• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

How can the same down load sound so different?

twiiii

twiiii

Lunatic Member
I have four sources for the same recording via iTunes. Each sounds different when sent to my Apple TV via airplay. . The oldest iMac 24" with maverick is just a very small fraction behind my New 27" iMac, followed by my 6 year old Mac book pro. The worst by far is the new iphone 10x. The music was originally down loaded by the 24 in iMac and transferred to the others via icloud. Now I assume the computers are using the lossless format I chose originally where the phone is using that God awful format every one complains about. My Sennheiser headphones have a nice roll off so even with the phone set in the flat position its tolerable. I use blu-tooth in the Acura and you can hear the highs being congested and rolled off, to make it listenable. Oh the word is MP3, I think think. But if its using air play like the computers , then whats going on. I was listening to a Pentatonic album, by the way which sounds great on the 27" i Mac.

What going on I'm curious as Hell!
 
Register to hide this ad
It "just works." We're not supposed to ask how and why it does. We are to believe that people far smarter than us have made all of the difficult and relevant decisions so we don't have to worry about such things.

It's kind of hard to follow your description. You bought the file once and it's now on four devices, one sends it to an Apple TV via airplay, two computers you listen to on Sennheisers, and the phone gets played via blu-tooth in the car? And they all sound different?

I vaguely recall reading a few years ago about Apple TV or airplay downsampling whatever was sent to it. I'm guessing the biggest variable is each device having different DACs. I know Apple used a bunch of different DACs in iPods, sometimes different ones in the same model.
 
Are you listening with the speakers on the MacBooks ?

Of so that probably has a lot to do with what you are/aren't hearing as the speakers and sound card in each are different. It's like comparing my HT system to my vinyl system to my car system, each sounds vastly different even using the same source material.
 
twiiii said:
I have four sources for the same recording via iTunes. Each sounds different when sent to my Apple TV via airplay. . The oldest iMac 24" with maverick is just a very small fraction behind my New 27" iMac, followed by my 6 year old Mac book pro. The worst by far is the new iphone 10x. The music was originally down loaded by the 24 in iMac and transferred to the others via icloud. Now I assume the computers are using the lossless format I chose originally where the phone is using that God awful format every one complains about. My Sennheiser headphones have a nice roll off so even with the phone set in the flat position its tolerable. I use blu-tooth in the Acura and you can hear the highs being congested and rolled off, to make it listenable. Oh the word is MP3, I think think. But if its using air play like the computers , then whats going on. I was listening to a Pentatonic album, by the way which sounds great on the 27" i Mac.

What going on I'm curious as Hell!
Click to expand...
Airplay is meant to be uncompressed audio, but it's not unusual for wireless protocols to gracefully degrade -- i.e., switch to compressed formats or elide data in order to reduce bandwidth requirements -- when the connection quality is poor. I don't know if Airplay gracefully degrades or not, but it's typical of networking systems to do so. This can result in audibly -- and notably -- poor audio when it happens.

A significant difference may also be the quality of the DACs in your various Apple devices.

The combination of different DACs and network degradation could easily account for dramatic differences in audio quality.
 
twiiii said:
The music was originally down loaded by the 24 in iMac and transferred to the others via icloud.
Click to expand...

Some cloud music systems do not copy your original file; they give you their copy of the track identified by metadata. Thus, you may have downloaded a lower quality version from iCloud.

Compare the file type and file sizes. Use a direct file transfer mechanism to exchange files; I'm sure Apple must support file exchange between machines on a local network.
 
cpt_paranoia said:
I interpreted his post to mean all devices were sending to his apple TV, and being compared that way. So only one DAC involved.
Click to expand...
I interpreted it as the Apple TV was also being compared to the iPhone X, various Macs, etc., but upon re-reading, I think you're right.

Therefore, it does sound like what was copied was the song signature and each device was downloading a different version -- or the same version at different levels of compression -- and casting it to the Apple TV.

Or... The network connection is flaky on the Apple TV, and it's sometimes working at full capacity and other times gracefully degrading.
 
twiiii said:
Now I assume the computers are using the lossless format
Click to expand...
All downloads from iTunes are lossy 256K AAC regardless of what you do with them later. It is Apple's MP3 equivalent format.

WiFi throughput is likely not the issue given the low bandwidth. I did, however, experience buffering issues in the garage when playing 192/24 FLAC content with its ~ 5100 kbps bitrate . Placing an extender there solved the problem.

twiiii said:
But if its using air play like the computers , then whats going on. What going on I'm curious as Hell!
Click to expand...
You're hearing each transport and its power supply! I use a general purpose computer for playback only in the office and as file server.

