Chesky records, a small record label, produces what is called audiophile friendly music. To learn how to create music with this high detail and how to play it right, just read some of their articles. There is a lot of talk there on tube amps and stereo mics and horn speakers etc. Forget the high-end gear, there is one complete article on how to set up your power supply for best audio experience. But mere mortals like us listen to the music on our PCs. Though it does not even begin to compare with all that exotic gear out there, it can be set up to deliver a surprisinlgy good quality sound.This is a highly simplified diagram of how the music flows:
mp3 file –> mp3 decoder –> processor –> soundcard –> speakers –> Ears
Each one of these stages has to be optimized if you want the best quality sound. If the mp3 file is low-bit rate to begin with, no amount of tweaking can make it sound lush and full. Or if you are using tin boxes for speakers, even a SACD would sound crappy on them. One weak link in the chain and there goes your listening experience.
Input:
MP3 technology was developed by Fraunhoffer Institute in Germany. They also provide software to encode/decode MP3 files and that is what is used in winamp by default. But these official decoders are not the best sounding ones. That honor goes to an open source (surprise! surprise!) decoder called mpg123. It supports ReplayGain and 24-bit output, both of which are lacking from the official decoder. The difference in quality may not be that apparent on ordinary songs but this plugin really shines when the MP3 file contains a lot of detail.
To use this plugin, Just copy the file ‘in_mpg123.dll’ to ‘C:\Program Files\Winamp\Plugins’ and launch winamp. Then under Preferences->input, choose mpg123 and double click on it to set its properties. Under Decoder tab, check ‘Enable’ and choose output format as ’16 bit’ (or ’24 bit’ if your sound card supports it). Under Replay Gain tab, check ‘Enable’, choose ‘Album gain’, and choose ‘Hard Limiter’. ReplayGain will make sure that all the songs play back at the same loudness, thus freeing you from changing the volume for different songs which had been mastered at different levels. No changes are made by ReplayGain to the original MP3 file. It just puts a tag on the file which is used by the decoder to decide whether to amplify or attenuate.
An alternative MP3 decoder called MAD also has a lot of following. There is no conclusive evidence that it is better than mpg123 but arguments go on.
PS – The official mp3 encoder is also not the best one out there! That honor goes (again) to an open source encoder called LAME.
Processing:
After a mp3 file gets decoded, it is fed to a Windows component called kmixer. It does exactly what its name implies – it mixes all the audio streams being fed to it and sends them to the soundcard. This is how we can hear those annoying Windows event sounds alongwith “sweet home alabama”. If this were the only thing that kmixer was doing, it won’t be so bad except that the kmixer resamples everything, just to make it easier for it to mix all those audio streams. It also fiddles with signal to noise ratio and does many other undesirable things. The end result is that the audio stream comes out sounding very flat and dull. So we need to find a way to bypass this mixer thing and send the audio stream directly to the soundcard. Enter kernel streaming.
Download this plugin and just copy it to ‘C:\Program Files\Winamp\Plugins’. Then choose ‘kernel streaming’ under Output preferences. There are no settings for this plugin (as yet). Since it bypasses the kmixer, the winamp volume control won’t work with this plugin. You have to use that little speaker icon in the task bar. This is also way too loud at normal volumes, so keep the volume to a minimum when you try it for the first time. In some cases, when you double click on a file after choosing kernel streaming, the winmap might quit. This is normal. Just relaunch the winamp and it should run fine from there on.
Output:
This section applies only if you have Creative soundcard or a soundcard based on 10K2 chipset.
After a mp3 files gets decoded by winamp, it gets fed to soundcard drivers. These drivers make sense of this audio data and convert it to a format suitable for the soundcard to play. But the drivers supplied by Creative are a bit crappy and bloated. So head over to kxproject homepage and download these drivers. These drivers have a cleaner sound ouput and are much more stable than the stock ones. You also gain much more control over your soundcard settings.
Ears:
Finally, the most critical link in the sound quality chain – the ears. If you cannot tell the difference between an MP3 file and the original CD it was encoded from, don’t despair. Most people can’t. It takes time and practice and a really good audio setup to hone the listening skills. I leave you here with some commentary from an audiophile who does not just hear music, but actually listens to it. Be inspired!
Credits:
The article that made me interested in all this fiddling is here. It deals specifically with foobar2000 but the concepts are same. Infact foobar2000 comes with kernel streaming preloaded. If you can stand its UI, it is said to be the best audio player around.
