I know this topic has been beaten to death with a bloody hammer repeatedly (wow, that's pretty graphic), but the fact remains I still know very little about some of the results of encoding.

I've read that Audiocatalyst chops off some of the higher frequencies. Does the encoding process chop off lower frequencies as well? I only ask because when I get my car's 10" sub hooked up I don't want to listen to muddled bass. Keep in mind that I encode at 128 because when I compared recordings to VBR I either couldn't hear or didn't mind the difference, and I'm using as good computer speakers as you can buy.


Also, just out of curiosity, what do you people listen to to test out your systems? I particularily like to crank some good 'ol Soul Coughing (Ruby Vroom). I don't know who recorded that but it's one of the best recorded albums I've ever heard. Crystal clarity with great mids and highs, and the bass (like on "Sugar Free Jazz") is absolutely nuts!!

DiGNAN

I've read that Audiocatalyst chops off some of the higher frequencies.

Not just Audiocatalyst, but pretty much every MP3 encoder out there. There is one exception. LAME claims to preserve the higher frequencies more accurately at high bit rates, and I'm evaluating LAME right now for use with my personal collection.

Does the encoding process chop off lower frequencies as well?

No. In theory, the lower a sound's frequency, the easier it is to data-compress accurately. I get great bass out of my Empeg, even on low-bitrate files.

Note that the Empeg's audio output circuits are EQ'd very "flat" by default, and the bass is not artificially boosted like it is on many car CD players. So the bass on the Empeg will start off sounding less pronounced than it is on the CD player it's replacing. Fortunately, the Empeg has an amazing equalizer that will let you dial in the exact settings you want for your car.

Keep in mind that I encode at 128 because when I compared recordings to VBR I either couldn't hear or didn't mind the difference, and I'm using as good computer speakers as you can buy.

I was content with 128 for a while, but after getting the Empeg, I realized that 128 wasn't enough. Getting the Empeg allowed me to listen in my car, which meant several things: a) The listening environment was more intimate, b) I tended to listen much much louder than at home, c) the amplification and speakers were better than at home. This allowed me to hear artifacts in the 128-encoded files that I hadn't noticed on the computer. These artifacts are usually in the high frequencies, and can be heard mostly on cymbals. Hi-hat and crash cymbals sometimes get swishy or crackly sounding at 128kbps. I'm now sold on VBR encoding at higher-than-128-equivalent quality settings to get rid of these artifacts. My only real choice now is between AudioCatalyst and LAME as my encoder.

Also, just out of curiosity, what do you people listen to to test out your systems?

I have an entire playlist of 40 songs on the Empeg devoted to nothing but test audio. They cover a range of frequencies, and I can listen for different characteristics in each one. Here are some highlights:

The Cape- Trevor Rabin
Subsitute For Love- Madonna
Ray of Light- Madonna
Vision of a Kiss- The B-52's
Follow Your Bliss- The B-52's
Red Rain- Peter Gabriel
Riot Proof- Tori Amos
Trip Like I Do- The Crystal Method
It Don't Hurt- Sheryl Crow
Papa Was a Rollin' Stone- Was (Not Was)
Modoc- Steve Morse
YYZ- Rush
La Villa Strangiato- Rush
Cut to the Chase- Rush
The Soul Cages- Sting
Copperline- James Taylor
Please- Nine Inch Nails

I'm sure Doug Burnside will chime in here with his favorite test audio tracks.

___________
Tony Fabris

OK, Tony, you pressured me into it...

I would say anything from the re-mastered Rush CD "Moving Pictures". (Got to be the re-master, though!)
There's a CD by Cyrstal Method called "Vegas" -- all electronic music, some pretty amazing bass on it, good production, great clarity.
Finally, almost any CD produced by Sheffield Labs.

Since I am not a rock and roll fan, I am clueless about Tony's recommendations except for two of the Rush songs.... ;-)

Also, I agree with Tony -- you are not going to be satisfied with 128 kbps for long once you get your empeg. The difference in file size between fixed 128 and the Normal-High VBR setting in AudioCatalyst isn't that great, and the sound quality is better. Also, hard drive space is cheap even on an empeg. You may have a really good sound card and speaker set with your PC, but I'd bet that it is really pretty sorry when compared to even a moderately good in-car stereo system. (Please -- no flame war here! I'm just expressing my probably ill-informed opinion, based only on my exposure to a lot of not-very-good PC sound systems! Maybe there is such a thing as a good-sounding system running from a PC sound card. I just haven't heard it.)

tanstaafl.

