[DDDamian]

The resolution on this is 0.5 Hz. I challenge anyone to be able to pick out a 0.5 Hz shift in a frequency tone.

That is easily done - remember, it's not just the frequency but the rate of change of the frequency which seperates music from a pure sinusoidal tone.

Re: SQL and the threads open - yeah, you can tell how easy it is just by reading those lol. So much fun migrating from your HTPC's primary drive to a SQL server. Programming not only performs a task it isolates the user from the 0's and 1's....

[branlr]

Regarding all of this audio discussion...

I have to say that this must be provable one way or another through double blind testing, and to disregard such testing is to spit in the face of science. Intuition and vague "feelings" lie to you, and tell you what you want to hear.

But even if it is shown that these differences in audio encoding are perceptible, there is a very real question of whether or not that difference is at all functional.

I have done a bit of audio mixing in my day, and this is what I know--audio sounds very different depending on circumstances:

1. Audio sounds different based on what I have listened to recently, how much, how loud, and in what circumstances (headphones, monitors, live).

2. Audio sounds VERY different when I am tired compared to when I am wide awake.

3. Audio sounds very different depending on system calibration. I pretty much find that I need to calibrate my system to some extent for the mix of a particular movie/album in order to get it sounding the way I'd like it. Not huge changes, but definitely a little tweaking.

To say that a human being could hear that they are listening to this or that codec without a direct comparison is ludicrous as long as any lossy formats are encoded at high quality settings, and a system is calibrated to sound good with that audio source.

And if a comparison is needed to differentiate at all, then I ask--when are you ever listening to something where you are constantly comparing it to another source except to prove a point?

Only audiophiles should be able to tell much of any difference at all, and I would be willing to bet that most would be surprised at their high rate of failure in differentiating audio sources under the right conditions.

Read the early hydrogenaudio forum debates about why 16bit/44.1khz is all the human ear can decipher. Then pop in a CD followed by an SACD or DVD-A of the same album. Night and day, and not solely due to the mastering.

I don't know how you can make this argument if the material has been remastered. How could you possibly attribute the difference on a given recording to media when it has been mastered differently? It is arguing that you can listen to multiple recordings, and average together the aural similarities while compensating for unknown differences in mastering to form this generalized opinion, and I just don't think that humans are really capable of that sort of long term comparison. Our perception gets too distorted over time--and of course, the unknown factor of how the audio was manipulated during mastering...

In the end, a simple test would prove this one way or the other. Even if the audiophiles being tested could be fooled by adjusting the settings during playback, you have already proven that we aren't actually arguing about fidelity, just a very vague feeling.

I think that it is better to pursue something that "sounds good" rather than something that "sounds the most like whatever the source supposedly actually sounds like." The fact of the matter is we all hear things differently, even moment to moment we ourselves may hear something differently. Then add different A/V equipment, different room, different situations leading up to listening... etc...

All of these things change the experience substantially.

[DDDamian]

I don't know how you can make this argument if the material has been remastered. How could you possibly attribute the difference on a given recording to media when it has been mastered differently?

Simple, and I've done it. Take the same source - in my case it was Neil Young Harvest at 24/192. I picked that as it is two channel and I know it intimately. Resample it to 16/44.1 and stitch that in. Re-encode it to MP3 at, say, 320kbps and stitch that in. Use Audacity - it's free.

It's like throwing a blanket over your speakers at each drop in fidelity. Look up "fidelity" and you'll have a good idea what's going on.

You've now eliminated equipment and mastering. Before you claim it's the resampler or the MP3 encoder consider that this is exactly what masterers do before they make a recording CD-ready (dumbed-down) from their higher quality 32/192 sources, or what your ripper does before you send it off to your HD or portable player. Same chain of fidelity reduction, same master, same equipment: audibly different result.

This debate has been going on forever on audio forums (like LP vs CD) but really, high fidelity sounds better.

And the users of those forums - many of whom would be here too if AE happens - love when their favourites are released in hi-def or are re-pressed using better technology.