Manish Bansal is a mainframe programmer by day and PC junkie by night. His other hobbies are Graphology and Photography.
If you would like to see your thoughts or experiences with technology published, please consider writing an article for OSNews.
…why focus so much on mp3 when Ogg Vorbis sounds 10x better? I just don’t get it. Someone enlighten me.
Why focus on only windows 😉 Im curious how this stuff works in Linux and Alsa.
All of this is pointless when you are listening to your mp3’s on the crappy speakers that came with your pc.
I doubt It’d make much of a difference even on the highest end pc speakers.
But if you’re running your pc through an amp to real speakers than this is well worth the effort.
Don’t forget Vorbis, AAC, Musepack.
BTW what’s a C: and what are its dependencies?
Anyone know where I apt-get it from?? 😉
BTW what’s a C: and what are its dependencies?
heh.. the world would be better if everyone asked that questions
I agree with all the above comments. I would also like to point out that good quality headphones beat any speaker system hands down. However, can anyone recommend some good, relatively cheap, portable headphones? I was looking at the Shure ones, but they are kinda pricey. Suggestions? thx
I’ve got a c: in ~/.wine/dosdevices/
😛
Why use MP3s? Real audiophiles use lossless audio. And Kernel Streaming makes my winamp stop working; foobar works fine, though.
this is a nice article, which i found useful.
but u can’t beat live music!!
Grado Labs SR80s headphones
http://www.gradolabs.com/product_pages/sr80.htm
is still crap in FreeBSD. I was in the market for some good audio card but they seem to be supported in Linux/Alsa but not FreeBSD. There are commercial OSS drivers, but will they even compile? So far it seems the “best” card is Audigy something, it’s still not what I want. So really, you can tweak your power supply all the way, but sound-card induced harmonic distrotions will stay regardless.
That’s why the best idea is still standalone players, such as this one: http://www.rokulabs.com/ but even this still needs a separate standalone DAC to really deliver good quality.
The cheapest Shures (e2c) are very good value for $60-$70. You can get cheaper stuff but none of it’s better. There’s Sennheiser PX100s and Koss PortaPros in the $20-$30 range, and Sharp MD33 or Sony EX71 / EX81 for cheap canalphones, but e2c are worth the money.
Really interesting article but you sure you ran it through a proof reading before submitting?
<em>…….. the winmap might quit. ……..</em>
Even with most stereo speakers these days, unless you’re paying major amounts of cash, then the differences between a high bitrate mp3 (192kbps or better) then you’re not going to notice much of a difference between that and a CD. Hell, with the low end kit you can’t even tell between a 64kbps mp3 and a CD… but I digress.
However, for those of us who listen to nothing except Flac encoded music, whose speakers pricetag could be mistook for a phone number then yes, this is a great article. For most of us out there, even if we have reasonable equipment, it probably wouldn’t get any noticeable benefit from the things suggested here.
Want audiophile quality from your PC? Ain’t gonna happen. Burn a CD, haul it over to some quality equipment.
To the person that says decent headphones beat any speaker, you’re wrong. You’re not listening to the right speakers (amp, pre-amp, cd, cables, etc).
RE: Burn a CD
While the speaker setup might beat a headphone setup in quality, better quality can be achieved for a much lower price with headphones.
@Luckett
The grados mentioned above are very good. I like my Sennheiser HD 570s which I got for $80. Take a look at http://www.headphone.com for comparisons and such.
Despite all of this listen to music for the music, not for the quality of the equipment.
I do not see it mentioned here, but audio quality can be improved by powering your PC and equipment from an online double-conversion UPS, which filters all the nine common power problems (as per the IEEE definition) and gives power off a battery at all times. In one notable case, I had a low-end computer motherboard that produced audible static noise when WinAmp was used, but which curiously wasn’t present when Windows Media Player was used. With the UPS, the static noise in WinAmp disappeared and played just as well.
It also makes other perceptible improvements in the sound.
“Despite all of this listen to music for the music, not for the quality of the equipment.”
That’s sage advice. I would say that a 16-bit card and stereo speakers for a couple of tens of dollars provide really a lot, although manufacturers of newer 24-bit, etc. equipment would like to convince us otherwise — IMHO largely to improve their profits.
Winamp couldn’t play .mod-files after installing the kernel output plugin.. so no more of that shit.
