Well, it won't give you more data, but it may push any noise up a few octaves, whether you like that or notice it or not may have a lot to do with your DAC. There's no 1 way to do upsampling though. Of course, you can draw a line between samples and extrapolate but companies like Wadia spent a lot of time making fancy, Bezier curve fitting algorithms as well.
I've always meant to try up and downsampling using a free tool like ffmpeg to try files at different resolutions. Why don't you give it a shot and see if you can tell any difference?
Any fans of upsampling out there? Anyone have experience with upsampling in which it actually makes your source sound better?
yeah, I have one CDP (a Wadia) that does upsampling well & I seem to like the sonics from it. It upsamples to 24/96 only. Like Erik mentioned Wadia has spent a lot of time coming up with some fancy algorithms that sound good. there are 3 choices (called A, B, C) that i can choose from. I believe choice A is their 'digimaster" algo & choices B & C put the listener at different locations in a hall. This unit always upsamples with no option to turn it off.
Personally I don't see how it can possible make any difference, b
you couldn't be more wrong!! ;-) Upsampling makes a big difference, often for the worse but there are several cases where the resulting sonics are very good. It all depends on the algorithm. Much work has been done in this arena by (Boothroyd-Stuart) Meridian, Wadia & EMM Labs. Of course there're many others now. I've seen a lot of people posting on A'gon that they prefer their DAC's sonics with the upsampling turned off & I've also seen a lot of other posts where people like the effect of upsampling. You are going to get a whole range of replies here. Better to search the archives here - there's a LOT written on this subject.....
On computeraudiophile the rage seems to be using HQ Player to upsample to quad DSD. There's about a 100pg long thread about it there, which I don't have the patience to read. I can't recommend it because I haven't heard it(would like to), but it's hard to conclude that upsampling definitely can't help unless you have tried that approach and don't hear any benefit. That approach seems to have captured the hearts and attention of many who spend much time focusing on getting the most out of their bits. Cheers, Spencer
DSD is another thing. The noise performance of normal DSD is normally pretty poor compared to 44/16 PCM, so it may in fact be a very good thing to do. In this case the argument is entirely about noise performance.
Who can hear it and when can they hear it, that's another thing.
Having played around obsessively with the settings on Audirvana+, I find no upsampling to deliver a more accurate sound in terms of timbre, as well as better imaging. Upsampling to my ears sounds initially thrilling and seductive but ultimately annoying. Like a hot girlfriend who turns out to be a total psycho.
phil9624, maybe one day your moniker will be phil9632 when you switch over to 32-b upsampling? Just kidding.... ;-) i don’t think anyone's going to laugh at you for * using a stop-light pen * or for thinking that cables make a big difference
many of us here have had similar experiences with both above items. Mostly, if it works for you, continue to use it. State your opinion & don’t try to convert anyone else. I’ve seen "fights" break out when some people try to convert others into their line of thinking or try to force a particular tweak down others' throats. In the one particular case in my personal experience, the Wadia 861SE, i also very much like their 24/96 upsampling implementation. I cannot say that for every 24/96 or 24/192 implementation I’ve heard.
but actually downsampled some 192/24 FLAC Neil Young albums from PONO to 96/24 WAV and think it sounds better, also smaller file. Go figure! Best, Phil
it's not entirely surprising. As you know, the fold-over frequency for 192K is 96K & your digital stream is incoming at 192K. The circuits involved in handling this digital stream & the corresponding analog circuits need to have 2X the bandwidth (compared to a 24/96 data stream). So, the real question is: in your electronics, were the circuits designed for higher bandwidth? If not, it is quite possible that the circuits are distorting & even tho' the 24/192 might be cleaner/better sonically, the circuits are unable to render this correctly. Dan Lavry (manuf of Lavry DACs. Lavry DACs are well regarded in the audiophile world) wrote a pretty nice white paper on this:
It seems pretty simple - try it and see what you think. On some systems it helps, on some systems it hurts and on some system there is not difference. People can write all they want about the theory of this, it all comes down to the implementation of the upsampling and the implementation of the DAC and other components. There is no answer - other than to try it and see how it works for you.
The technology is always worth knowing and understanding but in this case I would just listen and compare. Either way can be done well or not. What any individual will prefer soundwise cannot be predicted.
FWIW I have both oversampling and non oversampling DACs in that the technology interests me to compare, and I find each sounds much different but very good in its own way. Lots of good flavors of good quality sound out there.