ECDESIGNS, a Dutch DAC manufacturer made recently tests with different software for bit perfect playback with these results:
“ECD (Gordon) confirmed that they could not get Audirvana to play “bit-perfect” on Windows 10 either. Could be a setting, in which case they (Audirvana) are not making it obvious ! On Mac, Gordon confirmed Audirvana is bit-perfect, and JRiver on Windows with Wasapi.”
So Audirvana 3.5 for Mac seems to be bit perfect.
Matt
Just a quick quote that I did not come up with but necessary in this instance.
Opinions are like a$$holes, everybody’s got one.
If my man wants to say AS > A3.5, then it’s his opinion, which is true to him, it’s not meant to force you to share his opinion.
You have your problems which maybe he doesn’t so experiences are different hence the different opinion.
You know you’re being hateful when near the end of your opinion you state not meaning to be hateful. Then you end by stating that you would like to try AS but it’s too buggy? So you haven’t tried it and you find hard to believe the opinion of someone who has?
This would do AS a favour. The louder playing software appears to be superior sounding.
Matt
I won’t dare to claim that my observations are valuable at all or based on solid evidence. But if they were, any comparing listening would have to take that into account somehow. Like turning the AS’ volume a little bit down in the first place 
Very interesting, thanks! I will look this up.
Then I’m puzzled why these “bit perfect” apps sound different.
Bugs for the Mac were fixed in Studio V1.3.
I am listening to it and have the impression that V1.3 sounds better than the previous versions. But everything is not rosy.
In Bit perfect mode the sound is very good.
Upsampling 16/44.1 and 24/96 FLAC tracks, in r8brain mode, to DSD 256, makes only a subtle difference compared to the playback in Bit perfect mode.
Oddly, when I upsample 16/44.1 and 24/96 FLAC tracks to the option “Max DAC frequency”, I get crackles, both with r8brain and SoX. I rerouted the computer but the crackles remain.
When playing DSD tracks the sound is good.
I use an iFi micro iDSD Black Label DAC and Audeze LCD-2 headphones.
I don’t have A3.5, so I compared AS with A+ 2.5.
In Bit perfect mode both players sounded the same, when I played 16/44.1 and 24/96 FLAC tracks.
Unsurprisingly, there was no difference between the two.
As I said before, with AS V1.3, PCM upsamling provokes crackles. I did not have these crackles with V1-V1.2. For this reason, it was not possible to compare the new r8brain mode to the old iZotope mode of A+ 2.5 nor to upsampling modes of other players that I have.
I hope I could test r8brain mode with the next version of AS, if it is released before the end of the trial period.
When playing DSD64 tracks, ripped from a SACD, AS sounded better than A+ 2.5. The sound was clearer.
For the sake of comparison, I played the same DSF tracks with HQPlayer. I set it with Filter sinc-M. It sounded more natural than AS with more details and subtlety, more analogue like.
••••••••••
This was a SQ test, but I must also say that as an application A+ 2.5 is reactive and smooth, while AS is often displaying the spinning ball.
Thanks for the feedback!
I wonder if the beach balls were due to having the music files on the USB flash drive? Those can be pretty slow.
I didn’t experience any crackles with upsampling with my HD DAC1. But also it didn’t improve the sound for me so I don’t spend too much time on that.
VoyagerDude,
I all the tests, with the three players, I played the same files from the same USB-Key.
The spinning ball and the crackles were not due to the fact that the files were on the USB-Key.
First, because I played these tracks already from the USB-Key with the previous versions of AS, and just like you, there were no crackles.
Second, AS loads the track that it plays in RAM, so the drive on which the track is stored has no importance.
Third, I upsamled some tracks with A+ 2.5, in iZotope mode, just to see if the player plays them, and there were no crackles.
EDIT
In addition, as I said earlier, there were no crackles when AS upsampled PCM tracks to DSD256.
Ok cool. Thanks for clarifying.
With my iFi micro iDSD BL, it’s difficult to say if there’s a difference between Bit perfect and the upsampling of most players, except for HQPlayer.
If a new version of AS is released without these crackles, I’ll test the new r8brain upsampling mode also with an older DAC that I have and that I don’t use anymore. Thos DAC provides a much better SQ with an upsampling player. It’s the iFi micro iDAC. It does not support DSD and is limited to PCM 24/192.
