Protocol). TCP is what provides a reliable, sequenced stream with no holes of duplications.
The early RAAT implementations were on top of UDP (the InterNet User Datagram Protocol) which is not reliable, doesn't guarantee sequencing, duplicates, or prevent dropped packets).
The Roon server can work with Roon End Points/Bridges which directly implement RAAT. It can also work with Squeeze Clients. I'm currently using that mode with an old Logitech Squeeze Box Touch. I'm guessing that they probably did this to take advantage of the large installed base of Squeeze Clients in order to ramp up their business.
The Oppo UDP 203 originally had an undocumented Squeeze Client in the Firmware. Now it appears to have a RAAT End Point. It works, but is a little clunky in the sense that you need to power cycle the Oppo if you're using it for Roon and then want to play a Video Disc. At least that was what I had to do the one time I tried it.
Casey
|
|
|
|
Post by boomzilla on Mar 18, 2019 0:18:43 GMT -5
I experienced the same thing.
|
|
|
|
Post by wilburthegoose on Mar 18, 2019 5:43:34 GMT -5
Boom - you're seriously over-complicating a pretty simple concept.
|
|
|
|
Post by wilburthegoose on Mar 18, 2019 5:45:12 GMT -5
Send me $100, and I'll build you a Raspberry Pi with a HiFi Berry hat (a hat is akin to a sound card in a PC).  |
|
klinemj
Emo VIPs
Honorary Emofest Scribe
Posts: 14,740
|
Post by klinemj on Mar 18, 2019 7:28:05 GMT -5
I don't think B'zilla wants to tinker around. I know that's what turns me off, figuring out the different parts, chassis, software etc. I don''t want to speak for him but maybe he would be interested in a ready made one by you at a modest price? I know he doesn't want to build something. I was simply pointing out that, indeed, there is very low priced gear available for streaming. I also previously noted a couple options in the $500-1000 price range (no more than an ordinary DAC) that have ethernet into a DAC and are also Roon ready. It's not all about the super expensive stuff that boom started his premise on then built onto with his more recently comment that I responded to. There are many reasonable priced options and cost of gear for streaming is indeed coming down. Mark |
|
|
|
Post by wilburthegoose on Mar 18, 2019 9:23:58 GMT -5
Yeah - a Chromecast Audio costs about $40.
|
|
|
|
Post by Creature on Mar 18, 2019 17:18:07 GMT -5
Hi,
You have me reading more about DLNA. Here's some of what I've
learned thus far. Feel free to disagree, as my interpretation
of what I'm reading can very well be flawed.
DLNA layers on top of the UPnP Audio/Video extensions. There
are three content transport protocol choices, HTTP (mandatory
inclusion), DASH (optional) and RTP (optional). Apparently,
HTTP invariably uses TCP, though reserved port numbers typically
include both UDP and TCP reservations.
DASH similarly uses HTTP on top of TCP but adds an adaptive
bitrate feature (think adaptive video resolution when streaming
movies over the Internet). RTP apparently is seldom used but
is included as an option because RTP has a multicast feature.
RTP is over UDP, includes timestamps and monotonically increasing
segment numbers, but does not retransmit lost segments. The
optional companion protocol of RTCP enables the receiver to
communicate packet loss and jitter statistics to the sender but
RTCP does not add retransmission. The intent instead is to allow
the sender to adapt in an effort to improve QoS of future
transmission.
I found a white paper written as a thesis by a graduate student.
He characterized the performance profile of DLNA, including the
injection of packet loss (5%, 10% and 15% loss).
It did demonstrate that DLNA reliably transports the streaming
content, by virtue of HTTP using TCP. It also showed that the
content was pretty much just sent to a buffer in the DLNA
renderer as fast as the network allowed.
The HTTP/TCP session between DLNA server and renderer stayed
open during the entirety of the playback but had little to do
after the song was fully buffered in the DLNA renderer.
The 15% packet loss test was a bit of an exception due to quite
a bit of time lost to retransmission of missing TCP segments.
Something larger, like a movie, I'd guess fills the buffer and
then waits for the renderer's buffer to drain a bit before
sending more.