Instead, I use two small digital endpoints in my systems, a Raspberry Pi based unit with Allo DigiOne digital board and a Sonore microRendu. While the RPi is a very cost effective platform, it is not designed from the outset strictly for audio playback like the uRendu and does not offer the same level of transparency. Power supplies, especially with computers can make a pretty significant difference. Most use noisy switchers. I used an ultracapacitor based LPS-1 for the uRendu which is galvanically isolated from the AC and uses a low inductance starquad DC cable. With the RPi, I use either an inexpensive linear or a battery source for lowest noise.

A lot happens to the bit stream on the way to the DAC.

urendu_sm.jpg
 
Last edited:
I have a stereo system for analog, and a HT system for digital that shares the Left and right speakers with analog system. I have a switch to choose either the Analog pre amp or the digital processor. I plug my Apple TV piece into the HT processor using an optical input. Depending on which computer of 3 I am using or the iphone I can connect one at a time thru my home network to the apple TV. The new i Mac is in my Office, the Old I Mac was in Carol's sewing room office. And the Mac Book pro stays in the room with the main sound system. Most of the time I use the Mac Book pro feeding the HT processor via the Apple TV module. But once in a while I'll be in my Office and use the new i Mac to feed the Apple TV in the main listen room and my office through remote speakers. Carol could do the same thing from her sewing room office. And for the fun of it I can do it from my 10X even while taking a relaxing bath, via remote speakers. Well I got the new iphone 10 x and got it interfaced with my Acura and wondered how it would sound with the main system. So after connecting to the Home Network setting the password I played the identical music stored on the phone via icloud from the old 24 inch i Mac and the sound was not only different but 3rd rate. Mushy and distorted highs for one and poor definition elsewhere. That bothered me, so I decided to check out each computer played one at time via the home network feeding the Apple TV driving the HT processor in the Stereo mode. As I said earlier the new i Mac was the best as expected, then followed by the 9 year old 24" iMac, then the 6 year old Mac book pro and lastly trailing far behind the 10x iPhone.

My question still is, if the digital source is all the same originally from the iMac 24" distributed to the other three units and stored there, why is there so much difference in sound when plaid thru the digital HT system in the stereo mode. I would think all three computers would sound the same, but they don't. Oh and the EQ settings via i Tunes is set flat for all three computers. The iphone is set flat, too.

I am sorry I confused every one. Its similar to taking a CD and playing on 4 different units and having three sounding very close and the 4th being a dud. Why? Its not the Home network or the Apple TV or the HT system, nothing changes there. Its either the computers or the i phone. Some thing is different. I thought digital was digital. Yes sound cards can sound different, but I am not using them, am I? The HT processor does the decoding. Has the iphone stored the sound via MP3 to save speace where the three computers are using the same lossless Apple digital format which are supposed to be bit for bit. Something is going on I just wonder, what.

When I use the head phone out puts of the computers I know there will be a difference because of the sound cards. and that goes for the iphone, too. The same goes for the blu-tooth hook in the car. But I am using Air play when connecting to the apple TV and there are differences. why? Thanks for being patient with me?
 
Last edited:
twiiii said:
My question still if the digital source is all the same origially from the iMac 24" distributed to the other three units
Click to expand...

Unless you have compared the file sizes, extensions and done a binary file comparison, I don't think they are the same file, for the reasons explained above.

Have you done those checks?

Your 'mushy and distorted highs' is a classic problem of lossy compression.
 
twiiii said:
My question still is, if the digital source is all the same originally from the iMac 24" distributed to the other three units and stored there, why is there so much difference in sound when plaid thru the digital HT system in the stereo mode.
Click to expand...
In addition to what @cpt_paranoia suggested being a possibility, AirPlay relies on having a reliable network connection. I've heard anecdotally that AirPlay does not cope well with network problems. It may be that your various Apple devices are not consistent in reliably connecting to the network, with the result that with some devices the AirPlay protocol is being forced to compensate for reduced bandwidth due to network problems, resulting in a corresponding loss of sound quality.