2 cents, and 2 cents, will make 4 cents… or so…
Installed and tested AS 1.3 yesterday once again against 3.5.
Did it because I love the new interface better than the 3.5 one, and I keep telling myself : that’s not possible… why did they get it so wrong ?… they must realize it at some point…
But alas, no. Same thing.
AS deprives the sound of all its liveliness. Surgically.
Maybe some listeners will like it better this way, but I definitively tell myself that the direction taken is not for me.
I’ve spent a lot of time tuning my system according to the music I listen (choosing and testing each component, each cable etc…), and the way I want it to sound : analog, warm, without losing holographic imaging and a coherent amount of details.
I mostly listen to 50/60 jazz, classic rock and classical baroque, so for this type of music, to me, these criteria are paramount, and AS deprives the sound of its “analog” feeling and presence.
Library scan (4.5TB) takes forever, and to be honest, I don’t want any software to scan my library, and I don’t want it to mess with my tags.
We should be able to completely disable this privacy intrusion.
How is it possible that this option is not given to us, while we are being ask for a subscription ? If it’s free, you’re the product, but here, even if you pay, your data is scanned, for rather strange and unclear reasons, and your data is collected and send only God’s know where.
Can’t developpers imagine how long a 4.5 TB librarie to scan can take ?
Furthermore, because I think this is the real goal of this scanning, I don’t want to be offered tracks or album suggestions based on what I’m usually listening to. Thank you, I’m a big boy. I don’t stream. The real aim of this scan is marketting, and neither do I need it, neither do I want it. Audirvana should be transparent on this subject, and they aren’t. I don’t want my experience to be enhanced, I don’t want any of this marketting story telling.
I don’t do Tidal, Qobuz. I don’t do MQA, because MQA is pure marketting and a lossy format. I just want a good player for my music. And as I have to sadly admit, that’s what Audirvana was giving to me until this AS mayhem ! But that’s no longer the case.
On a side note, I still would like to know how 2 “alleged” bitperfect streams (Im running MacOs, so it should be bitperfect) with no treatment whatsover, no upsampling, no volume control etc…, can sound so different on a system where nothnig else has been changed between the listening experience ?
Stil scratching my head…
Maybe I’ll get an answer someday. But the silence of the developper on this subject is overwhelming, while maybe it’s the more important thing to consider when dealing with a community of passionate people for whom each aspect of their listening experience is important.
I dreamt of a company were programmers and developpers who address audiophile as they say, are kind of transparent with what they are really doing behind the scene, and why their software is scanning your data. And to what use ?
Anyway, in fact, one can consider himself lucky, sticking to 3.5 for the sake of the sound will spare me some money… even better, some montly money, that I’ll be able to use to buy music. ah ! 
Just downloaded the 1.3 update and compared to 3.5.46. No upsampling or processing.
Same difference as previous: AS has very flat soundstage, instruments lose fine texture. 3.5 sounds more natural, especially voices.
Really scratching my head on how they can be soooo different.
Whatever their ages, some will never grow up and tolerate gracefully other points of view.
We do suffer the culprit for a very little time in our lives, He lives with himself all of the time.
Please let’s show him mercy.
Checking your 12 posts in this thread, I can confirm that a minority of them were “On topic”
I’m confusing nothing.
I was merely pointing out the fact that users have problems, whilst you do not. Yet you seem to take enjoyment in telling those of us that may be unhappy with the performance of Studio how well it works for you.
Fine. Fair play. But if it works so well for you, why take the time in having a go at those of us who aren’t quite as lucky as you appear to be?
And before you defend yourself, do you regard this as a satisfactory response?
That was an interesting contribution, thank you for it.
Such contributions always move me to stay at 3.5 because they have been addressed in a factual and expert manner.
Your description of the sound also seems objective and shows that you know your music.
Since there is another update, the current situation actually shows that it is not worth it at the moment, at least for me to test it at all.
Thanks again
I’ve no problem whatsoever with you attempting to sell Studio to the masses.
But let me get this straight. You think it’s acceptable of accusing a user of crying ( a deregetory term where I come from) for posting a grievance about a piece of software that you apparently adore and works for you, whilst a fair few of us are having problems with, is acceptable?
I’ll give odds of 5:1 of this post being flagged.
Any takers?
The pavlovian flagger has excelled himself 