"There Ain't No Such Thing As A Free Lunch"

You may have a really good sound card and speaker set with your PC, but I'd bet that it is really pretty sorry when compared to even a moderately good in-car stereo system

Actually, it isn't. My current car stereo is PITIFUL. It includes two really crappy speakers and an 80 watt head unit that overheats the CD's. I've got a Sound Blaster Live!, and more importantly, 400 watt Klipsch ProMedia's. What's more is that my car doesn't even have a subwoofer as opposed to the 160 watt in my Pro's.

Now when I get the new system installed, it will only be as good as my PC speakers. The amp won't even be as powerful, and I'll still have the same speakers. Trust me, I won't be dissatisfied with 128

DiGNAN

Besides, honestly I can hear the artifacts in 128kb even through cheaper speakers...

Me, too, but only because I'm practiced in what to listen for. It's easier to hear the artifacts on a really good system, but it still takes time and practice to pick them out.

The problem is that, although I was once satisfied with 128-fixed, the high quality of the Empeg playing through a good car system has trained my ears to hear the artifacts. Now that I know what to listen for, an artifact-y cymbal crash sticks out like a sore thumb and it really bugs me.

It's like when someone points out a glitch in an otherwise good movie. You could go your whole life enjoying the movie on its own level, but as soon as your friend points out how you can see the wires holding up the spaceship in that one scene, it'll bug you every time you see it from then on.

Some people will never get to that point. I mean, look at all the memory-stick MP3 players they're selling. The owners of these devices have to be satisfied with low bitrate encoding because of the limited storage. So some folks are happy with 128, 112, or even 96. I'd say "ignorance is bliss", but it's not ignorance, it's just that they never listened closely enough to hear the artifacts, or never had it pointed out to them.

So don't get down on folks who are satisfied with lower bitrates. Envy them.

___________
Tony Fabris

Well, I've gotta say that I'm not that bad. I have a friend who claims that because he has horrible speakers, he just encodes everything at 56! I'm sorry, but you'd have to be playing music through a PC speaker not to know the difference between 56 and 128

DiGNAN

I use AudioGrabber 1.62.2 with the LameEnc DLL engine 3.70 at the 192kbit/s bitrate (joint stereo). I went through the stuff www.r3mix.net 'articles' and came to the following conclusions:

I 'should' be using EAC (Exact Audio Copy) to rip the music pure (100%) from the CD. However, I got hooked with the AudioGrabber -application (easy to use, has CDDB and .mp3 encoding options ready, just copy the LAME .dll into the dir). Their difference is that EAC reads the CD's raw as AudioGrabber does DAC (Digital Audio Copy). I will not go into talking about this but people have pointed out that EAC goes through the error checksums etc, somehow utilizing the raw-read mode or something like that. So far I have not noticed any clicks or other artifacts that the www.r3mix.net guy says the Digital Audio Copy may produce, I trust it quite much depends on the speed you are using in reading, the quality of your CD-rom AND the particular CD. Most of my CDs have been used very little so I am in luck by having no visible 'bad' scratching in most. Am I wrong if I claim that 'Digital Audio Copying' (ripping) is like playing the CD on your own CD-player ie. listening to the record?

I have locked my HP 9310i series writer (which I use in the ripping process) to 8X reading speed in the AudioGrabber program. Higher speeds seem to "give possible read errors" on the program as my p3-600 can not keep up with the ripping pace - the disk needs to re-spin and the laser will have to resume reading at the very same spot it left off. Although I noticed no problems with higher speeds either, I went for the 8X option. Some people would claim even 4X is too fast..? (the audiograbber guy says that quality drives should handle this ok) CD-drives should be rated for 'read music' speeds but I didn't bother to go find more about this.

Sometimes the AudioGrabber or my CD-drive does not comply with the 8X read-speed I've issued to it and the 'possible problem' -count changes to a positive number from 0 but I have not heard any problems on the songs this has happened.