Use OSS.
Will it compile? Uhm, it’s binary and comes with a installer.
That “mpg123” mp3 decoder sounds very interesting.
Does there anyway to use it with my iTunes?
Buy an amplifier which can act as an USB audio device. Then you will get a pure digital signal out of your PC and have the amplifier do the D/A work. No PC noise!
I disagree. 24 bits is a lot better as soon as you start mixing more than one track together, otherwise everything sounds overcompressed.
Only use a SP/DIF output, an a cheap A/V receiver. You’ll get real quality, and upgrade is very easy : when you get more money, change the receiver, or the loudspeackers.
Something doesn’t quite add up here. Why would you go to all that effort to get Winamp to skip kmixer and change your sound card drivers etc for some pretty subtle changes in quality, and then play mp3’s?
As someone said, Ogg Vorbis sounds better, but if you’re going to go through all this surely nothing less than FLAC would be acceptable?
“Winamp couldn’t play .mod-files after installing the kernel output plugin.. so no more of that shit.”
Winamp couldn’t play modules for toffee anyway. If you must use windows, then use the oldsk00l winamp plugin with built the emulation core from UAE built into it. Make sure to delete the standard winamp module plugin first.
http://fuel.adsl.dk/Oldsk00l.htm
For the *nixes, OSX and MorphOS, use UADE which does the same.
http://uade.ton.tut.fi/
Sennheiser HD 565 (dunno if these are still in production), sounds crystalclear and has these IMHO extremey comfy earpieces I’ve never seen on other brands. Plus, It’s a open design, so you still hear something from the outside.
“If this were the only thing that kmixer was doing, it won’t be so bad except that the kmixer resamples everything, just to make it easier for it to mix all those audio streams.”
This is not really true, and misleading.
Because…
Kmixer does not need to resample if the sound card’s sample rate can be set to the same rate as the audio data. (It will always try to clock the card to the correct sample rate to avoid resampling.)
If the sound card cannot be reclocked (It’s cheap and fixed at 48k) then *any* driver will have to resample to play files at other sample rates.
It’s impossible to digitally mix audio streams of different sample rates without resampling. It’s not the ‘easiest’ way, it’s the *only* way.
I can, however, well belive that Microsoft screwed up the mixing algorithm and lost a couple of bits of dynamic range. That’s a seperate issue from resampling though.
The situation on Linux is perhaps a little better, but only really because of the stubborn nature of the ALSA developers. They refuse to let ALSA do the ‘wrong thing’ when faced with the possibility of playing multiple files at different sample rates, so the card simply blocks new requests until whatever program was first using it has finished.
This is the default behaviour, and so most people use dmix or a sound daemon to make a more usable system. If the card supports hardware mixing, ALSA will use that, and the results are often better than software src.
Even with most stereo speakers these days, unless you’re paying major amounts of cash, then the differences between a high bitrate mp3 (192kbps or better) then you’re not going to notice much of a difference between that and a CD. Hell, with the low end kit you can’t even tell between a 64kbps mp3 and a CD… but I digress.
64kbps… dude, if you can’t hear the difference in distortion and bass then you gotta check you ears. It’s such a difference that a low end kit makes it even more obvious.
192kbps MP3s, now to start with that’s gotta be a good quality encode done with ABR otherwise you’re gonna hear the difference quite often. It is quite easy to do tests yourself. Put two versions (one FLAC/APE and the other 192kbps CBR) of a tune on a playlist close your eyes and hit the “Next” button a lot of times (no counting). Now listen and try to determine which version is which and you’re gonna get it right quite easily. I’m not saying 192kbps MP3 CBR is bad, it’s just not CD-grade. 192kbps Ogg or MP4, that’s a different issue but the difference is to me unhearable.
128kbps Ogg or MP4 is still good but NOT CD-grade. 128kbps Ogg or MP4 is an excellent level of compromise (size/quality) for portable players tho.
But it is embarrasing that an audiophile company even mentions MP3 (except if only to declare how “bad” it is).
MP3 is NOT acceptable any more for lossy audio. Ogg Vorbis and MP4 have much better size/quality ratio than MP3 and that is all that SHOULD matter when lossy audio is the issue. If size is not an issue then FLAC/APE is the shit (and I hope you all have backups of your CDs in those formats right? I’ve had borrowed returns come back scratched badly and my backups saved me from getting mad).