It is true that practical implementations of UPnP/DLNA do not
include synchronization of simultaneous playback on multiple
renderers. RAAT does include such a feature, which sounds
useful for whole home implementations. Another legitimate
argument is that DLNA basically has no (or weak) authentication
and authorization mechanisms. There's also an old IGDP
vulnerability in old (or poor) Internet firewall/modem/router
implementations.
As an aside, has anyone checked out the miniDSP SHD Series?
A bit pricey but an amazing collection of capability. I
like the inclusion of bass management and Dirac. There is
a RoonBridge plugin available for Volumio but doesn't appear
to be included out of the box. I've been auditioning
Volumio as a VM in VirtualBox. Haven't been disappointed
with its feature set and UI thus far. Having fun with this
hobby.
Cheers
|
|
|
|
Post by Creature on Mar 18, 2019 17:43:10 GMT -5
As an exercise, I FTP transfered a music file to my DLNA server.
It was a CD audio 2.0 PCM rip put into a FLAC container and
included tags and cover art.
I then browsed the DLNA server with VLC and found the URL by
which the renderer would request the streamed song. I then used
curl to transport and save the resultant stream.
The saved stream was an absolutely verbatim copy of the original
FLAC file, confirmed by comparing the checksum of the original
and the copy having made the round trip.
The means that a DLNA renderer is simply reading the original
media file. Similar to putting the file on a USB thumb drive
and reading the file from a directly attached USB drive, with
the exception that it was read across the network instead.
No different than having been read off of a NFS or SMB share.
|
|
klinemj
Emo VIPs
Honorary Emofest Scribe
Posts: 14,740
|
Post by klinemj on Mar 23, 2019 19:53:27 GMT -5
Well, darn it...this silly thread got me itching to build something again. I decided I am going to buy a HifiBerry DAC+ Pro and add it onto my R-Pi3. I want to see how it sounds vs. my DC-1. If it's close, I will upgrade the HifiBerry DAC+ to use an external power supply. It's a simple mod...a few $ in parts and 4 solders (two on the board, two to the power input) + drilling a hole in the case for the power input.
It will be a fun experiment. If I like it well enough, I'll build a 2nd, sell the DC-1 and get a 2nd BasX A100 and set up a 2nd headphone station.
Mark
|
|
KeithL
Administrator 
Posts: 9,928
|
Post by KeithL on Mar 26, 2019 16:52:50 GMT -5
The issue I have with DLNA is that it isn't always implemented the same - and it has too many options. In the particular setup you chose to test the file was transferred exactly. However, not all DLNA devices use the same options, and it can often be difficult to know exactly what it's doing at any given time. For example, if there is network congestion, can I trust DLNA to stop playing, rather than continue to play a degraded stream with packets missing from it? And, for that matter, the manufacturers of many DLNA devices don't actually tell you which transport mechanism they use by default... This means that, regardless of what DLNA is capable of, there is an issue with knowing exactly what it's doing at any given moment.
At best, as you state, it is the same as reading a file from a network share. However, it can be difficult to know if that is how it's performing right now. And, as you said, "it's no different than reading the file off a network share" - so why not just do exactly that?
I haven't played with Roon, but I am told that it can be explicitly configured, and actually indicates what protocols it's using, and whether any resampling occurs.
This would suggest to me that it offers additional benefits, without the drawbacks of DLNA (but at a higher cost in terms of its actual price tag).
As an exercise, I FTP transfered a music file to my DLNA server. It was a CD audio 2.0 PCM rip put into a FLAC container and included tags and cover art. I then browsed the DLNA server with VLC and found the URL by which the renderer would request the streamed song. I then used curl to transport and save the resultant stream. The saved stream was an absolutely verbatim copy of the original FLAC file, confirmed by comparing the checksum of the original and the copy having made the round trip. The means that a DLNA renderer is simply reading the original media file. Similar to putting the file on a USB thumb drive and reading the file from a directly attached USB drive, with the exception that it was read across the network instead. No different than having been read off of a NFS or SMB share. |
|
KeithL
Administrator
Posts: 9,928
|
Post by KeithL on Mar 26, 2019 16:55:15 GMT -5
Personally, I would be more inclined to compare the sound of the HiFiBerry Digi Pro+ to the sound of the USB output directly on the RPi.... both connected via the DC-1. (I may do just that when I get around to it.)