I am NOT using VBR currently as the www.r3mix.net guy claims that the new LAME encoder engine's (though mine is the 'stable' version) VBR produces some serious artifacts and still needs some tweaking. This made me a bit worried as I do NOT want to go through all my (roughly) 250 CDs more than this one time. 192kbit/s seems to produce clearer sounds as I listen to them at home with my Pioneer (can't remember the model, one of the top ones of 1998) amplifier and OR-300Hi (these guys are huge and have a really good sound) speakers. The difference with 128kbit/s and 192kbit/s can be heard.. also, I have never been in need of HD space.. lucky me I guess (other than that the Empeg I have in mind only has a 12GB drive.. hmmm, I think it will fit about 150 full albums encoded with 192kbit/s). I'm switching to VBR as soon as the more close-to-the-tech people (or atleast ones talking about it) flash green lights.. though I only have about 100 CDs more to go..

I'm also using the 'Encode with Highest Quality (reduced speed)', 'Use ID3 Tag' and the 'Normalize peak level to 98%' -options. This may somewhat slow the encoding process, I have not really experimented with these but they all sounded good. Ripping & Encoding one CD takes a while but well.. I have other comps to play with during that time, I'm not even touching the one I'm doing the encoding on while the CD is in the oven. I never do anything with the machines I'm 'burning' stuff on, I somehow equalize encoding with burning.. I reckon reading something about the peak level normalizing but I really can not recall what it was. The ID3 tags are used by the Empeg (I hear) so why not keep them there.

I guess I'll have to go for the 18GB model.. however, there is this question about the money.. I'm hoping I can find the money for the Empeg in time, I'm hoping the blue screened one will fit perfectly in a VW Golf V6 4motion that I should be getting from the factory very very soon now..(hopeful)

On the other hand, are you sure you want to keep -ALL- your encoded CDs on the Empeg-machine only? A moderately sized IDE hard-disk costs next to nothing these days.. I'm looking forward to putting the 'diamonds' of my collection on the Empeg and leaving some of the music I know I will not be listening (even on the Empeg) somewhere else.


Antti 'Amarth' Luostarinen
#14229 in the waiting queue
play BatMUD (telnet bat.org)

Oh, that's a nice car. I have a 2000 VW Golf GLS, not a V6 4Motion though, I can only imagine. Where are you located? We can't even get those yet in Canada. I hope the blue will match too, does anybody know?

Mike Cousins
#14979 on the waiting list

The blue screen is slightly darker in tone than the current VW "instrument blue" lighting colour. Howver, I found that if you turn down the illumination on the console of your car, the colour difference is not so notable.

Anyway, you want people to look at your empeg, not at your speedo...

One of the few remaining Mk1 owners... #00015

Actually, the production blue is lighter and more blue-green. It doesn't match the beetle's blue illumination, but is much more readable in sunlight. I'd have liked an exact match for hte beetle, but I prefer to be able to read the display :)

Hugo

Actually, the production blue is lighter and more blue-green. It doesn't match the beetle's blue illumination, but is much more readable in sunlight.

Mark 3: RGB Display, with preset "color schemes" for different makes/models of car.

It's the only real solution. You now have no choice but to implement it.

___________
Tony Fabris

Amarth said:
>I use AudioGrabber 1.62.2 with the LameEnc DLL engine 3.70 at
I also use AG 1.62 but I use the Fraunhofer (sp?) encoder. Where does one get LAMEEnc?

I have one question about AG. For the type of copy or buffering, what do you choose? I can not recall what I have but I think it is the default of buffered burst or something.

>#14229 in the waiting queue
You are ahead of me. Have you been contacted yet? I was told possibly late August, early September. In time for my b-day. EMPEG give discounts for birthday gifts?!?!? :)

Thanks!

I am a number, not a free man! And that number is:

#15562

I have one question about AG. For the type of copy or buffering, what do you choose? I can not recall what I have but I think it is the default of buffered burst or something.

CD ripping is still a bit of a black art. Different CD drives on different systems will need different settings.

In my personal experience, I can leave Audiocatalyst (which is the same as Audiograbber) at its default settings, but only if I leave the computer alone while it's ripping. If I try to multitask other software during the process, I get skips and pops in my rips.

Asking someone else what their copy/buffer settings are is kind of useless, since each system has its own best settings. The important thing is to listen carefully to your rips and see if there are any of the following:

- Skips/pops
- bursts of noise
- sudden shifting of the stereo image between the left and right channels that's not present on the original recording.

All of those can happen with ripping errors. If you get those, look into altering the settings from their defaults, otherwise don't touch them. Look in AudioGrabber's online help under "Troubleshooting" for more information. For a technical discussion of why these problems occur, check out the Jitter section of the CD-Recordable FAQ.