I guess someone (like quite a few of those “audiophiles”) overlooked the fact what MP3 is actually doing. 24 bit sampling resolution with MPEG Audio Layer III is pointless given the quantization resolution and quantization error introduced in those typical (and useful) bitrates. And beside that, as a transformation coder simply is not designed for the quality scalability to a SNR(24 bit) of 144 dB.
So why the hunt for 24 bit decoders? Again another hunt for pointless numbers by “audiophiles”?
128kbps Ogg or MP4 is still good but NOT CD-grade. 128kbps Ogg or MP4 is an excellent level of compromise (size/quality) for portable players tho.
Whether it is comparable depends heavily on the audio source and ears/headphones/speakers combination you have. I think the best is if everyone do a set of double-blind tests themselves. But yes, at least for my ears, 128 kb/s is noticably different from PCM both for MPEG Audio Layer III and Ogg Vorbis.
The other (rather funny) thing is if you start to rank the quality among PCM and various lossy compressions. There are situations where lossy compressions simply sounds “better” than PCM – and then hunting “CD-grade” becomes pointless.
“Winamp couldn’t play modules for toffee anyway. If you must use windows, then use the oldsk00l winamp plugin with built the emulation core from UAE built into it. Make sure to delete the standard winamp module plugin first.”
Will that work with the kernel output plugin also? I’ll try it anyway. It did. However after playing a module winamp can’t play any more MP3s.
Regarding sound quality I think the KX drivers makes the sound weird at full volume, I might be wrong thought. Are they much louder? Anyway I’ll decreased the volume.
The KX drivers made my VOIP sound quality much better. (less choppy in my end, probably better sound quality in the other aswell, i use a crap mic, anyone could recommend a great headset for music, games and voip, or maybe just a simple portable one aswell. Or is better headphones + cheap headset the only way to go? Price must be quite low).
It sucks that the winamp sound controller doesn’t work with the kernel output plugin, however winamps always on top never work either so it doesn’t matter so much if i bring back the old volume controller to my systray.
And no, I don’t have to use Windows, I do since the WoW beta but will go back soon, to OSS however, and amarok, dunno if it plays modules to. Also my A1200 lies in the corner of the same room.
Regarding bit rates I hear a huge difference on 128 and 160kbps mp3s, dunno about 192kbps vs CD. Setup is SB audigy (what was that with no real 24 bit? please explain, does it suck?) + denon avr-1801 + b&w dm603s2.. i also have 601s2 surrounds of which my cat floored one yesterday, i hope it’s ok anyway
, should have taped/glued them but i didn’t wanted to, but it’s a much better thing to do then have broken ones.
The most important factor in music listening is the Price-Tag Effect. This is where deluded folks think you will get great sound by spending huge bucks. The reality is most people are as deaf as posts and can hear sfa above 11KHz. It is impossible in a home situation to get anything approaching proper sound because of shocking acoustics and the high ambient noise levels (40dB+) in most houses. Studio recordings sound nothing like live music anyway.
There are quite a lot of high-end sound alternatives for the PC. The best bang for the buck is prolly Chaintech AV710 and EMU 0404, while the best quality alternatives are EMU 1212M, RME HDSP 9632, Lynx L22. After that the way to go is probably using an external DAC.
Noteworthy is that in the same price range, pc gear beats dedicated audio gear hands down, if configured well. I much prefer foobar over winamp. It gives more detailed sound, prolly due to higher inner precision and a better piping process going through less components. I also prefer the modded mpg123 decoder of foobar over the winamp fraunhofer one and MAD. Also, if you have a AC97 sound card (which includes all onboard and creatives cards, even Audigy4 pro), the card will force upsample of 44.1khz to 48khz internally, often with a lousy algorithm that optimizes speed over quality. In that situation, using a software upsampler like SSRC and thus bypassing the hardware upsampler will improve the sound a lot.
IMO vorbis doesnt sound better than mp3, just different. It has more detail in mids but suffers in the highs, also due to being vbr, the music gets a funny character, not resolute enough, as if it cant make up its mind. It is more dynamic than mp3 though. Personally I prefer mp3.
I can’t really agree on the difference between mp3 and lossless only being obvious when using high-end gear. It depends on the sensitivity of the gear, but not nessacarily on the quality of the gear. The difference is more apparent in winamp than in foobar though, due to worse decoding.