It's not an audiophile snob who walks into Best Buy and wants a decent system. He wants to enjoy good media reproduction in the man-cave. And how many haven't been thrilled with the sound when they bought their first SACD, DVDA or BluRay? Why? It sounds better.

To return to the thread intent, my hope is that XBMC advances to the sonic quality of 10 year old technology (SACD or DVDA) and further on to 5 year old technology like BluRay.

The CD has turned 20....Really guys, just listen to better quality sources and you'll believe without having pull out your 8-tracks. The difference is soooooo obvious and sounds so good.

[natethomas]

Alright, here's my compromise.

DDD, when you have your system professionally tuned and ensure that the walls around your system are sound proof, and then you perform a double blind test, in which somebody else plays multiple HD tracks and 1.5 mbps DTS tracks (that are exactly the same in every way except for how they are encoded), where neither of you know which track is HD and which is not, and then, when you are able to pick out the HD tracks and have confirmation of your picks after the fact, I'm willing to grant your argument. If, after that point, you were not able to pick out the HD tracks, then you will be forced to concede my argument.

Until then, as we say in America, we are all just pissing in the wind.

That's a fairly simple request. I'm going with the scientific stance: It isn't real until the facts say it is real. The null hypothesis is that there is no difference between DTS at 1.5mbps and DTS-HD for you. The simple experiment described above is the only (and best) way to reject that null hypothesis. Go!

[DDDamian]

My test was simpler lol - and very easy to hear the difference.

I note you keep going back to the DTS (1500kbps) and DTS-MA example. I haven't tried that as they are both movie codecs - not used for music. That makes them very hard to assess for sound. Will Smith still sounds like Will Smith, although the explosions have much more dynamic range with the DTS-MA lol. Not to mention the increased soundfield 7.1 DTS-MA brings.

I'm talking about music, which is my main use of XBMC - until I play anything better than a CD-quality flac. There the differences are very obvious. Like a said before - like throwing a blanket over your speakers.

Ah well - put in my hopes for where XBMC goes. MPC can do it, WMC can do it, Foobar can do it, Winamp can do it. XBMC wants weather and window-dressing. I think the most I've ever watched a single movies is about four times for TLOTR, whereas I have albums or songs I've heard hundreds of times. Why? Ear-candy has more lasting power than eye-candy (except in porn lol).

And I've heard way more people say "Wow that sounds amazing" than say "Wow that looks amazing" when they watch BluRay. When it's SACD or DVDA people tend to say "Wow, I've never heard that bit before" when they've known a song all their lives. These are people who listen to clock radios without flinching, not audiophiles.

What SACD's or DVD-A's do you have that you can't hear the difference on?

[darkscout]

You are a very minor outlier of XBMC users.

People that want dedicated music playing app should probably stick to apps that were designed for music playing.

[DDDamian]

Fair enough. At least until more people hear good quality over MP3's that is likely true. And that's currently what I do: XBMC for old-tech vids and MP3's, Foobar or Winamp or the Denon Universal player for good sound, and the standalone BR player for good vid & sound. The man-cave thanks me and I love it. I got XBMC for the eye-candy but didn't get the ear-candy. Thought it was a current media player too.

Get the binary add-ons working and some kid in China will have the quality stuff working in a flash as happened with the "other" software.

As long as XBMC wants to play on a cellphone gnif will have to jump thru more hoops than his time obviously allows to get XBMC audio up to the 2005 era.

[natethomas]

Ah, see, we're clearly having two entirely different discussions, and given the current dev team, I imagine that's going to be the case for a while. The vast majority of current devs, excluding possibly spiff and maybe one or two others, use XBMC primarily for video. We watch the heck out of movies and TV, and tend to relegate music to secondary status. I, personally, almost exclusively listen to music via headphones.

One of the downsides (if you want to call it that) to open source development is that developers work on what interests them. They aren't getting paid to work on things that don't. Foobar and Winamp are pieces of software developed by people love music to the detriment of all else. XBMC is a piece of software developed by people who appreciate music, but care an awful lot more about seeing and watching a car explode, than picking out the timbre of a perfectly tuned violin.