___________
Tony Fabris

Following up on the ripping issue (i.e., extracting the digital bits from the CD), I'm going to have to go back and start over because I've been noticing ripping artifacts from my setup. I'm using Linux, grip (with built-in cdparanoia), and lame. The problem, after hours of experimentation, turns out to be my CD-ROM drive, a no-name "40X" drive. No matter how you tweak the cdparanoia options, it's really impossible to get a truly perfect rip. From one particularly scratched CD I have, the rips were comically bad. (It was techno music, so for a while I thought maybe it was just a strange remix...)

My current plan (once my EMPEG shows up) is to go out and buy...

- a Plextor PlexWriter (available SCSI or ATAPI -- the new ATAPI one is $305 at CDW)
- an IBM 75GB drive (enough space to eventually rip all 1000 of my CDs, $620 at CDW)

Some people I've talked to seem to think that SCSI is better than ATAPI, but I have yet to hear anything definitive. However, there seems to be a consensus that the Plextor drives are the ones to get. You can safely turn off all the "paranoia" features and count on getting the bits right every time (or so I'm told).

I've also played with various setting of Lame for encoding the same tracks over and over. 128kb/sec is an absolute minimum. Below that, all kinds of things start sounding funky if you listen closely enough. Above that, it's harder to say. With my weak sound card, and expensive Grado Labs headphones, I can't tell the difference between original WAV files and MP3s coded above 128kb/sec. Part of why I can't wait to get my EMPEG is to see for myself just exactly what settings on Lame give the best quality.


Why not go out and buy all this hardware today? Every month I wait, the hardware gets cheaper and/or faster. And, until I have the EMPEG, I won't really know what encoder settings I want to use. I want to rip my music once and be done with it.

Dan

On the subject of new CD-ROM drives... I just got a new Ricoh drive that's both a CD-RW and a DVD-ROM player. And it rips pretty decent, too. Impressive pieve of technology.

___________
Tony Fabris

I've got the old Ultraplex (32 speed) and Plexwriter 412c (Both SCSI) and I've been very happy with them both...

With MusicMatch 5 and my Celery 450a (Thanks for startup screen Tony =) I can rip a cd perfectly at 6.5 speed (I use 100% VBR)...
I suppose some people will find that slow, but I'm sure it's a processor limit at the moment... =)

Under windows, you can use the Audio FS mode with the plextor software, and it makes an audio cd appear as a bunch of wav files... This means you can use a command line encoder without using a ripper... =)

Jazz
(List 112, Mk2 12 gig #40. Mk1 for sale 4 gig #30, apply within)

and my Celery 450a (Thanks for startup screen Tony =)

You're welcome. For those interested, he's talking about this.

Under windows, you can use the Audio FS mode with the plextor software, and it makes an audio cd appear as a bunch of wav files... This means you can use a command line encoder without using a ripper...

Yeah, that's what I built my Jack software around. I found a sys driver that does the same thing (dunno where it came from though). My only problem is that I have trouble using WinAmp as my CD player if I use this driver. If I try to hit the Playlist function in WinAmp, it gives me the CD tracks but then tries to recursively catalog the .WAV files on the CD, which screws up the playback.

___________
Tony Fabris

I`d recommend AG or ACAT on defaults. Works fine in most cases. You may want to limit speed after a bit of trial and error to get rid of any pops or clicks. (buffered burst is defualt)
I have Pioneer 6x dvd rom, and it rips at 12x, but its accurate to, meaning i can cock my comp up and it just re reads, i get no pops or clicks.
Also have a Plextor 412 (?) 4x CDR, which holds 12x, but doesnt seem to be able to resync accurately, so if ripping with that i`d have to leave the comp alone.
Lame http://www.sulaco.org/mp3

Ian

> My current plan (once my EMPEG shows up) is to go out and buy...
>
> - a Plextor PlexWriter (available SCSI or ATAPI -- the new ATAPI one is $305 at CDW)
> - an IBM 75GB drive (enough space to eventually rip all 1000 of my CDs, $620 at CDW)

If you're using Linux, as you say you are, you're missing an opportunity to get better performance for less money. Buy 2 or more cheaper disks and RAID them using software-RAID. The best value disks I can see at the moment are the Maxtor 30Gb disks. 2 of these will cost you probably half what the 75Gb disk will cost, and in a RAID-0 configuration will near enough fit your 1000 CDs, and easily outperform any single disk in terms of I/O speed. Better still, get 3 or 4 of the things and join them in a RAID-5 to still give good read speed and added security.