“MP3 is NOT acceptable any more for lossy audio. Ogg Vorbis and MP4 have much better size/quality ratio than MP3 and that is all that SHOULD matter when lossy audio is the issue.”
I’ve seen this claim a few times, but I haven’t seen it backed up with evidence. Are there any listening tests that actually prove that there’s a big difference in quality between MP3 and other formats?
From what I’ve seen the difference between Ogg, AAC and LAME –alt-preset MP3s is pretty small. With the advantage of being supported by far more hardware and software, MP3s still seem like the best option.
Its kinda hard to measure the size-quality ratio really. The difference between mp3 and newer formats is much like that between svcd (mpg2) and divx (mpg4). I presonally much prefer svcd over divx, is much more consistent, with constant quality and no drops when things get messy (fast movements).
1. Don’t use on-board sound chips, they’re crap.
2. Get a sound card supported by ALSA. Most cards are supported.
3. Turn PCM volume down to about 70%.
4. There you are, with OGG and FLAC goodness installed by default.
Don’t bother with the EMU/Creative cards (It’s the same company since Creative bought EMU).
You can pick up an M-audio audiophile 24/96 new for around £60 nowadays, and it will blow most consumer cards out of the water. It has no headphone outs, no preamped mic in, no 5.1 surround, no analog input level vca, it’s crap for gaming, but…
It actually does what it says it does, with no resampling required and decent ad/das. It can also pass *bit accurate* data over it’s spdif i/o and be externaly clocked. That means you could digitally transfer data to DAT and back without a single bit being changed.
While most people may think that’s an esoteric requirement, only useful for recording studios, I think the bare minimun you should expect from a sound card is that it should not change a single bit of the audio data unless you tell it to.
Unfortunately I should have written an article or two about my experiences with PC audio but I have gotten caught up in the rough and tumble of daily life. Here’s my suggestions and what I use and do with my music.
All you really need is an external USB Soundcard to isolate the audio signals outside of your computer case. The cards all work converting the signals into audio outside of the noisy computer case. So you won’t get any static when you move your mouse or when your harddrive spins.
Simply hooking up the RCA outs to a stereo should be fine for mid-fi setups with a receiver or something similar.
I went a step further. I took the high-end route and used the fiber optic out of the soundcard and put that through a Digital to Analogue converter (DAC). What this does is bypass the rather cheap DAC inside the soundcard that cost about $29 to make. If your system is revealing enough you will notice a difference.
Don’t want to get into too much detail but there is less flutter than when using a Compact Disk.
File format should be lossless, I use FLAC and sometimes Windows Media Lossless. I have also tried Real Audio Losless and that sounds the same as the other two.
Thanks for the good article.
REAL audiophiles buy tickets.
For classical music, not only to all compressed formats suck, sometimes even non-compressed ones do as well. Some types of voices for example, are just too big and too rich in overtones for ANY recording to sound like a live performance. It’s amazing what is lost even in a fully professional recording.
There are definitively. The statistically most significant one I know is done my the German computer magazine c’t performed few years ago, which included several thousand (6000?) double-blind samples from its readers. In that test, Ogg Vorbis is among the best lossy compression available, while MPEG Audio Layer III the worst, and that includes even the MP3Pro for the streaming typical bitrates.
The quality of a sound card is important but the quality of the driver is not be underestimated. The best available sound drivers today are in ALSA: http://www.alsa-project.org
http://stereo-link.com/ ‘Nuff said.
i use mp3 for 2 thinks:
listen to music on the go. with all the background noise from the street you dont even hear the difference between 128 and 160kbs
and when i listen to it on my pc
i use sennheiser hd650 at home. because of its open design you always hear the noise from your pc. again a losless format doesn’t provide better quality.
and when i realy want to listen to music: i power off everything i don’t need in my room. connect my hd650 to my yamaha cd-player and just relax
>Why focus on only windows 😉
You can’t expect every article to include every OS? If you do, I expect you to ask the same question next time an article with focus on Linux is posted.
Thanks AdamW, I think ill go with those Shures, and a set of the Grado SR-80’s when I have the money.
Sweet.
“Want audiophile quality from your PC? Ain’t gonna happen. Burn a CD, haul it over to some quality equipment.”
Crap. I know several audiophiles who have a PC source. A source is a DAC. You can get lots of good DACs to put in (or outside, with a USB external DAC) PCs; they sit on high-end soundcards. Then hook it up to the same preamp / amp / output stage you would any other hi-fi system.
