Source should never offer a clock. Clock should be always something owned by the DAC, sitting right next to the D/A conversion stage.
For example NAA doesn’t transport any audio clocks and it is not offering any audio clocks either.
Already because HQPlayer server can be loud due to multiple fans. Use of NAA allows you to run HQPlayer server in a machine room, far away from listening place.
As usual, my development is objectively driven. So I’ve done quite a bunch of measurements on this area too. But how much NAA makes impact from electrical noise point of view, depends a lot on the overall system.
Thanks! I run a PC. My DAC; RME is however limited to DSD256. Already today, usb from the desktop pc with sparse noise treatment, HQPe upsampling to RME max rate sounds really great. Lokking for the last drops …
My pc is very silent. ~ 15dB, ventilation is louder. More concerned about the electric and airborn noise. Which is present in both my HQPe PC and in any endpoint, more or less. Close coupled by usb cable shield to DAC in both cases.
So agreed, the total noise from my PC or a NAA is the thing to determine. I mean I can feed a NAA from linear psu. But I can install a Matrix Matrix Audio Element-H USB output – Thomas Audio Örebro Card with expernal supply, a card that is also sealed/shielded inside the pc, with quite smaller leaking noise. Cost is almost equal, but one less device to spill while dusting
Most of mine too, but the newest i9-12900K + RTX2080Ti combo not so much anymore at full CPU + GPU load. Although I made it as quiet as possible. Likely newer RTX3080Ti could be a bit louder again at full load. These machines blow out few hundred watts worth of heat.
My HQPlayer Embedded servers are usually connected directly to DACs. While with HQPlayer Desktop machines I tend to use NAAs.
NAAs also give me possibility to easily use pretty much any of the DACs from any of the servers.
But anyway, you have the flexibility to decide which way you like to go.
Smallest NAA I have is powered from two AA batteries.
Next week, I hope to be able to bring this home from work and try it; Antipodes K50 - Antipodes Audio It will be highly interesting, and I am anxious to see if their combination of softwares are able to boost every drop out from HQPe installed from factory. Won’t be able to purchase though, way out of my league, but still the challange is to come as close as possible to the prime stuff without having to pay the price. Will check out the S40 too, would be impressive if that is capable of running upsample as I do now DSD256/ASDM7ECv2/poly-sinc-gauss-xla/DoP/48K SDM, ALSA backend. Your newest HQPe really made the difference for CPU loads and ability to push higher. Prior to that update it would not even launch playback.
Could you for reference hint me about brand and modell of the tiny 2 battery NAA?
Today I experienced something I cannot explain. I have two albums on NAS, strangely enough is one 24-96, the other 24-192. Both in wave-format.
When I played back the 192 kHz, I got drop-outs (5 sec music, 5 sec buffer … type-ish), at 96 kHz I did not. Running ASDM7ECv2, poly-sinc-gauss-xla, DoP, 48k SDM, 12,880,000 Hz (DSD256)
Can anyone explain why the PC’s having a harder time with 192 kHz compared to 96? And it is only on wave I noticed this. FLAC, flows perfectly irrespective of the two sample rates, I might add.
//My life has gone from having a hard time understanding women, to the point where HQPe is my main concern. Is that age dependent, perhaps?//
It is the same case for FLAC as well. With double sampling rate, you get about double the filter load with those two stage filters, because the first stage needs to operate at twice higher data (sample) rates.
You’re right, off course. Also FLAC… Are there any change I can do, that will cure this problem, but with only the slightest difference in filter choice/SQ? I am deadly in love with the complete silence of this gauss-xla-filter. Wonder if this precision filter can be the reason why some tracks now on sounds like lacking a fraction of treble energy, when it is now so clean? Perhaps the mix, master monitor system was not as clean in studio? Anyhow I don’t mind, it makes my day when I feel like listening to the gothic rock genre, where bombastic sound usually are all over the place, but now makes so much sense when the arrangement becomes easy to hear especially when large classical orchestras includes.
If you change your filter to the -2s (E.G. gauss-xla becomes gauss-xla-2s) it will reduce the load on your processor. It may reduce it enough that the lag disappears. But will introduce a 2 second lag between you hitting play and it actually doing so.
