I read this on another forum:
> > I had a chance to compare 3 different playback software on the Mac OS
> > - Songbird, iTunes and Pure Music all used with M2Tech HiFace,
> > yesterday and they sound TOTALLY different!
> >
> > Songbird is by far the worst. No only that the sound stage is flat
> > but the max volume is significantly lower.
>
> > iTunes has much better resolution and dynamics but sound aggressive
> > and thin compare to CD.
> >
> > Pure Music is clearly the best among all 3. It is much smoother than
> > iTunes but retains the same level of dynamics and resolution.
> >
> > For comparison, I also compared the above against FooBar + KS HiFace
> > under Windows XP on Bootcamp, and found Pure Music on Mac OS still sound better.
fyi: http://www.pure-music-player.com/ $129
I was sceptical, so asked the poster:
"Can you outline the test: ie music selections, amp and speakers, and especially the file formats".
He replied:
File format was AIFF on Mac OSX and WAV (ripped by EAC) on Windows XP.
We tested music mainly from the tracks ripped from FIM "Five Songbirds" HDCD.
The Macbook was connected to the DAC via M2Tech HiFace USB-BNC interface.
DAC used is the new Eastern Electric DAC with ESS9108 DAC and TFK ribbed plate 12AU7. The DAC was feeding the Eastern Electric BBA with TFK ECL82 and GEC 6X4W.
Power amp was Graaf GM20 OTL driving Ensemble Reference (very early Pawel bi-wire Neutrik version) on UK made Foundation Designer Stands specially made for Ensemble Reference (not the version for LS3/5A).
We also tried out a few other DAC like Sonic Frontiers SFD1 Mk2 and PS Audio Ultralink.
Comments from a guy who's used, Foobar, iTunes, album player, Winamp;
and on Macs ecoute, iTunes, and heard vlc . .
Better Sound for Mac Users- Pure Music Player - Head-Fi.org Community
And also from HK tube folk
You may try VLC (Version 1.1.0 latest and support 64 bit too) for Mac OS X.
I running mine on a 10.6.4 (64 bit mode) with 4 GB RAM, I found it better sounding than iTunes 9.2. Settings are 100% audio output (as one can overdrive it to 400%) and no decoded output (guess it's like buffer memory here). I haven't compared it with Pure Music.
For ripping in Mac, you may try Max. Again have not compare it with EAC for PC.
In general I have a feeling that Mac running Mac OS X 10.5 or above have better sounding than PC due to the OS is truly optimize for audio, graphic since it launched and Core Audio, again no real comparison.
> > I had a chance to compare 3 different playback software on the Mac OS
> > - Songbird, iTunes and Pure Music all used with M2Tech HiFace,
> > yesterday and they sound TOTALLY different!
> >
> > Songbird is by far the worst. No only that the sound stage is flat
> > but the max volume is significantly lower.
>
> > iTunes has much better resolution and dynamics but sound aggressive
> > and thin compare to CD.
> >
> > Pure Music is clearly the best among all 3. It is much smoother than
> > iTunes but retains the same level of dynamics and resolution.
> >
> > For comparison, I also compared the above against FooBar + KS HiFace
> > under Windows XP on Bootcamp, and found Pure Music on Mac OS still sound better.
fyi: http://www.pure-music-player.com/ $129
I was sceptical, so asked the poster:
"Can you outline the test: ie music selections, amp and speakers, and especially the file formats".
He replied:
File format was AIFF on Mac OSX and WAV (ripped by EAC) on Windows XP.
We tested music mainly from the tracks ripped from FIM "Five Songbirds" HDCD.
The Macbook was connected to the DAC via M2Tech HiFace USB-BNC interface.
DAC used is the new Eastern Electric DAC with ESS9108 DAC and TFK ribbed plate 12AU7. The DAC was feeding the Eastern Electric BBA with TFK ECL82 and GEC 6X4W.
Power amp was Graaf GM20 OTL driving Ensemble Reference (very early Pawel bi-wire Neutrik version) on UK made Foundation Designer Stands specially made for Ensemble Reference (not the version for LS3/5A).
