Low performance PC VS. High performance PC + BIOS Mod (Paul Pang concept) What do you think?

[iago]

What do you think?

 

First the disclaimer: I have no experience with Paul Pang's components, so I cannot advise you on their merits. I have some experience building audio servers, using laptops, embedded computers and currently a standard ITX (SoC) main board. You might consider me as a proponent of low power solutions.

 

Low performance vs. High performance PC

As firedog mentioned, define your use case. Low power PCs are absolutely sufficient for playback, but if you want to try resampling, room correction or DSD decoding, you need something with more performance.

 

BIOS Mod (e.g. Paul Pang)

I think you expect too much from this single modification. There are other hardware optimizations which will give you a better return for money spent. Consider a better power supply, an optimized operating system, diskless operation and signal reconstitution on the output of the server (I prefer S/PDIF, but USB is also possible - it's discussed extensively on this forum) - probably in this order. While I could not detect changes in audio performance when modifying BIOS settings (even if some of them had measurable impact on system performance), I did hear differences between main board brands. (*)

 

Actually i want a kind of Audio pc which approach to Hi-end Audio player

You should consider the whole system, consisting of playback hardware and DAC. The CD player has an unfair advantage over multiple box solutions, because it can use a more efficient internal signal path between it's components. I have a Linn CD player for comparison, and computer playback needed a lot of effort to make it perform better. The major part of playback performance comes from the DAC, put at least 2/3 of the budget in this component, the difference in sound is more noticeable than investing in computer hardware (exceptions: see above).

 

Budget 2000$

With this budget in mind, you could also look at commercial options, e.g. Aries or SotM/Sonore.

 

(*) I mostly listen for changes in sound stage and transients in acoustic guitars or drum kits, because I find them most easily recognized.

[iago]

I have choices. Have you ever tried them?

Which one sounds the best?

 

Sorry, I have not tried any of them; my last experiments were exclusively with J1800 boards.

 

The current incarnation of my audio server uses a Gigabyte GA-J1800N-D2P board, obviously made for the embedded market. Since my operating system would not boot reliably and the Gigabyte support was not really helpful, I also acquired similar boards from Asus (ASUS J1800I-C (the A was not available then)) and Asrock (ASRock D1800B-ITX) to see if it was a problem with the main board or a software problem. After fixing an embarrassing design bug, I tried all of them and found the Gigabyte board to have the most preferable sound. Since it also had the best real-time performance [1], I presume this might account for the difference.

 

[1] Essentially, I use a tool to measure and display the interval between the time a process is scheduled for execution and the time it actually starts. The output gives an overview of scheduling latency & jitter.

[iago]

Do you want to share your findings? Could be the start of a nice "sticky" with the mobo performance of various models perhaps?

 

Certainly. Give me until end of next week, by then I should be back home and have access to hardware and change logs.