So if you care about the musical side of things, you have two options. Either you can wait, patiently, for a developer who does care to finish his work, or you can personally roll up your sleeves and do it yourself. That's pretty much it. Complaining about the software and making negative remarks about development cycles may very effectively get rises out of people, but it's not going to result in software going the way you want it to go.

Edit: I should probably also note that you lose a lot of credibility in your last post in two different places. Snake oil buying audiophiles always seem to have this belief that everyone and everything agrees with them on the miracle of better audio, to the extent that they say weird things like, "the man-cave thanks me," and they begin going down that insanely annoying road that all hipsters go, wherein they assume that people who don't understand must be idiots, without ever once thinking, "Hey, maybe the reason other people don't understand is because there really isn't anything to understand, other than the magical construct that's in my head." Thus, cloaked in this imaginary gauze of invincible pigheadedness, audiophiles proceed to disparage and put down the very people that they demand help them for free by saying things like, "As long as XBMC wants to play on a cellphone." Somehow, audiophiles don't comprehend that saying things like that make the developers, who are only doing this work because they enjoy doing it, contemplate going the exact opposite path from that which the audiophile wants.

When I read your cloaked put-downs, my immediate reaction is to suggest a complete halt of all audio development. My secondary reaction is to remember that there really are people who like having decent sound, who aren't suffering from some kind of moronic, non-participatory sense of entitlement, so I calm down.

gnif will finish his baby when he can. At that time, many people will be thankful. And you will finally get what you appear to think you deserve by birthright.

[davilla]

I once spent weeks tweaking speaker positioning to achieve that perfect sonic stage. That was many years ago and quite frankly I just don't have the patience for those games anymore. There's just no way to really replicate the sounds and presence of the real, physical thing. You can get close but that's all you can do. So it becomes a waste of time, waste of money for the never ending pursue of perfection. As long as it does not sound like crap, I'm happy.

[GreenEyez]

I once spent weeks tweaking speaker positioning to achieve that perfect sonic stage. That was many years ago and quite frankly I just don't have the patience for those games anymore. There's just no way to really replicate the sounds and presence of the real, physical thing. You can get close but that's all you can do. So it becomes a waste of time, waste of money for the never ending pursue of perfection. As long as it does not sound like crap, I'm happy.

Well, today, you don`t need to. Modern AVR receivers have Audissey MultiEQ room correction, with a provided calibration mic, which does it for you, 15 minutes to run the setup and you get much better audio with a calibrated sistem, no matter if the audio is mp3 or DTS-HD, the calibration is very good and improves all codecs.

[GreenEyez]

Fair enough. At least until more people hear good quality over MP3's that is likely true. And that's currently what I do: XBMC for old-tech vids and MP3's, Foobar or Winamp or the Denon Universal player for good sound, and the standalone BR player for good vid & sound. The man-cave thanks me and I love it. I got XBMC for the eye-candy but didn't get the ear-candy. Thought it was a current media player too.

Get the binary add-ons working and some kid in China will have the quality stuff working in a flash as happened with the "other" software.

As long as XBMC wants to play on a cellphone gnif will have to jump thru more hoops than his time obviously allows to get XBMC audio up to the 2005 era.

Just to let you know, you can use an external player in XBMC, like MPC-HC with LAV filters to passtrough any audio type you want, ranging from DD to DTS-HD (what i do every day). XBMCs internal player isn`t the end to all means .

[DDDamian]

@greeneyz - Isn't Audessey excellent? Works magic with very few needs to tweak afterwards. Even my older Denon 3805 had a decent self-calibration, although not on par with Audessey.

The fact that with a sub-$2,000 speaker setup I can appreciate the difference in high-def audio means many of us can enjoy it. Hardly audiophile gear!