Most people I know who’ve heard both rank the e-mu 0404 over the m-audio 2496. The 192 is getting nice reviews, though.
I have SR80s myself, but I don’t think they’d be a _really_ significant upgrade from e2c’s, and they’re not intended for portable use (though you can use them that way if you don’t care about everyone else in the bus / train / library / whatever :>). If you really want to get an upgrade save your money for something significantly better, like a higher pair of Shures, Ety ER4, or $150-$200 headphones (Grado SR225 would be a more sensible upgrade as Grados go). If you don’t want to spend that much just stay with the Shures and be happy.
Just to throw in my 2 cents, AKG K141 headphones are pretty standard studio headphones. They are very comfortable and sound great. They aren’t even that expensive, around $90 I believe. For the price range they are what I recommend.
OGG is only better than MP3 for equivalent bitrates. I have plenty of MP3s that are 256k, 320k and VBR that sound great. The size maybe larger than OGG of similar quality (lower bitrate) but I can play all those MP3s on my PC, DVD player and portable players.
No mention in article on Winamp DSP/effect plugins. I’ve tried SRS WOW and am currently using Enhancer v0.17 (dsp_enh.dll).
If these open source decoders and encoders are available (and free?), why are Linux distros not using them and complaining about having to pay licensing fees to the Fraunhoffer Institute to have mp3 support since mp3 is copyrighted?
You can’t copyright MP3 encoding. You can, however, patent it, which is what the Fraunhofer Institute does. Distributing lame (in compiled form) without paying Fraunhofer a license fee for the *patent* is against the law in countries which enforce software patents (i.e., the U.S.A.).
Am I missing something or is there not a link to “in_mpg123.dll” on the Japanese page the article links to? I’ve downloaded “in_mpg123_118ot82.7z” from there, but Winamp doesn’t seem to have any idea what that is…
http://www.7-zip.org/
Ah, thank ye kindly!
Shucks. I wish you had told me sooner. I record and mix audio on a PC. Outboard I have all quality Pro equipment, as well as inside the PC. M-Audio makes some great stuff and the SB Audigy has a good quality to it. For some it’s the noise floor and latency issues that drive them nuts… and stay away from anything compressed. Use WAV files and fast drives.
Sweetwater has been in the music business for years and they are now building quality-tuned PC’s for high end audio production. You CAN get quality on a PC. They have done it, so have I.
A great pair of headphones? AKG K171 studio closed circumaural. Cuts out the background noise and sound great. Get yourself a good monitor amp and pair of nearfield monitor speakers. Or a pair of powered monitors. Event or Alesis monitors sound great and won’t break the bank – much.
Mplayer has alot of these features built in such as gain, resampling, and volume normalizing. You can find them under the -af flag in it’s user manual, eg man mplayer. There are some real interesting stuff there, no one has beeen able to exploit it by building a *good* gui over mplayer. Kinda sad.
Here’s my ~/.mplayer/config if anyone would like to make their stuff sound/look better:
# Use SDL for video output, sadly no HW accel though.
vo=sdl
# Audio Filter. Gain, Surround, and normalization.
af=volume=8,surround=15,volnorm
# Video Filters.
vf=pp=hb/vb/dr/al/lb
# Four channel out put. Values accepted are from 1 – 6.
channels=4
# Video letter boxing.
aspect=16:9
# Framedrop on, not hard.
framedrop=on
I would also like to point out that good quality headphones beat any speaker system hands down.
The dynamic response of headphones will always be awful compared to a proper loudspeaker system, and this is coupled with the fact that headphones destroy any tangability to the music.
My suggestion has been and will continue to be a nice external USB or FireWire device, preferably one with a nice, isolated DC power supply. Here is an excellent little 8 channel FireWire mixer/multitrack recorder for $250 from Alesis:
http://www.samedaymusic.com/product–ALEMULTIMIX8FW
yes, because I can sure touch the music coming out of my speakers. *shakes head* “tangibility”…whatever.
In fact, it’s not easy to even find Creative Drivers for 64bit Windows right now… I wonder how long it will take to have them…
BTW what’s a C: and what are its dependencies?
heh.. the world would be better if everyone asked that questions
————————————————————–
The scary thing is that the world still asks that question. It would be better if it just didn’t exist.