According to https://en.wikipedia.org/wiki/AirPlay, audio streams are transmitted using Apple's proprietary lossless audio format, but it's typical for UDP-based network protocols -- which AirPlay's audio uses -- to degrade gracefully in the presence of intermittent transmission failures. Degrading gracefully means transmission doesn't fail outright; but does mean the sound quality may be compromised as the protocol tries to compensate for network problems.
 
twiiii said:
My question still is, if the digital source is all the same originally from the iMac 24" distributed to the other three units and stored there, why is there so much difference in sound when plaid thru the digital HT system in the stereo mode.
Click to expand...
While they may be fed the same file from your music server, each unit is effectively a digital transport - which is where the difference lies. A Chromecast player will never match the quality of a dCS Network Bridge. I run several digital endpoints all sourced from the same music server and find variance between them as well. The app I use for remote control of the LMS server also turns either an iPhone or iPad into a player. The phone most definitely does not deliver the same quality as the microRendu.

twiiii said:
MI am sorry I confused every one.
Click to expand...
No confusion here. I have experimented greatly with using different endpoints and even different power supplies for a given one and in the case of the RPi player, fine tuned a number of kernel parameters for optimum playback. I declocked the CPU for minimum noise and power consumption, disable all manner of unnecessary peripherals and configured a bunch of its memory for a large 600 MB output bufffer. For Redbook content, it offers memory playback.

twiiii said:
I thought digital was digital.
Click to expand...
I'll repeat: A lot happens to the bit stream on the way to the DAC. The scenario really isn't any different than comparing different optical disk transports and finding differences with them. Digital transports vary widely in their performance due to a number of factors.

twiiii said:
Has the iphone stored the sound via MP3 to save speace where the three computers are using the same lossless Apple digital format which are supposed to be bit for bit.
Click to expand...
Once again - if the recording is "for the same recording via iTunes", it is lossy 256 kbps AAC. You cannot download or stream any lossless content from iTunes or their music service. Period. Let me say that again since you missed it the first time around. APPLE DOES NOT OFFER LOSSLESS CONTENT!

For that, you need Tidal or Quobuz. I use Tidal and get 44/16 lossless FLAC. Perhaps you might be facing more latency with streaming the lossy content via iCloud to the phone.
 
Last edited:
DaveVoorhis said:
According to https://en.wikipedia.org/wiki/AirPlay, audio streams are transmitted using Apple's proprietary lossless audio format, but it's typical for UDP-based network protocols -- which AirPlay's audio uses -- to degrade gracefully in the presence of intermittent transmission failures.
Click to expand...
Which is perhaps why Apple abandoned the product. Why compress network traffic at all? While the main player connects to the music server via 1 gbit ethernet, the garage system connects via a standard 802.11 based access point and extender. There's plenty of speed available to stream high data content packets without inducing overhead at both ends.
 
It is not clear to me what your source is and what app you are using to play the file. It sounds like you stored a single version of a music file on iCloud. And you are using iTunes on each device to access and play that music file.

If you are relying upon the Apple iTunes ecosystem to play your music, I think all bets are off. For example, iPhone quality takes a hit when streaming from iTunes because of adaptive streaming to minimize bandwidth. I understand that can be bypassed. Have you? iTunes also plays local files differently than remote files. I do not trust iTunes and Apple to let me play my music unadulterated. It may be possible to set up iTunes to play everything HQ, but the anomalies you are experiencing and my own difficulties lead me to believe that other methods are more desireable. I have abandoned iTunes for music.

For a better test, I suggest you use a bit perfect app to access a single file stored somewhere. This will isolate you from iTunes, which sometimes does strange things with music depending upon where it is streaming from.

I have music stored on a Mac Pro with music server software (I use PLEX, some prefer JRiver) running on it. I access those files from multiple computers, laptops, and mobile devices and it all sounds the same when I stream the music to the same amp and speakers. But I use a non-Apple app to access the music server, so I do not worry about Apple manipulating my stream.
 
DaveVoorhis said:
According to https://en.wikipedia.org/wiki/AirPlay, audio streams are transmitted using Apple's proprietary lossless audio format, but it's typical for UDP-based network protocols -- which AirPlay's audio uses -- to degrade gracefully in the presence of intermittent transmission failures. Degrading gracefully means transmission doesn't fail outright; but does mean the sound quality may be compromised as the protocol tries to compensate for network problems.
Click to expand...

UDP is connectionless with only data integrity via checksum. Nothing graceful about it in adverse conditions.
 
JP said:
UDP is connectionless with only data integrity via checksum. Nothing graceful about it in adverse conditions.
Click to expand...

That's why the higher level protocols do the clever stuff. The protocols that use UDP as an underlying bearer, like Dave said.
 
You must log in or register to reply here.
Back
Top Bottom