I've had this running for a year and a half on my home server (root on RAID-5, MP3 storage on RAID-0) and I can happily have a couple of workstations ripping/encoding to the server, whilst listening to a third.

By the way, what backup are you going to use for 75Gb? I've got a 10/20Gb Travan drive (not the best performance, I know, but easily the cheapest option for the capacity), and backing up my MP3s (only 35Gb of them currently) is extremely tedious.

Cheers,

Bruno
bruno@prior.ftech.co.uk

My god, that makes my home linux server seem small. One day I plan to get some bigger hard drives for it, since mine at this point has an 8, 6, 5, and 4 gig. (3 IDE, one SCSI and one SCSI CD-Rom) Also, I need to get a tape drive for it, since I get more and more worried about loosing data every day.

backing up my MP3s (only 35Gb of them currently) is extremely tedious.

Think how tedious it would be if you were to back up to floppy disks -- I figure it would take right around 25,000 of them.

tanstaafl.

"There Ain't No Such Thing As A Free Lunch"

Although not conclusive, here's a link that shows an attempt at measuring for objective results between different encoder engines:

http://www.ars-technica.com/wankerdesk/1q00/mp3/mp3-1.html

(no, I didn't make the address up, it is truly "ars-technica" and "wankerdesk"!).

Anyway, I was equally curious what MP3 encoding at different rates would do so I did a little experiment:

First ripped 3 different pieces of music as WAV's and then converted each to MP3 at resp. 192, 128, and 96k. These MP3 files I converted back to WAV and burned the whole lot on to a CDR (12 tracks total; 1 x WAV, 1 x 196k; 1 x 128k; 1 x 96k for each piece of music).

This way I could listen for myself on my own (pretty decent) stereo system, using the original CDs I ripped from as a benchmark.

I know you can blow wholes in this procedure as a means of assessing MP3 compression ratios, but the results were pretty interesting none-the-less and showed up much more differences than using my PC sound card (cheap generic) run through the stereo system.

The differences are very noticable in my opinion, but I decided that 128k will be good enough for me for the time being (I'm also thinking of using same MP3s in diddy solid state portable, so size does matter).

Whatever is good enough is a personal choice anyway, it's not someone can tell you; what I may find a big difference you may find barely noticeable or important and vice-versa. Try it out for yourself, perhaps method described above gives a little bit more insight.

On the issue of ripping; I've found that using my samsung 40x DVD-Rom gives worse results than using the Philips CD burner to rip from.

Cas

Could someone explain what makes a CD drive better or worse than another (other than speed). I was under the impression that it was simply a digital pickup. Do some drives misread bits more often? If so I thought CD audio had a large amount of error correction built in.

Thanks,

Alex Lear

Alex,
From what I gather from reading the material on the net, CD Audio does indeed have error correction built into it. However, most cd ripping programs just read the digital information from the audio disk without attempting to use any form of error correction.
That's why you would use a program like Exact Audio Copy http://www.exactaudiocopy.de/, which does attempt to do error correcting. This way you have a wave file which is much closer to the original form from the cd.

As for which bitrate is good for you, I agree that it's a personal choice. I must say though that after reading the comparison above, and also at http://www.r3mix.net, I decided to play with variable bitrate compression.

When compressing Tom Sawyer from the remastered Moving Pictures, I told LAME to use VBR at the highest bitrate. At one point in the intro to the song, the bitrate for a frame went as high as 320kbs, so I figure if I were to limit it to 128kbs, I would be missing a whole lot of info. Unfortunately, I'm nowhere near my home audio equipment, so I can't do a good listening test.

Even without listening to it on a good system, I could still tell the difference between a 128k encoded file and a VBR file at highest setting, even though the file sizes were so close that the difference is not even worth mentioning.

Just adding my 2 cents to the whole discussion.

-Trevor

Alex: Could someone explain what makes a CD drive better or worse than another (other than speed). I was under the impression that it was simply a digital pickup. Do some drives misread bits more often? If so I thought CD audio had a large amount of error correction built in.