Well, darn it...this silly thread got me itching to build something again. I decided I am going to buy a HifiBerry DAC+ Pro and add it onto my R-Pi3. I want to see how it sounds vs. my DC-1. If it's close, I will upgrade the HifiBerry DAC+ to use an external power supply. It's a simple mod...a few $ in parts and 4 solders (two on the board, two to the power input) + drilling a hole in the case for the power input. It will be a fun experiment. If I like it well enough, I'll build a 2nd, sell the DC-1 and get a 2nd BasX A100 and set up a 2nd headphone station. Mark |
|
klinemj
Emo VIPs
Honorary Emofest Scribe
Posts: 14,740
|
Post by klinemj on Mar 26, 2019 19:08:55 GMT -5
Personally, I would be more inclined to compare the sound of the HiFiBerry Digi Pro+ to the sound of the USB output directly on the RPi.... both connected via the DC-1. (I may do just that when I get around to it.)
Well, darn it...this silly thread got me itching to build something again. I decided I am going to buy a HifiBerry DAC+ Pro and add it onto my R-Pi3. I want to see how it sounds vs. my DC-1. If it's close, I will upgrade the HifiBerry DAC+ to use an external power supply. It's a simple mod...a few $ in parts and 4 solders (two on the board, two to the power input) + drilling a hole in the case for the power input. It will be a fun experiment. If I like it well enough, I'll build a 2nd, sell the DC-1 and get a 2nd BasX A100 and set up a 2nd headphone station. Mark So, why would you do that? Mark |
|
Chris
Emo VIPs
Posts: 424
|
Post by Chris on Mar 26, 2019 22:43:22 GMT -5
|
|
KeithL
Administrator
Posts: 9,928
|
Post by KeithL on Mar 27, 2019 10:23:23 GMT -5
To be honest - just for fun. I've been using a Raspberry Pi as my music source for quite some time... I use its built-in USB output, which seems to work flawlessly for me, with either my DC-1 or my XMC-1... I don't do any network streaming - and I use Volumio and a local USB hard drive. I would expect the Coax connection to sound the same - or, if slightly different, then pretty much equal, compared to USB. So, to me, it would be on the same level of interest as test driving friend's cars... to see how they drive. I have very little expectation that the HiFiBerry DAC would match either the DC-1 or the DACs in the XMC-1 - but I would expect it to be at least pretty good.
However, since I have no inclination to use an outboard DAC, rather than the excellent ones built into my XMC-1, I'm not especially interested in how close the HiFiBerry one might be.
I should also point out that, while the options that support Roon are somewhat limited, one of the fun things about the Raspberry Pi is that you have many software choices... Most of them are free... Most of them don't take much effort to install... And, best of all, you can install each on its own micro-SD card... (This allows you to install, listen to, and even collect, a virtually unlimited number of different "virtual players".... micro-SD cards are really cheap nowadays).
This means that, rather than having to wipe your settings, or replace your current firmware and configuration files, and lose your work, every time you try something new... You can simply start with a new micro-SD card, save the one that you liked, and plug it back in later when you want to use it again. Essentially, you can have a whole collection of different Raspberry Pis, with different software on each, for the cost of a few extra micro-SD cards.
(Buy two Raspberry Pi units, and a dozen SD cards, and you can compare any two whenever you feel like it...)
I should also point out something to anyone considering a dedicated streamer device... A lot of the software for the Raspberry Pi is "headless"... When you run Volumio, the software runs on the Raspberry Pi, and you control it from a web browser on some other device... The down-side of this is that you MUST have a computer or smart phone turned on to control Volumio... However, once you tell it what to play, you can close your web browser, and even turn your computer off if you like, and it will continue to do what you've instructed it to do... The up-side is that very few all-in-one players have a 24" full-HD display on the front (which is what I have on the computer in my living room at the moment). (And, if you want a "dedicated display", you can purchase a small laptop or tablet, and leave it permanently connected, for far less than the cost of a purpose-built player with a big display on the front.)
Personally, I would be more inclined to compare the sound of the HiFiBerry Digi Pro+ to the sound of the USB output directly on the RPi.... both connected via the DC-1. (I may do just that when I get around to it.)
So, why would you do that? Mark |
|
