Conversions to MP3

  • Thread starter Thread starter miroslav
  • Start date Start date
I render it in Reaper and leave it at 24bit and 88200. I create a 320 mp3. that's just leaving everything at it's highest quality, right? so isn't that best?

but that's at the mastering phase

Eh, I saw a comparison done years ago where they tested mp3 codecs at various sampling rates - I think it was a simple pure tone test, so I'm not sure how it'd stack up in the real world, but Lame at 256kbps actually outperformed Lame at 320kbps - there was a little more distortion present.

I haven't thought seriously about this in a long time, but the last time I tried testing myself (ripping a fairly sparse track from CD to .wav, then encoding the .wav in varying rates and opening them both in winamp, putting it on random, and hitting "track advance" a couple times with my eyes closed to pull up a random version that would either be a .wav file or the mp3 frequency I was testing) I could distinguish between 128kbps and .wav with almost perfect accuracy, and between 192 and .wav with a statistically significant frequency (around 2 out of 3). 256kbps was the point where there ceased to be any statistically significant difference to my ears (if I started listening and thought, "ok, that's a mp3, I was only right about half the time), so that's what I encode at today whenever possible.

I use Goldwave, with the lame.dll codec. Goldwave is just a prettier interface. :D
 
...anything above 192k tends to be a waste of bandwidth, IMHO.

So you say really don't hear much difference from same-source 192 VS 320Kbps?

I've done some 160 and 256Kps, but never split the difference to try a 192.

The 160 don't sound *bad*, but I can hear the sound degrade...that phasey thing that happens to the HF. So that's what I like to use for my Lo-Q MP3 files, with 320 for the Hi-Q.
The 256Kbps was also pretty good...but I was being "mathematical" in my choices... :D so I figured I would just double the Lo-Q 160 rate and that’s how I chose 320 as my Hi-Q....though I think it does offer the a noticeable sound improvement.


That's the limit for free accounts. For paid accounts you can upload at 320.

I can't recall...how much do they charge for a paid account?

I think I ended up just going with my own website, which lets me do a lot more, plus the audio files at any rate I want...rather than paying SoundClick just for a basic file-share service....though I guess SoundClick has become somewhat of a larger musician's community for people that like the browse for music.
 
I can't recall...how much do they charge for a paid account?

I think I ended up just going with my own website, which lets me do a lot more, plus the audio files at any rate I want...rather than paying SoundClick just for a basic file-share service....though I guess SoundClick has become somewhat of a larger musician's community for people that like the browse for music.
Soundclick offers several of levels of paid service. For about $10 every 3 months, you can upload bigger files and I think they rotate your music through the category playlists in some fashion. You can also pay extra to have a band banner or be a "featured band."
 
I'm the Legionserial school of MP3 creation.

My primary objective is to mix for CD. If I am creating an album's worth of material, the tracks get mixed once, then go through a post-production process to get the set sounding cohesive as a whole. From their I convert individual tracks to MP3s as and when required.

I am disinclined to creating a set of MP3s at mixdown . . . because that means then subjecting these to the same post-production process as the WAVs . . . that, to me, represents double-handling which is decidently unappealling.

I tend not to use MP3s as a primary distribution means, and use them more for providing examples and demos.
 
So you say really don't hear much difference from same-source 192 VS 320Kbps?
What I'm saying is that - for me - Fraunhoffer at 192 is good enough for MP3. I can also taste the difference between a Quarter Pounder and an Angus Burger at McDonalds, and I can taste the difference between Budweiser and Miller. Who cares? Nobody is going to McDonalds or the 7-11 to buy that stuff expecting a gourmand's experience.

I don't do MP3 at all unless I have to use it to e-mail or up/download a sample to someone, or unless I buy a song from iTunes, and in eather case, 192 is just fine for me. In the studio or on audiophile playback, however, I stay away from MP3 altogether.

If you want fidelity, go lossless and forget the MP3 altogether. If you want MP3 to save the bandwidth, then save the bandwidth. When I'm listening to MP3, what difference there may be between 192 and 320 is no more relevant than the difference between a QP and an Angus is to someone who's scarfing it down behind the steering wheel while texting and trying not to crash into the car in front of them.