Good question, Alex. We should probably put this in the FAQ. Yes, some drives mis-read more often than others, but when we're talking about Digital Audio Extraction, we're not so much worried about the bits as we are worried about the blocks. Read on...

Trevor: From what I gather from reading the material on the net, CD Audio does indeed have error correction built into it. However, most cd ripping programs just read the digital information from the audio disk without attempting to use any form of error correction.

Not strictly true, Trevor. CD Audio does use error correction (but on a much smaller scale than ROM data CDs do). It's just that the error correction information can't be sent back to the computer with most CD-ROM drives. Most CD-ROM drives aren't built with Digital Audio Extraction in mind, so the circuitry and drivers to perform block-accurate addressing on audio tracks simply isn't there. There are some notable exceptions (Plextor), but for the most part, it's up to the ripping software to perform after-the fact error correction by overlapping the reads and sliding the samples together. A discussion of this can be found at the CD-Recordable FAQ.

Which is why every decent ripper out there has error correction of some kind or another built-in. It's just that some do it more agressively than others. Some default to an agressive error correction scheme, others default to a faster method with less error correction, and require you to tweak the settings to error-correct more agressively. In my personal experience, I've found that different pieces of software will/won't work on different drives, regardless of whether or not they claim to do great error correction.

For instance, on my work computer, the only software that can work at all is WinDac32. Exactaudiocopy and Audiograbber/Audiocatalyst choke completely on it. On my home computer, EAC and WinDac die, but AG does fine.

Something I've noticed is that many CD-recorder drives support digital audio extraction quite well, especially if you use their bundled software to perform the extraction. For instance, I can rip quite well on my new DVD+CD-RW drive using the functions in Easy CD creator 4.

___________
Tony Fabris

Tony,
Thanks for setting me straight. I wasn't real sure on this, but I do notice a difference when I use EAC on scratched disks. I know that this is because it is more agressive with the error correction.

I haven't even tried it on my CDR yet, as it's in a box with the rest of my components, as my desktop is on loan to the Subaru Impreza Owner's Club, running their copy of UBB. I do know that my current drive does NOT support error correction codes.

Thanks again,
-Trevor

...The differences are very noticable in my opinion...

Cas -- don't leave us hanging like that. What did you notice? At what bit rates? How noticeable was it? Was the sound better, worse, or just "different"? Most pronounced at higher frequencies or lower frequencies? Was the sound brighter or muddier?

C'mon, let us know what you found!

In fact, if you could bear to part with it, I would love to borrow that actual CD that you burned so I could perform the same sort of listening tests you did. (Hmmm... do you think RIAA would come after us?)

tanstaafl.

"There Ain't No Such Thing As A Free Lunch"

Hi Tanstaafl,

C'mon, let us know what you found!

Well, Between WAV and 192k encoding, youn loose top-high frequency detail (cymbals get a bit fizzy); slight degradation of stereo image. I also use(d) MiniDisc (Sony MZ-R35) and found the two quite comparable in degradation but different in character (MD is bit cleaner, MP3 warmer).

192k vs. 128k:
I felt this difference was bigger between original WAV and 192k. Same applies, but in higher order; particularly narrowing of soundstage was noticeable. Stuff that was firmly located to the left made a shift inwards. Dynamics seemed to have suffered as well; loud passages aren't simply as loud as the original with far less "slam" when they hit.
Biggest difference in the bass is that it looses its firmness and less defined; there isn't a loss in amount of bass and the lowest frequencies still seem to be there. if anything, subjectively, there seems to be more, as it's more coloured.

128k vs. 96k.
This is just very poor, not worth discussing in detail, read above and multiply by several factors.

The results vary with the type of music. The first piece I used was a small jazz ensemble with lots of detail and ambience (from Chesky recordings); second came large scale orchestra (from "West Side Story" on EMI); third: light pop with woman's voice (again Chesky).

It effects of datareduction appear to wreak the greatest havoc with the classical piece, least with the fairly uncomplicated track with woman's voice.

The deciding factor for me to stick with 128k fixed was that the 128k quality wasn't objectionalbe and it makes compatibility with other stuff (such as solid state portables) easier. Also, when listening in the car imaging and soundstaging in the audiophile sense become pretty moot anyway.

For your info; ripping and MP3 encoding done with MusicMatch 5.00 (supposedly Fraunhofer encoder). Evaluating was done on my domestic music system which consists of NAD S100 pre, S200 power, S500 CD, Dali Diva speakers.

