|
|
Post by sycraft on May 19, 2016 12:41:39 GMT -5
I've whined a bit about the noise levels on the XMC-1 before (because I am whiny about noise) but I've discovered a major factor in it: Dirac Live. With it enabled, the noise floor jumps 10dB or more. Not sure what is going on, perhaps an implementation bug? Something you might wish to look in to.
|
|
|
|
Post by AudioHTIT on May 19, 2016 12:55:34 GMT -5
I've whined a bit about the noise levels on the XMC-1 before (because I am whiny about noise) but I've discovered a major factor in it: Dirac Live. With it enabled, the noise floor jumps 10dB or more. Not sure what is going on, perhaps an implementation bug? Something you might wish to look in to. Could it be that, for instance, certain treble frequencies are being boosted thereby making the existing noise more apparent? So technically the noise isn't increasing, you just hear it more because Dirac has equalized an area where it's easy to hear. Not saying your premise is wrong, just another possibility. |
|
tesla
Sensei
Amplitude Fetishist
Posts: 152
|
Post by tesla on May 19, 2016 16:24:59 GMT -5
I've whined a bit about the noise levels on the XMC-1 before (because I am whiny about noise) but I've discovered a major factor in it: Dirac Live. With it enabled, the noise floor jumps 10dB or more. Not sure what is going on, perhaps an implementation bug? Something you might wish to look in to. How did you determine this? Oscilloscope? SPL meter? What source? All sources or just a few? I'm curious about your methodology. Cheers! T- |
|
|
|
Post by sycraft on May 19, 2016 18:23:08 GMT -5
Turned on the XMC-1, HDMI input from the computer, no signal. Turned the volume up to 0dB, changed between Dirac and no-Dirac settings, heard major audible difference in the noise level.
|
|
LCSeminole
Global Moderator 
Res firma mitescere nescit.
Posts: 20,850
|
Post by LCSeminole on May 19, 2016 19:11:08 GMT -5
Turned on the XMC-1, HDMI input from the computer, no signal. Turned the volume up to 0dB, changed between Dirac and no-Dirac settings, heard major audible difference in the noise level. Should I expect the noise floor to jump 10db if I were to use an input from another source, say my Oppo BDP-93, with no signal as well? |
|
|
|
Post by sycraft on May 19, 2016 20:02:10 GMT -5
Not sure, I'll have to try it with another source. Doesn't happen on S/PDIF since the auto-mutes engage and quiet everything down. However here's more indication that something may be up: Have a look at this graph from Hometheater Hifi. They are testing Dirac's range and it, as specified, stops at 24kHz since it processes at 48kHz. However look at the graph beyond 24kHz: The SNR is up in the -95dB range, what you'd expect for 16-bit:  It looks to me kinda like Dirac is limiting output to 16-bit words, which would then cause an increase in SNR. |
|
|
|
Post by sycraft on May 19, 2016 20:18:09 GMT -5
I also can replicate the behavior on the S/PDIF input, though with lower noise overall. Here when there is nothing playing, it is extremely silent, as the mutes engage. In Sound Forge I generate a -80dBFS 20kHz sine wave, which is going to be basically totally inaudible to me even with the volume cranked since my hearing doesn't really go past 17kHz and that is very low level. When I play back the sound at full volume (using ASIO to a Focurite card) on the non-Dirac preset, the noise level raises just a touch. When I play it back on the Dirac preset, it raises a good bit more. In both cases is drops away when the signal stops since the mute kicks on.
|
|