We also tried out a few other DAC like Sonic Frontiers SFD1 Mk2 and PS Audio Ultralink.
Comments from a guy who's used, Foobar, iTunes, album player, Winamp;
and on Macs ecoute, iTunes, and heard vlc . .
Better Sound for Mac Users- Pure Music Player - Head-Fi.org Community
And also from HK tube folk
You may try VLC (Version 1.1.0 latest and support 64 bit too) for Mac OS X.
I running mine on a 10.6.4 (64 bit mode) with 4 GB RAM, I found it better sounding than iTunes 9.2. Settings are 100% audio output (as one can overdrive it to 400%) and no decoded output (guess it's like buffer memory here). I haven't compared it with Pure Music.
For ripping in Mac, you may try Max. Again have not compare it with EAC for PC.
In general I have a feeling that Mac running Mac OS X 10.5 or above have better sounding than PC due to the OS is truly optimize for audio, graphic since it launched and Core Audio, again no real comparison.
So, software players and OS sound different. And so do various uncompressed formats. The interface between PC and dac is far from ideal. Is this news?
And so do different motherboards, different PSUs, different CPUs etc. Because they all produce different amount and pattern of noise which is the only factor affecting sound for any bit-perfect chain. I do not belive anyone can tell "my PC sounds better with this RAM setup therefore yours will sound better with this setup too". Or different OS, or different lossless format, or any other factor not affecting the bit-perfectness of the chain. Because it is all about noise affecting the sound card clock/ SPDIF output signal jitter.
do (any, all?) of the listed players decode HDCD?
HDCD is a strange choice - it is a dynamic range compressed format that while not unlistenable without decoding is really meant to be best when played through a HDCD decoder - a HDCD decoder chip in a desktop CD or "universal" disc player or Microsoft's media player SW (Microsoft bought HDCD rights)
it is necessary to level match SPL to 0.1 dB to avoid perceptual difference due to playback level alone - larger loudness differences can be perceived as frequency imbalance due to Fletcher-Munson effect
HDCD is a strange choice - it is a dynamic range compressed format that while not unlistenable without decoding is really meant to be best when played through a HDCD decoder - a HDCD decoder chip in a desktop CD or "universal" disc player or Microsoft's media player SW (Microsoft bought HDCD rights)
it is necessary to level match SPL to 0.1 dB to avoid perceptual difference due to playback level alone - larger loudness differences can be perceived as frequency imbalance due to Fletcher-Munson effect
Last edited:
I agree. It is interesting however that they used the async Hiface with the MAC. It should in theory eliminate most problems with jitter (local clock) but clearly the factory implemenation is not so great. It would be interesting to try a modded Hiface with proper power supplies as well.
It looks as if Hiface is powered from the PC USB port, right?
On one of my work PC, with T-amp powered with 12V from the case, I can hear moving mouse (USB interrupts?), letters appearing in linux GUI terminal (GPU interrupts?, basically any change on the screen produces audible noise), HDD seeking etc.
But the solution IMO is not to tweak the SW/OS/PC HW to modify the noise pattern as that is virtually unpredictable and vastly differs among setups but to isolate the sound converter from the PC noise, keep it the master clock (i.e. PCI(e), USB asynchronous) and make sure the chain is bit-perfect under all circumstances. IMO this key condition can be provided only by transparent open source software all the way down to kernel and drivers.
On one of my work PC, with T-amp powered with 12V from the case, I can hear moving mouse (USB interrupts?), letters appearing in linux GUI terminal (GPU interrupts?, basically any change on the screen produces audible noise), HDD seeking etc.
But the solution IMO is not to tweak the SW/OS/PC HW to modify the noise pattern as that is virtually unpredictable and vastly differs among setups but to isolate the sound converter from the PC noise, keep it the master clock (i.e. PCI(e), USB asynchronous) and make sure the chain is bit-perfect under all circumstances. IMO this key condition can be provided only by transparent open source software all the way down to kernel and drivers.
Looks like the usual sighted tests that are subject to well known problems with unintended bias etc. Take their comments with a liberal (bucketful?) of salt.
- Status
- Not open for further replies.