In fact, if you could bear to part with it, I would love to borrow that actual CD that you burned so I could perform the same sort of listening tests you did.

well, if you really can't toast a CD yourself I may considedr sending you a copy all the way from good 'ol Holland to Alaska!



Cas

Whoa..., Doug I'm from Alaska too. I go to school in Montana but I have permanent residence in Anchorage. Alaska is the last place I would expect another empeg fan because of its reletively small population. I never would have checked you bio. This forum needs to put the state or country by the user name or have a page that lists users by state/country.

Anybody else from Alaska, or even Montana?

Alex Lear

192k vs. 128k:

Was 192 as high as you went with the tests?

It might be fun to take one or two pieces which exhibited the worst aliasing (to your ear) and try them at higher bitrates, or at highest-quality VBR rates using LAME or Xing.

When I encode in high-quality VBR, I've seen some frames jump to 320kbps...

Biggest difference in the bass is that it looses its firmness and less defined;

I'm starting to hear more and more people say this. It sounds illogical to me (the low frequencies should be reproduced quite well at any bit rate), but these comments are starting to reach a critical mass. I haven't ever noticed any problems with the bass, but now I'll start listening more closely.

In some cases, you can attribute those comments to the fact that many people are previewing the MP3s on different sound reproduction equipment than the CDs they're comparing them to. But in your case you did the Right Thing and tested the encoder only by decoding the MP3 back down to a wav file and burning that to a CD. So it's not an equipment thing in your case.

Same applies, but in higher order; particularly narrowing of soundstage was noticeable. Stuff that was firmly located to the left made a shift inwards.

Important information for the "soundstage" discussion we were having a while ago. Of course, this is the MP3-soundstage-limitation question, not the Empeg-soundstage-limitation question.

Also, when listening in the car imaging and soundstaging in the audiophile sense become pretty moot anyway.

Don't let Doug hear you say that! Many of us car audio enthusiasts have better sound systems in our cars than in our homes.

It effects of datareduction appear to wreak the greatest havoc with the classical piece, least with the fairly uncomplicated track with woman's voice.

Interesting. I've noticed that solo voice can really show up artifacting. Just yesterday I was noticing some compression artifacts on a solo female voice track (The album happened to be "Mouth Music" by Swan and Mackenzie). I didn't think the track was complex enough to cause audible artifacting, the voice was firmly in the midrange and very clean. But what it showed was that having all the "air" around the voice allowed faint artifacts to become audible, especially on note decays, that would otherwise have been buried in the rest of the mixed instruments.

And this was egregious artifacting, too. It wasn't subtle. You know that gurgly, alien-talking-underwater kind of sound you hear on really-heavily-compressed streaming RealAudio tracks? I could hear that as her voice trailed off on certain notes. And this is at 128k, not at lower bitrates where you would expect to hear it.

Mind you, I had to be listening at full volume to hear it. And I was running the voladj kernel, so it was cranking up the note decays well above the noise floor, so it's kind of unfair, I guess. Still, it was there.

___________
Tony Fabris

Hi!

I'm in Anchorage as well. I also have the only Mark I that is in Alaska.

Reg_2845 Serial #00173

Also, when listening in the car imaging and soundstaging in the audiophile sense become pretty moot anyway.

Don't let Doug hear you say that! Many of us car audio enthusiasts have better sound systems in our cars than in our homes.

But you've gotta admit, in most people's cases this is true. The soundstage isn't too bad for the driver in relation to the rear speakers in most cars, but with speakers in the door panels it's way off. If you've got kick panels it's alot better because you'll move the stage forward a bit, but there's no way around the fact that--in most vehicles--the placement of the speakers is not ideal (unless you happen to own a car where the driver's seat is in the middle)


DiGNAN

...but with speakers in the door panels it's way off.

Not necessarily.

Being basically selfish (and since most of my passengers could care less about staging and imaging) I shift the left-right balance one notch towards the passenger side -- i.e, the right side of the car gets a little more volume than the left side. This has the effect of exactly centering the image for the driver, at the expense of the passenger -- but it is a small change, and unless the passenger heard it both ways, he would never know.

In competition judging, one of the tracks is a series of seven snare drum beats that travel left to right across the sound stage. Ideally, they would be positioned for both driver and passenger like this:

x........x........x........x........x........x........x

And there are some cars that compete that do, indeed, achieve this, albeit expensively.

