Yep, that's me - excellent amplifier, a really captivating sound that gets your foot tapping.
Well, that is certainly the case. But I wonder how you can control that on a consistent basis from a single playback application on a complex desktop system running dozens of other processes and kernel threads, many of them even outside of reach of the user the player runs under.
Honestly, I have never seen any credible explanation by authors of closed-source players claiming their product is sonically superior to other bit-perfect players.
@npetralis
Sorry, when I missed it.
What kind of codec are you using for flac decoding and playing?
Does your player provide gapless playback?
I wrote a player with ASIO support and gapless playback and I am interested how others will do that.
Sorry, when I missed it.
What kind of codec are you using for flac decoding and playing?
Does your player provide gapless playback?
I wrote a player with ASIO support and gapless playback and I am interested how others will do that.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
In my opinion, there are 4 factors that may improve the sound during playback.
1) Media. Sound varies if the source id a normal hard disk, SSD disk , a USD stick, Cd-Rom etc. The best source is Memory, especially when paging to disk is disabled.
2)Track format. The best sounding format is wav. You may decompress a flac and compare the 2 files using Foobar, the difference is obvious.
3)Other processes-CPU utilization. It's hard to control other processes, so one solution is to run your player in High process priority.
4)Chunk size and latency when "talking" to Windows wave out. This is real alchemy-Voodoo, so trial-audition-error seems to be the only way...
A good non-switching power supply when possible helps too
Flac decoding and playback engine is written in C, using the "official" libflac.
What do you mean by gapless? The player buffers the whole file and then plays it. You can check the readme file for more information (or click the red pepper).
Just a wild guess, but I think he may be referring to back to back file playback and loading times creating gaps.
Sounds like a credible explanation for the percieved difference between players to me.
You mean Pureplayer?
It buffers the whole file to memory, so it plays only from memory.
Oh i see.
No, Pureplayer introduces actually large gaps between tracks because it decodes the track first, then loads it to memory and then begins playback. There is a "preprocess" option too that decodes all the selected files first. In that case the gap is reduced to the time needed for buffering.
In my opinion, there are 4 factors that may improve the sound during playback.
1) Media. Sound varies if the source id a normal hard disk, SSD disk , a USD stick, Cd-Rom etc. The best source is Memory, especially when paging to disk is disabled.
2)Track format. The best sounding format is wav. You may decompress a flac and compare the 2 files using Foobar, the difference is obvious.
3)Other processes-CPU utilization. It's hard to control other processes, so one solution is to run your player in High process priority.
4)Chunk size and latency when "talking" to Windows wave out. This is real alchemy-Voodoo, so trial-audition-error seems to be the only way...
A good non-switching power supply when possible helps too
Thanks, npetralis,
Do you now reformat everything to WAV before playing? Some posts ago you were reticent about any difference between FLAC & WAV - you have now heard it? Me, I'm gone the other way, not so sure anymore
I need to do some blind testing.I presume you address 1 to 4 in your player, pity you can't adddress the SMPS
Thanks, npetralis,
Do you now reformat everything to WAV before playing? Some posts ago you were reticent about any difference between FLAC & WAV - you have now heard it? Me, I'm gone the other way, not so sure anymore
I presume you address 1 to 4 in your player, pity you can't adddress the SMPS
The files in PurePlayer were "reformatted" from the beginning. That's why i can't hear any differences between flac and wav, but with other players of course i do. Obviously a misunderstaning
SMPS...it's ok for web surfing
@jkeny
Maybe we can build a killer source:
A low consumption fanless mini motherboard, SSD disk, 3GB RAM, preinstalled WinXP with everything disabled, CD-ROM, remote control, Pureplayer, a modified m2tech hiface embedded to the internal USB and of course an extremely good old school power supply. All these (except the transformer) in a heavy and beautiful aluminium case...
Maybe we can build a killer source:
A low consumption fanless mini motherboard, SSD disk, 3GB RAM, preinstalled WinXP with everything disabled, CD-ROM, remote control, Pureplayer, a modified m2tech hiface embedded to the internal USB and of course an extremely good old school power supply. All these (except the transformer) in a heavy and beautiful aluminium case...
Last edited:
The files in PurePlayer were "reformatted" from the beginning. That's why i can't hear any differences between flac and wav, but with other players of course i do. Obviously a misunderstaning
SMPS...it's ok for web surfing
Ah, yes, I misunderstood. You are still using SOX for WAV processing/playback & it still gives the message about 24bit WAC files I reported earlier. Great to see that you are going to Wasapi - do you reckon it makes a difference.
Re your next post - I'll PM
I do not want to start any flame but let me ask a few questions.
Proper decoding flac yields samples identical to the wav. The only difference is higher CPU load. Is that the cause you mean?
It has no effect on overall CPU utilisation, the player will not use more CPU cycles, since it is timed by the sound card in any case. Of course I assume the system is powerful enough and tuned well to ensure timely delivery of data for the card which for pure playback is no difficult task these days.
I do not know the windows wave-out technology. But in any case I cannot imagine how buffer sizes could affect the sound quality. Of course incorrectly chosen sizes (i.e too small in most cases) cause non-deliveries resulting in audible dropouts but when these happen we cannot talk about audio quality at all. I am talking about those "beter lows", "clearer mids" etc (just an example, no quoting here )
Proper decoding flac yields samples identical to the wav. The only difference is higher CPU load. Is that the cause you mean?
It has no effect on overall CPU utilisation, the player will not use more CPU cycles, since it is timed by the sound card in any case. Of course I assume the system is powerful enough and tuned well to ensure timely delivery of data for the card which for pure playback is no difficult task these days.
I do not know the windows wave-out technology. But in any case I cannot imagine how buffer sizes could affect the sound quality. Of course incorrectly chosen sizes (i.e too small in most cases) cause non-deliveries resulting in audible dropouts but when these happen we cannot talk about audio quality at all. I am talking about those "beter lows", "clearer mids" etc (just an example, no quoting here
)