Equipment vs. encoding

[blueturtl]

So after reading and replying to the nth Windows vs. Linux thread I thought I'd like to discuss something else for a change.

Anyway, I've been wondering for a while about encoding vs. equipment, or more spesifically, which is the bottle neck in high quality audio in home computer systems. I used to think that fundamentally a lossy encoding, like MP3, was the main culprit in reduced fidelity.

However recently having purchased a new soundcard for my HTPC I noticed that in fact the ceiling still seems to be pushed simply by upgrading the hardware responsible for digital-to-analog conversion. The source files regardless of codec quality all sound much better than they did before. The difference is much more pronounced between soundcards than it is when comparing audio files of different encoding methods or bitrates (at least this has been the case for me). This applies even to lossless formats such as FLAC!

Of course some effects are accumulating; eg. a low quality MP3 will sound better on high quality equipment than on low quality equipment. Still, I'm surprised that even 128 kbps MP3s still seem to clearly benefit from the higher articulation and reproduction of higher quality DACs.

All lossy algorithms are after all based on the idea of removing sounds masked by other sounds, so in theory we shouldn't really be able to pick up the difference between the original and the encoded version. It seems the poor quality reproduction of most speakers and audio cards removes a bigger audible portion of the sound than even aggressive encodings do.

Of course my equipment is not really high end (I know true audiophiles can blow tens of thousands of euros on just speakers), but I am still to find the limits of the speakers. Every time I've upgraded the sound card the speakers have shown they can infact do better audio.

I'd love to hear what some of you enthusiasts (music or hi-fi) think of this? Does everyone simply use high end headphones (much easier to get good quality than speakers for the money)? Can someone confirm what I've found or do you have opposite experiences?

Why waste disk space on lossless encodings when most of the computer audio hardware can't even flawlessly reproduce lossy encodings?

[LowSky]

I'm no audiophile. But my issue is more in the highs and lows produced by encoding. Its really annoying when a song or a video sounds to low when there isn't much going on and then too load when there is a good deal going on. Its very annoying having to adjust the speakers while listening or watching something.

[cchhrriiss121212]

I was not aware there was a bottleneck in high end audio on home computers. Get yourself a card with digital output then hook it up to a decent up-sampling DAC by Cambridge Audio or something and enjoy.

[blueturtl]

I was not aware there was a bottleneck in high end audio on home computers. Get yourself a card with digital output then hook it up to a decent up-sampling DAC by Cambridge Audio or something and enjoy.

What I mean is that it seems somewhat silly to see the same people who weigh MP3 Ultra preset files against FLAC still use integrated audio on their mobos hooked up to what is at best a mediocre speaker system like the ones most commonly on sold with PCs today. It seems to me upgrading the equipment brings much more fidelity even to low quality sound files than swapping between MP3 or OGG or FLAC. This is what I meant to highlite.

The original post was a bit incoherent, my apologies.

[cchhrriiss121212]

What I mean is that it seems somewhat silly to see the same people who weigh MP3 Ultra preset files against FLAC still use integrated audio on their mobos hooked up to what is at best a mediocre speaker system like the ones most commonly on sold with PCs today. It seems to me upgrading the equipment brings much more fidelity even to low quality sound files than swapping between MP3 or OGG or FLAC. This is what I meant to highlite.

The original post was a bit incoherent, my apologies.
OK I see what you mean now. I would say that one advantage of FLAC is that you will have a direct copy of the source material. So ripping all your CDs to FLAC will give you a back up of your collection in addition to being playable on your average sounding desktop system.

What kind of high-end headphones are you using BTW?

[eriktheblu]

To an extent that is true. An MP3 encoded at very low sample rates will demonstrate it's inferiorities on all but the cheapest of sound systems.

Unfortunately, while you can always upgrade your sound equipment, you can never improve the fidelity of a recording. Don't rip to FLAC because it sounds better on your system, do so because it might sound better on the system you have 10 years from now (or perhaps the system your friend has when you share your creative commons licensed music).

