Looking for great DSD samples, audio advice
- Thread starter onlyconnect
- Start date
tomek
pfm Member
If dsd is made from pcm, than it can't sound better, than pcm.
onlyconnect
pfm Member
I tend to agree but there are other things to consider:
DSD could add euphonic colouration and thus sound "better"
DSD could sound better in the special case where there is no PCM conversion eg a direct to DSD live recording
Some people think that PCM sounds better converted to DSD and played through a DSD DAC. If so, I suppose you could argue that this is the "best" way to design a PCM DAC.
Tim
Paul R
pfm Member
foobar2000 gives you 1 dB resolution on the conversion gain. I think it should just come out 'right' though.
Paul
Barry williams
Banned
I wouldn't move your SBLs if you have them sounding great at home.
It will cost you two new sealing kits £150 as they should not be transported built.
Also if you have a hotel room with no solid wall to stand them against they won't sound any good.
It will cost you two new sealing kits £150 as they should not be transported built.
Also if you have a hotel room with no solid wall to stand them against they won't sound any good.
onlyconnect
pfm Member
Just did a conversion in Foobar2000 and the Flac is definitely quieter than the DSF. Will check the options.
Tim
onlyconnect
pfm Member
Hmm, this is a problem.
I am finding, with my particular DAC, that the conversion to PCM needs to be around +6dB to maintain similar volume. However at that level it is also clips. Maybe I need to find some "quiet" DSD samples!
Tim
I am finding, with my particular DAC, that the conversion to PCM needs to be around +6dB to maintain similar volume. However at that level it is also clips. Maybe I need to find some "quiet" DSD samples!
Tim
onlyconnect
pfm Member
Dear me, look at that ultrasonic noise!
Tim
Tim
Paul R
pfm Member
That's slightly surprising. I think DSD is nominally -6dB, although many are 'hotter'. I guess you DAC is 'true' DSD and Teac want DSD to sound 'better'...
Paul
onlyconnect
pfm Member
It is an interesting thought. But I think I picked a bad sample, the Elton John track I used is relatively loud vs the average. I'll just have to be careful.
Tim
Barry williams
Banned
Steve,
TBH and not meaning to be rude here but i don't think the OP has a clue when it comes to real DSD music files let alone moving and SBL set up?
onlyconnect
pfm Member
How kind. Care to give me one?
Yes I did set up my SBLs with new gaskets and silicone sealant when I last moved them and have followed Naim's advice never to move them assembled other than short distances around the house, and no, I don't intend to take them to Scalford.
Tim
Barry williams
Banned
Tim,
Thats good to hear.
Now to the true DSD files/not PCM.
How are you going to play these ?
What source DAC and Amps are you going to use?
One of the problems when dealing with DSD is that, unlike PCM, the absolute maximum level is not easily defined and is slightly programme dependent. For this reason, the SACD specification insists that there must always be a 6dB margin.
In addition, there is no rule governing the points at which your DAC ought to level-match between PCM & DSD; the DAC makers will, themselves, be in a bit of a dilemma over the ideal gain compromise between the two formats so a simple and elegant answer may not be forthcoming. For instance, the maker may well have chosen to reduce the gain of the DSD path by 6dB to play safe, while not doing too much harm to the PCM noise performance. In such a case, you will not be able to achieve a level match with a file conversion that, in theory, ought to be within ≤1dB.
Not really what you want to hear, but it may well be necessary to effect an external (i.e. analogue domain) gain adjustment to get two notionally similar files (one PCM, the other DSD) to play back at the same level.
onlyconnect
pfm Member
I have thought about this but it is error-prone. I think the best solution may be the Meyer/Moran approach of inserting an additional PCM conversion into the signal path after the DAC output but still looking into it.
Tim
Not if you work out the required attenuation factor and create it out of fixed resistors.
Trouble is, that approach arguably negates the validity of the entire investigation!
onlyconnect
pfm Member
Argue it then
Note that DSD has the advantage since you hear it without any intervening switch box or processing. The latter only comes in for PCM. So it tilts the balance in favour of DSD - maybe that is your objection?Tim
Well, based on your demonstrations at Scalford 2 years ago, those who expect to hear a big difference, and then don't, will assume there is some trickery afoot rather than adjusting/correcting their existing opinion/prejudice...
Not really. A good experiment is useful, a poor one less so. That's all really.
If you actually wanted to test the performance of DSD vs PCM, all you need do is run a loop of each (A → D → A) but insert a 40dB attenuator at the input to the test rig and a 40dB amplifier at the output (-40dB → A → D → A → +40dB). Following such a test, not much doubt will remain