“Don’t bother with the EMU/Creative cards (It’s the same company since Creative bought EMU).
You can pick up an M-audio audiophile 24/96 new for around £60 nowadays, and it will blow most consumer cards out of the water. It has no headphone outs, no preamped mic in, no 5.1 surround, no analog input level vca, it’s crap for gaming, but…”
I’ve directly compared the M-Audio 24/96 and the Emu-0404, to me the Emu is easily superior. Reviews and coments on boards like Head-fi and Hydrogen Audio seem to agree with me.
“There are definitively. The statistically most significant one I know is done my the German computer magazine c’t performed few years ago, which included several thousand (6000?) double-blind samples from its readers. In that test, Ogg Vorbis is among the best lossy compression available, while MPEG Audio Layer III the worst, and that includes even the MP3Pro for the streaming typical bitrates.”
Do you have a link to this or any other test? What MP3 encoder did they use when comparing MP3 to other formats? There are big differences between different MP3 encoders, recent versions of LAME are far superior to the encoders most people were using a few years ago.
http://www.heise.de/ct/02/18/020/
That’s the announcement of the listening test (in German), but the results are only in the magazine.
http://www.heise.de/ct/02/19/094/
Seems they also made a blind test with audio professionals.
That test is several years old, all it shows is that the particular MP3 encoder they used back then was inferior to the alternatives.
Things have moved on a lot since then, this test says nothing about the relative quality of the modern LAME encoder and VBR MP3s.
That is about the only sensible discussion I have seen in this tread… Your PC is never going to give you audiophile quality sound – at least if you use its analogue outputs.
Why? Well, it has a whole bunch of cheap, crappy DACs in it to start with. It also has a whole bunch of noise generating bits and pieces. The cheap crappy DACs and copper tracks pick up all that noise and send it on out the analog plug. These motherboards with valve amplifiers on them are nothing more than wank for idiots who don’t have a clue. Do not be fooled.
What you need to do is get a soundcard that has a digital (optical) output and feed that through a long piece of fibre (anyone who tries to tell you that you need “monster fibre” is also full of wank) to an external DAC connected to a high quality power supply and a high quality amplifier/speakers/headphones.
And dispense with the MP3 crap. It is old technology that was useful when it was released because it was the only technology that could get reasonable quality audio down at telephone (64k) bitrates. There are better lossy coders now – look into OGG Vorbis. If you want true perfect quality, look into FLAC and others.
I never figured out why people are obsessed with MP3 vs Ogg Vorbis vs WMA vs AAC vs whatever else.
At the same bitrate, the difference between MP3 vs anything else is MUCH less than the difference between your current crappy sound card and (for example) an external box with much less interference. It’s also much less than the difference between your current crappy headphones and (for example) anything made by Sennheiser that costs over $100.
It’s what the article said. Fix the weakest link in the chain first. Once you have a great sound card and great output, then you can worry about whether 192Kbps Ogg Vorbis sounds better than 256Kbps LAME MP3.
Quote: “Want audiophile quality from your PC? Ain’t gonna happen. Burn a CD, haul it over to some quality equipment.
To the person that says decent headphones beat any speaker, you’re wrong. You’re not listening to the right speakers (amp, pre-amp, cd, cables, etc).”
Your first comment is absolutely correct. mp3s are compressed. So are oggs. So are CDs for that matter. LPs are as well, but are much more natural sounding (and lesser compressed to boot, no surprise there). You really need decent equipment to get the very best out of music, and a good room as well. The room is very, very important. If the room is poor, the sound will always be a compromise.
Some headphones, in some ways, can be better than speakers. There are some things that speakers do better, and some that headphones do better.
Resolution – unless you own megabuck speakers, headphones will provide better resolution
Bass – tough one. Depending on the quality of the headphones or speakers it can go either way from my experience. My Sennheiser HD40 Gold references have a nice and taught bass. My Sonus Faber Elector Amator speakers have a nice bass, but a tiny bit tubby. Still relatively good bass depth for such small speakers. I’d rate my headphones slightly sharper, or tighter in the bass, but the speakers have slightly more depth, and in impact I find. Of course, if I add my REL Stadium II subwoofer into the equation we get true bass 😉 Neither headphones or main speakers come near it in terms of bass reproduction. This baby goes down to around -3db @ 12hz or so. Providing the room is big enough to accommodate bass that low that is.