I've tried the external player route (including DSPlayer builds which in effect gave you binary add-ons via FFDShow) but it does make thing a lot more clunky, and of course DSPlayer is no longer supported. Using MPC for the same purpose is fine for movies 2hrs long but less appealing for songs where you're in-and-out of the XBMC interface.

Well, I've put in my request for the original thread title here and been resoundly spanked by some devs who are more video-oriented (although not TrueHD or DTS-MA oriented). I thought my OP was on-topic whether is was for AE, binary add-ons or server/client architecture. Binary add-ons would allow other devs to work on the audio issues outside the need to keep things cellphone-friendly.

At the risk of - clearly stated - spoiling the pot for those who count music as media I'll bow out and hope AE doesn't go the way of DSPlayer.

Aside from my moronic, self-entitled, audiosnob 2-cents worth on the hopes for the future of XBMC I do think it's a nice front-end and hope the development continues.

[Drae]

You are a very minor outlier of XBMC users.

People that want dedicated music playing app should probably stick to apps that were designed for music playing.

To be fair I imagine (no proof!) that you could class iOS users as outliers compared to the installed Windows/Linux/Live/OpenElec base but that's still getting a lot of attention.

At the end of the day XBMC is essentially about three things; Video decoding/presentation, Audio decoding/presentation and a nice UI to wrap it altogether (I lump in organisation/scraping with that). It's the combination of all three that imho place it head and shoulders above pretty much everything else out there (that I'm aware of anyway).

Audioengine will bring audio right up to date - be it bitperfect (least in terms of what your amp allows) audio, lack of fiddling with asound.conf's or bitstreaming HD formats. With regard the later the simple fact is regardless of the science if another product comes along with TrueHD and DTS HD Master/HR slapped on it the average "layman" will likely consider it a superior solution and use it.

Personally I look forward to Audioengine and many many props to Gnif, Anssi and co. for working on it as he/they have. I sincerely hope that once Eden is out this kind of discussion (who needs DTS HD?! who needs 24bit?! blah blah) will cease to occur - because frankly it's a disservice to their hard work - and that instead everyone will get right behind its implementation. Same goes for an update to ffpmeg

I'll now throw in the usual but none the less very true - many thanks to the really very dedicated developers of XBMC for ALL their hard work. Having been heavily involved in the development and management of a fairly well known web based app I can appreciate all the issues that you put up with day in, day out. For me at least it's VERY much appreciated, thanks.

[darkscout]

To be fair I imagine (no proof!) that you could class iOS users as outliers compared to the installed Windows/Linux/Live/OpenElec base but that's still getting a lot of attention.

One of the downsides (if you want to call it that) to open source development is that developers work on what interests them.

Emphasis mine.

[Drae]

As I say - having worked on an opensource app for a long time I'm well aware of that fact That however doesn't detract from the point I was making. It is simply wrong to go around claiming "well X isn't important you're best off using Y instead" - when it's quite possible to point at various aspects of a project and say "well imo Z isn't important either". It's swings and roundabouts. I'm not suggesting B0n0bo should learn all about audio decoding and give up on perfecting the video functionality. Or Opdenkamp give up on cec and pvr (another thing I'm following closely). I _am_ saying all XBMC users should be right behind Audioengine.

Audioengine isn't an "outlier" it's intended to replace one of what I consider to be the three core aspects of XBMC. Regardless of whether someone thinks HD movie codecs are necessary or not - they exist, they are popular. Regardless of whether someone "believes" they can hear a difference between 16bit and 24bit, 96KHz or 192Khz - such differences in opinion exist and should wherever possible be catered too. And with Audioengine they can and hopefully will be - as Nate says above, when it's done, if it's done such arguments cease to be an issue.

These ongoing (not limited to this topic) discussions of "Well DTS HD is pointless" imho - and as someone who's had to put up with similar comments levelled at my own work - are frankly ball breaking to the peeps who are working on said projects. If I were Gnif I'd actually be thinking "why do I bother" and move on - that's of course just me, I imagine Gnif has a much stronger constitution than myself