Gav

There were two reasons why I was interested in the outcome of the vote on versions A, B and C: Firstly, to understand the direction that Audirvana will evolve. I voted C as being my least favourite, so the result is not ideal for me. Secondly, to see if there was any concrete evidence revealed about subtle changes in the digital domain leading to SQ effects. I hear differences, but also accept that this may be 'mind over matter'. If the votes were 33%, 33% and 33% then maybe this would show that there is no real SQ difference. However, this could also show that people's preferences are random even if there are real SQ differences being heard. A result that favours one player is an interesting result but if the number of participants is small then this could be chance alone. To give a feel for the role that chance played it would be interesting to know the actual number of votes rather than the percentages. I could not really distinguish A from B and voted them both as number 1. Version C stood out to me as being different. C seemed forward in the mid-range/upper mi-range, particularly spotlighting vocals. A and B both seemed to present music better as a coherent whole. All pretty subtle though. While testing, I could understand why one might prefer C to A/B and also why one might prefer A/B to C (as is the case for me). I expected the results of A and B to be similar and C to stand out as either a definite winner or loser. The vote percents seem fairly consistent with my opinion. So I must have been hearing a real difference, right? On the other hand though, A is LLVM and B is GCC and C is some kind of half-breed. So logically I now expect A and B should sound the most different with C as a half way compromise. This is not what I heard, so it must be all in my head? Gav

I have 3 FW cables that I have tried between my MBP and Weiss INT202 (feeding a naim DAC): a super-cheap unbranded cable; a belkin; and an Oyaide neo+. All three sound different. I completely agree with Blu's description of the sound from the Oyaide, however I like the effect through my system which can get a bit bass heavy, so that is my cable of choice at the moment. I have limited computer and electrical engineering knowledge but I just figure that some RF noise finds its way from computer to DAC and the cable can influence the form of the noise. Gav

Tom, I updated to the latest Weiss driver software (3.5.5.10185). You're guess was right, it doesn't support integer mode! Oh well maybe next time. Tonight was my first chance to properly listen to Audirvana with 'Use Max I/O buffer' active. This was with 'normal' operation mode set in the latest Weiss software. Wow, I'm a happy man. I have tended to switch between Audirvana and Decibel, maybe favouring Decibel overall. Now Audirvana is clearly top of the tree for SQ in my system. A donation going to Damien from me. Gav

Hi Chilest, You're absolutely right. I set 'normal' in the Weiss application on my MBP and the 'Use Max I/O buffer' now works flawlessly. How the heck did you ever work that out?! Thanks for your help on this. Now, how do we convince Weiss to produce a new driver for the INT that supports integer mode? Gav

It's a great project that you are working on here Damien. I have an ongoing issue though with "Use Max I/O buffer" across all of the audirvana versions I have used. With this checked the sound is a 'chattery' 'stuttery' mess. Uncheck it and it's great. Playing around with the other pref settings does not seem to solve it. I can see that this has been discussed earlier in the thread but I can't work out whether an answer was found. I am using a mid 2009 macbookpro with latest up to date OSX feeding a Weiss INT202 FW to spdif interface. If I disconnect the INT202 and select built-in output then there is no problem using Max I/O buffer. Gav.