Depth – Generally speaking I find that speakers will have much better depth than any headphone set, even a very expensive and good headphone set. Small speakers, properly setup, mounted on solid speaker stands, with the correct distance between them, and the listener and also the right amount of toe-in (or no toe-in for some speakers), as well as room size, shape and also wall/floor covering will all impact on how the illusion of depth is received by the Human ear and brain.
Soundstage – Both speakers and headphones can be winners here, with headphones it comes down to quality of the transducers, for speakers it’s partly the transducers, the electronics, the recording, the room etc. It does seem to be easier to pick out soundstage with headphones than speakers (at least on a reliable basis). The better speakers deal with soundstage better, but you have to pay money 😉 The rules that apply for Depth and Height also apply here, although from experience not to the same extent (ie. this part of aural perception seems more tolerant than the other areas).
Height – Similar to depth, although headphones from my experience fare better than the depth perception. From experience though, speakers win here as well. The same rules for speaker placement etc apply here as for Depth.
Image placement – This one…is well…Difficult. Usually i’d say that headphones have much better image placement than speakers, and i’d be correct in most instances. However, the very best speakers can and do match the best that headphones can offer. The rules that apply for depth/height also apply here.
Do cables, amps, source make a difference? Sure they do. Is it a subtle difference? Yes, in most instances. And in most instances the average person isn’t going to really care, or be this pedantic. In reality, the recording/performance is the most important thing – if it isn’t on the recording, you’ll *never* hear it no matter how good the equipment.
Dave
PS For those audiophiles out there, my dream system would be:
A pair of Mark Levinson No. 31 mono amps
Pink Triangle Pip II pre amp
Michell ISO phone headamp
Pink Triangle Anniversary turntable
SME IV tonearm
Lyra Clavis MC cartridge (I already have one of these hehehe)
For a DAC, the good old DPA pdm-2 series 1 (I think it was 1, or maybe it was series 3, hell it’s been that long I can’t remember lol) DAC can’t be beaten imho, not even by the very latest and greatest in optical technology.
A pair of Sonus Faber Amati Homage speakers (sexy!!!)
I hereby declare this thread moved to http://www.audiowankers.org , thank you and goodnight.
Yes Sir, mr. thread policeperson. Sig Heil mein kommandant!
Dave
This one was indeed very appealing to me as I’ve been planning on building my audiophile equipment quite soon. I started out by making sure I have loads of music in FLAC format (currently 150gb) so there is at least some motivation for it, as in I want to note the difference.
It seems like the wisest move to do is to go for a smart DAC solution. I just wonder if there are any benchmarks for them to see at what level is reasonable to settle for..
Anyone have any ideas?
Disclaimer: I am not an audiophile, and computer uses its (surprisingly good) onboard audio through middle-of-the-line 2.1 speakers by Logitech – not audiophile quality but quite decent. I mostly use Winamp 5 for Shoutcast radio at 128kbps, which I have playing most of the time. I never lower the volume with Winamp’s control, prefering the manual control on the speakers.
Given that, I tried the applicable suggestions in this article. The mpg123 plugin seemed to have some positive effect, but not drastically so. The kernel streaming plugin had a noticable, significantly positive effect, especially on the higher end, adding definition and detail. If there are stability issues I haven’t experienced them yet.
Pending some trials with other formats, I would have to give this advice a plus. For my purposes this has helped a lot.
I’ve seen this claim a few times, but I haven’t seen it backed up with evidence. Are there any listening tests that actually prove that there’s a big difference in quality between MP3 and other formats?
Test for yourself: Rip a CD to 192 kbit CBR MP3, then listen to it on a pair of quality boxes (MFBs, for example) or quality headphones (e.g. Sennheiser). Now do the same with the original CD. Perform a blind test if you want. Choice of CD may matter ofcourse, but that is a duh.
You may also want to Google for tests (German mag C’T performed some, as did r3mix.net). Or go to some ‘underground’ party where the DJs are beginners who can’t afford vinyl (or perform live act!). Instead they leech MP3s from the Internet and use those. Even though the audio hardware is of good quality, it’ll be noticable, and you’ll feel yuck.
Worked for me too once direct kernel streaming was enabled.
For a person that just likes to listen to music to fill the emptiness around him, the new sound gain that I hear sounds great. And that’s all I care about.
Fuck Audiophile l33t sp34k, just give me the fucking music.