G.
 
Last edited:
I don't do MP3 at all unless I have to use it to e-mail or up/download a sample to someone, or unless I buy a song from iTunes, and in eather case, 192 is just fine for me. In the studio or on audiophile playback, however, I stay away from MP3 altogether.


No...I'm not trying to find “fidelity” in MP3 files. :D

But like you pointed out, at some point, the studio work we do will often have to end up as an MP3 if we want to offer the same downloading convenience that others do.
So...my questions stem from that, since I would rather create my own MP3s using the encoder and quality level that works for me, rather than leaving that up to some automated process on some website...or some else's ears.

That's why I'm looking for the "best" possible MP3 creation process that people have decided on.
I was curious how others were dealing with it.
 
No...I'm not trying to find “fidelity” in MP3 files. :D
...
That's why I'm looking for the "best" possible MP3 creation process that people have decided on.
I was curious how others were dealing with it.
And that's what I was relating to you. I'm not arguing, I'm telling you how I deal with the whole MP3 issue, and what I feel the "best" creation process is (Fraunhoffer 192k) and why.

One thing that many of us tech-heads with enough coinage to rub together to be able to afford modest recording gear forget that there is still a significant number of Internet users out there that do NOT have broadband access. I regularly check the usage statistics on my website, which one could argue attracts even more technically-bent users than the average, because it does indeed deal with audio recording. I average between 7k and 10k visits a month (not huge, I know, but a large enough database for significant statistics), and a good 1 out of 3 of my visitors are running at speeds of iDSL (not DSL, iDSL) or less.

That's a huge amount of folks out there for whom the audio difference between 192k and 320k on a pop or rock tune isn't going to be worth the cost in bandwidth/download time, especially since most of them, broadband or not, will never hear it anyway in the situations in which MP3 is intended.

And on the other end of the spectrum, those with cable, T1 or better have little-to-no issue grabbing multi-megabyte WAV files which are lossless and full-fidelity. So setting up 320k MP3s for them just for the incremental fidelity is silly; I'll just send them the real thing.

MP3 is for convenience and convenience only in my mind. If the receiver needs/wants full-fidelity, I'll send then a stereo WAV. If they have only dial up or iDSL, I'll either send them a CDA or WAV data disc, or if they just want a sample, send them a decent-but-not-best fidelity sample FrauMP3 at 192, and they'll be perfectly happy with it...at least so far that's been 100% true.

This is not to say that anyone else is wrong when they look at it different. I'm just answering your request :).

This whole MP3 thing is right up there with the sample rate thing IMHO, there's waaaaay too much energy,time and bandwidth spent worrying about, discussing and "optimizing" such minor B-list issues. Just make a freakin' MP3 and get on with it. Nobody really cares about the rest if the song is worth listening to; and if it's not, then it REALLY doesn't matter.

IMHO YMMV LSD MDMA etc.

G.
 
VBR = Very Bad Result - seriously!
I use 320 if I want someone to listen at the end of an email or when sending UP to a streamer. I didn't originally but then I found a LAME prog that had 320 & UP graded (Easy CD Extractor).
I bought myself a nice little player that does all the MP3s, Realvideo (that's a bummer - but what do I really want quality wise in a video player smaller than my palm) but most importantly plays wave files & does it well. So I have the portability but reasonably good quality - I don't use ear buds though. I use it to listen to mixes etc when wandering down to the shops or in the garden etc.
I mix to wav & then rip from that as needed.
I'm no fan of MP3s but they have their uses - & as mentioned, if the limitations are understood & the use warrants the limitation I use them.
 
I'm not arguing, I'm telling you how I deal with the whole MP3 issue...

:)

No...no arguing…shit always gets lost in the translation on forums! :D


My main questions pertained to any differences in encoding MP3 files from the higher resolution masters (assuming the mastering was done at the 24bit/32bit stage)...VS....encoding MP3 files from the finished 16bit WAV files.