[blueturtl]

OK I see what you mean now. I would say that one advantage of FLAC is that you will have a direct copy of the source material. So ripping all your CDs to FLAC will give you a back up of your collection in addition to being playable on your average sounding desktop system.

What kind of high-end headphones are you using BTW?

I don't, I actually do use speakers. Huge Cerwin Vegas in the livingroom and Boston Acoustics BA635 in the office. While a real audiophile might cringe at these (I consider myself an enthusiast at most), I've found they are still not the bottleneck. Every time I swap a new soundcard there is an audible difference in sound quality. However distinguishing between say a 160 kbps OGG file and a FLAC file I can't say there is really a difference at all.


An MP3 encoded at very low sample rates will demonstrate it's inferiorities on all but the cheapest of sound systems.

That is what I'm sort of calling to question here. My subjective experiments indicate that there is plenty of headroom on all but the lowest quality recordings (I count 128 kbps the limit), meaning changing the speakers or soundcard will show improvement before changing the codec or samplerate does. One would have to have really high end equipment, or maybe good quality headphones come cheaper than do good quality speakers.

[bonzini]

In my experience, you have a point.

Generally my ripped audio files are FLACs. I have some MP3s that were give-aways eg from SixDegreesRecords (wonderful people, wonderful music by the way) and from download coupons that come with LPs I have bought in the last few years. By and large these don't sound too bad on my pretty decent audio system.

I believe I can hear the difference between the same songs I have in MP3 and FLAC (where I've bought an album based on an MP3 promo).

I believe my LPs sound better, generally, than the same music on CD as well. And, I have a good CD player and a good turntable / cartridge.

But there is no doubt in my mind whatsoever that better equipment dramatically improves the sound.

Of course, determining what is "better" than what you have is easy in some cases, not so much in others!

Yes, my HeadRoom TotalBitHead is way better than the sound card built into my 2005 Toshiba laptop. But the kids' Toshiba laptop through the Pro-ject USB audio box into my Linn pre-amp - that's not noticeably different than my TotalBitHead.

Probably because the Pro-ject USB audio box is not dramatically better / worse than the TotalBitHead.

Having said all of that, a few years ago I went crazy and bought a second power amplifier and a set of active cross-overs and rewired my speakers to take the passive cross-overs out, and then I spent all my spare time over the next few weeks re-discovering my music collection.

What a huge difference that made to the small details in the recorded music; and how much more clear was the sound when the volume was set pretty high.

Having said that, I return to my earlier comment - determining whether a new component is really better than an old one - that's not always so easy, and it may require you to sit down over an extended time frame and listen carefully for differences.

Or, you may get it right away, like when you plug in your TotalBitHead and compare it to your built-in audio.

Interesting thread!

[NightwishFan]

I generally rip all my music to a lossless format as I am keen on retaining originality at the expense of space. The trick is to decide what is more important; Quality or space. Generally if you are uninterested in preserving or archiving files and just like to listen to music, a lossy format will do just fine. Even an "extreme" quality mp3 file at 320kb/s is smaller than a 700kb-900kb/s flac file. How low you go is merely a matter of how much space you have. I would say 160kb/s is a good size for vorbis audio files.

[AoSteve]

Coming from a guy with two sound cards... LOL


Hardware isn't an issue anymore. The capabilities of todays sound processors is absolutely insane. Take the idea, running my Sound Blaster Live Platinum in 5.1 surround playing a blu-ray. Compared to a Sound Blaster X-Fi Titanium; none. The SBLive is nearly 12 years older.

It's all about the encoding and capabilities of the software pushing the sound.

The biggest thing I've ever noticed with hardware is latency when recording/playback at the same time. I'm not a big time audiophile; but pressing a 5.1 surround system at 980ws with extremely high frequency range speakers is a good way to test the personal quality. I however, can tell the difference between playing a blu-ray disc and an h264 file...