AFA which delivery rate to use for Internet...IMO you got to get past 160Kbps to NOT hear the *more obvious* sound degradation. Whether that “better” MP3 rate is 192, 256 or 320....mmmmm, I haven't decided on that just yet, but I do get your point about 320Kbps being a fairly heavy load on file transfers with so-so connections, and that 320 MP3 file size can be pretty large too, so there may not be much to gain VS just letting people download the full WAV....though it's still over a 4-to-1 ratio (WAV to 320 MP3) in file size.

And then here’s the other thing…I(we) may not want to allow full WAV downloads unless people are paying proportionately for them, whereas the MP3 downloads are more of a “bargain” version, no matter what rate…IMO.
So…my thinking is that the MP3 has to be of good enough quality to entice people to buy the whole enchilada…but still lacking enough in quality for more discriminating ears to notice.
Maybe 192Kbps is the best “break-point” for that….?

But it may all be academic, as it seems many young ears are being trained that 128kBps is “good enough”…so I’m not sure how you entice those people to go for the full enchilada…???
 
AFA which delivery rate to use for Internet...IMO you got to get past 160Kbps to NOT hear the *more obvious* sound degradation.
Which is why I go to 192k.

I think another thing that's kind of being danced around here but not quite directly stated is that there's a difference in philosophy by what's meant by "for Internet".

I am neither a self-recording act nor an internet "record label". I do not do that swamp called meSpace except as another place to hang out a token link shingle for a real website. My use of MP3 is limited to playback of tracks I have bought from iTunes or as a "sample" format for sharing facsimile demos with folks. In both of those cases, I'm fine with 192k, and so are the folks I send them to, because it's more than good enough for those purposes (the difference between a 192k and a 320k MP3 when played in a noisy car going down the road or on mePod earbuds while jogging, or when wanting to hear if the vocla-to-accompaniment balance in a mix is what the client wants, is a distinction without a difference, because that incremental fidelity is irrelevant in those circumstances.)

And yes, WAV may be 4 times or so larger than 320k, but that's minor when compared to the 10:1 or greater difference in download speed from V.90 or iDSL to true broadband.

What it comes down to is that I think it's a mistake for anyone to think of MP3 as anything more than a convenience format at the least, or a temporary kludge format whose only reasons for existence are because

a) it allows or acceptable download speeds (at 128 or 192) for non-broadband users (who, remember, were the vast majority when MP3 was adopted), and

2) because it's lossy compression scheme was attractive to those who were worried about the lack of solid digital copyright protection (whether one agrees with the whole copyright thing is not the point; it's that this was a big reason for MP3's acceptance by the MPEG committee.)

In neither of these reasons is there even a hint that MP3 was ever meant as a final master distribution mechanism or as a bastion of fidelity, and to try and fit the square peg of lossy compression into the round hole of master distribution format is folly. It is NOT a serious format.

G.
 
My use of MP3 is limited to playback of tracks I have bought from iTunes....


.......



What it comes down to is that I think it's a mistake for anyone to think of MP3 as anything more than a convenience format at the least, or a temporary kludge format whose only reasons for existence are because

a) it allows or acceptable download speeds (at 128 or 192) for non-broadband users (who, remember, were the vast majority when MP3 was adopted), and

2) because it's lossy compression scheme was attractive to those who were worried about the lack of solid digital copyright protection (whether one agrees with the whole copyright thing is not the point; it's that this was a big reason for MP3's acceptance by the MPEG committee.)

In neither of these reasons is there even a hint that MP3 was ever meant as a final master distribution mechanism or as a bastion of fidelity, and to try and fit the square peg of lossy compression into the round hole of master distribution format is folly. It is NOT a serious format.

I agree with you that it should NOT be considered a serious format...BUT...like you mentioned iTunes...there are a lot of folks that DO get all their music in that format, and mainly/only that format.
So....even if it's not for your own self-recorded music and it's for some artist you are working with...I think you/they/we are almost *forced* to consider that the MP3 files WILL be the "master" distribution format for many. :eek:
I know most people will still churn out a CD format too...but it appears a lot of "convenience" listeners just go for the iTunes/downloadable versions, and they don't ever buy the CD, so in many ways, the ugly MP3 is the only version they ever get to hear.

I’m just trying to walk the line…though I would rather not deal with MP3 files at all.
 
I agree with you that it should NOT be considered a serious format...BUT...like you mentioned iTunes...there are a lot of folks that DO get all their music in that format, and mainly/only that format.
So....even if it's not for your own self-recorded music and it's for some artist you are working with...I think you/they/we are almost *forced* to consider that the MP3 files WILL be the "master" distribution format for many.
That's circular logic.

It's much ado about nothing. There's no line to walk; or if there is, if you fall off, nobody is going to get hurt. Just hold your nose, cut an MP3, and get on with it, IMHO. Nobody is going to miss the stuff that 192k may potentially leave out.

G.
 
Just hold your nose, cut an MP3, and get on with it, IMHO. Nobody is going to miss the stuff that 192k may potentially leave out.

:D

Yeah...maybe that's the best way to deal with them.

I guess I just wonder why we bother if nobody is missing the stuff beyond 192Kbps….?

I'm really curious (not sure if there are even any hard numbers)...how many people ARE buying that full CDs after sampling some 128/160/192/320Kbps MP3 files...?
How many are saying..."Sounds pretty good, I think I'll get the CD so I can hear how this music sound as the Hi-Q version."...???
While l I do know that lots of folks still have crappy Internet...this kinda’ gets back to that recent video with George Massenburg at that Australian audio conference, where he's pretty much asking the question why we are not moving on past MP3 files and going with full CD quality WAV download systems...instead of the iTune MP3 crap....and that the technology is here now to do that (we just have to improve the Internet speed in some areas).

IMO...this is all about milking the MP3 format for all it's worth...and then watch, maybe 5 years from now they’ll start offering up pure CD-quality (or even better, using 24bit instead of 16)...and then they can milk everyone for more money. :)
 
I think the point here is that the reason you make an mp3 is to cut down on storage space and bandwidth.

I personally would find it pretty annoying if I was forced to download 320kpbs mp3s without the option of 160 or 192. When I download an mp3, I'm either streaming it, which means I'm listening to it on headphones at work, or I'm saving it to my hard drive or iPod, which means I don't want it taking up unnecessary space.

Anything above 192, I generally just toss, because if all my songs were encoded at 320, my iPod would have been full long ago. If I'm streaming a song and it stutters because the connection can't keep up with it, I move on to a new song.

As Glen said, it's best to find the best sounding encoder at the lowest bitrate possible. I think choosing the highest bitrate because it's higher makes the whole mp3 thing kind of pointless.
 
I make them off of the 44.1 masters.

Mp3's sound like shit and there's nothing you can do about it. To me all digital audio sounds horrible so far. Technically I would describe the sound as "fucked".

I don't have an iPod. This whole thing of "how many songs do you have on yours" and making them less than 320 so you can get 10,000 songs on one is plain stupid.

We're in a really stupid phase right now and it has more to do with being in the "pre-paradygm-shift" era than anything musical. There's a massive paradygm shift coming up the likes of which we've never seen. It's going to be a really different world in 5 years. I mean REALLY DIFFERENT, probably as different as from 1985 til now, or more, so just get through this period and all this mp3 crap will seem like 8-track tapes very soon.
 
you're right, it will be different. Bitrates will just get lower and lower.
 
I personally would find it pretty annoying if I was forced to download 320kpbs mp3s without the option of 160 or 192. When I download an mp3, I'm either streaming it, which means I'm listening to it on headphones at work, or I'm saving it to my hard drive or iPod, which means I don't want it taking up unnecessary space.

I think choosing the highest bitrate because it's higher makes the whole mp3 thing kind of pointless.

So then...you're saying that you don't care much for anything of higher quality...it's all about the convenience of lading a gazillion *crushed* MP3 files on you iPod...THAT is basically what is important to you?

I'm just trying to understand the current focus on audio quality.
Like Glen said...he avoids MP3 files whenever possible, and prefers to let the people have the Hi-Q WAV files.

But you seem to be on the other end of the spectrum, where quantity and convenience is more important than quality of audio.
 
I think it's important to understand the purpose of the mp3 codec. It is designed to have a small footprint while sounding as close as possible to the original content. The goal is to get it as small as possible while minimizing the negative audio artifacts. If you don't care about footprint, then there is no need to compress it.

I also think it's important to understand who is going to be listening to it. If you want a copy for your own gratification to listen to and admire the recording quality, you won't be doing that with the mp3 version.

If you want to distribute it for other people to hear, then you need to cater to their needs, not yours, and in this case, you will have to sacrifice what you consider to be quality for what the end user considers to be convenience. Just because you can hear the difference, doesn't mean they can. If the end user can't hear the difference between 160 and 320, then the file is twice as large as it needs to be, and you are doing them a disservice by encoding at a higher bitrate.

I think it's also important to think of the people you are recording for. If they say they want their stuff encoded at 128, then by all means, encode at 128. I think it's a good idea to educate them on the advantages/disadvantages of different bitrates, but if your advice is "bigger is better" then clearly you're missing the picture.

I think the 320kpbs mp3 is catering to a nonexistent audience. That's basically saying "I want it as big as possible as long as it's not as big as the original."
 
To steer away from the "who can hear what" subject and get back on the best way to get the best quality from your mp3 encoder, I'd say the first step is to find which encoder sounds the best. All encoders are not made equal. One may sound better at 112 than another sounds at 320.

Also, variable bitrate is inherently better than constant bitrate. I'm curious to hear some of the issues that people have had with vbr.
 
I've never heard anything in the 160 range or below that sounded better than 320...no matter which encoder you use. And I never heard ANYONE specifically demanding lower bit rates. That's just something that was forced on people as a limitation of the inexpensive technologies being chosed/used…but like George Massenburg was pointing out, it's already possible to go with higher quality...the technology is hear now, manufacturers just don’t want to make the move yet…(they’re not done milking everyone with the current technology). :D
So again...are listeners really "enjoying" 128...or...if the technology stepped up (which it soon may) would they be enjoying higher quality even more?
YES...BIGGER IS BETTER :) when talking about compressed audio...and no I'm not missing the picture, I think everyone that has *accepted* crappy formats just because they want a million files on their iPod instead of a thousand (or some such thing)...those are the folks missing the picture, IMO.
They opted for convenience and quantity over quality.

Use to be, people would carry around their favorite 20-30tapes or CDs...and life was good. I don't recall anyone really complaining about how few choices they had or that they wanted more choices at the sacrifice of quality, not to mention that radio was use to be much better and there were some great stations at one time.

So my question again to everyone...is the convenience and quantity more important than great sound quality?

See...back in the day, when a "stereo system" was something to almost worship, everyone was into the quality...now days, not so much.
That said...if I have to toss out MP3 files of my music, I will choose the rate that I think does the most justice to the music. Maybe 320 is not needed...maybe like Glen said, 192 will cut it, but no one can tell me they can't hear how crappy the lower rates sound. If you can't hear the sound degradation when you get down around 160 and lower….it just means you've gotten use to it.
It's like listening to music in the car while driving. The road noise creates serious masking, no matter how good your car or sound system...but after a mile or so, you don't notice it as much...and pretty soon your ears are only focusing on the main sonic elements, which will be enough to get the music *across*, but certainly not anywhere near what the music really sounds like when listened to in the right environment. All those sonic nuances, those tonal variations, the subtleties, the ear candy…are lost or altered by the encoding/compression process.

My underlying point here is that somehow people who have accepted the quantity/convenience over quality, need to be drawn back over the quality side…to once again appreciate good audio…like it use to be. Otherwise...we will all get use to listening to ambient "noise.
Maybe they’ll push out those “new-n-improved” technologies soon….

I never use IPods to listen to music…mostly ‘cuz I think they sound like shit, but I do acknowledge that a lot of folks use them, and that’s why I started this thread…to get a sense of how people were prepping their MP3 files and how much concern they had over MP3 sound quality.
 
Back
Top