Most cars come out like this:

x....x....x....x.................x....x....x....... for the driver and the mirror image for the passenger.

According to the judging sheets, my car comes out more like this:

..x......x......x..........x..........x....x....x..... for the driver, and
........x....x....x........x........x......x......x.. for the passenger.

In other words, the stage is a bit narrower than optimal, and the separation gets compressed on the left and right sides, and compressed a little more on the side opposite the listener. (When in championship judging mode with two judges, one in each front seat, I don't shift the left-right balance, but leave it centered.) Because of the acoustics of my car (a fairly large station wagon) my staging and imaging is better than most, due more to blind luck than any cleverness on my part.

Finally, my car is quieter than most -- no audible engine noise, lots of sound deadening, aerodynamically shaped enough that wind noise is pretty minimal, so even at highway speeds I keep most subtle nuances of the music. rjlov's "normalizer" program will make it even better, but I have to actually have an empeg first, I guess... :-(

I guess the above rambling discourse is all just to say that a car can be a better listening environment than most people give credit, if you have the proper car and the proper equipment and tuning. I think the fact that the car is such a controlled environment helps: you don't have to deal with asymmetrical rooms or rooms with some walls close by and other walls effectively not there at all as in an open living room environment.

tanstaafl.


"There Ain't No Such Thing As A Free Lunch"

Being basically selfish (and since most of my passengers could care less about staging and imaging) I shift the left-right balance one notch towards the passenger side -- i.e, the right side of the car gets a little more volume than the left side

So I was right. I knew you could do that, and I would but I typically do have picky passengers .

DiGNAN

Was 192 as high as you went with the tests?

Yes, I wanted to stay close to 128k to see how much "damage" "internet-standard" would do.

It might be fun to take one or two pieces which exhibited the worst aliasing (to your ear) and try them at higher bitrates, or at highest-quality VBR rates using LAME or Xing.

When I have a bit of time (!) that's part of the next experiment. I'm also very curious to see what the Xing, Blade, Lame, etc. encoders sound like in various modes compared to an original WAV. Don't hold your breath for the results though...

Any other suggestions to add to the experiment?

Cas

Bruno writes: Buy 2 or more cheaper disks and RAID them using software-RAID.

I hadn't seriously considered this before, but it's an interesting idea. I don't really care about speed, per se. The sole purpose of this huge disk (or disk group) is to hold MP3s.

As far as price goes, let's check again at CDW:

- IBM 75GB 7200rpm: $589 ($7.85/GB)
- IBM 60GB 7200rpm: $480 ($8.00/GB)
- IBM 45GB 7200rpm: $299 ($6.64/GB)
- IBM 30GB 7200rpm: $209 ($6.97/GB)

- Maxtor 60GB 5400rpm: $300 ($5.00/GB)
- Maxtor 40GB 7200rpm: $265 ($6.63/GB)
- Maxtor 30GB 5400rpm: $145 ($4.83/GB)

- Western Digital 45GB 5400rpm: $199 ($4.42/GB)

So, at 7200rpm, the best deal seems to be a tie between IBM and Maxtor. If you're trying to trade off speed for space, the Western Digital disks have the best price/storage ratio.

By the way, what backup are you going to use for 75Gb?

Backups? I've still got a huge bookcase full of the original CDs. I don't really care so much about my home machine. Anything that really matters is at work on the NetApp box (RAID with all the bells and whistles). Besides, the EMPEG (once I can buy large-enough laptop hard drives) will be the real backup.

($4.42/GB)

I once spent (actually, the company I worked for spent...) $1200 for an 80 MB hard drive. That works out to be about $15,000/GB, that was in the mid 1980's.

In the 1970's, another company I worked for used DataPoint mini-computers whose 5 MB hard drives (the size of a small filing cabinet) cost over $8,000. That's $1,600,000/GB.

And you want me to comparison shop so I can save something like 40 cents per gigabyte.... Ri i i g h t....



tanstaafl.



"There Ain't No Such Thing As A Free Lunch"

Good prices! Still, I intend to wait until the EMPEG is in danger of arriving. Then I start spending inordinate amounts of time feeding CDs to the machine. By then, prices will only be cheaper.

Although i`ve never performed a real blind test, i`ve done test like this before and noticed the same things.
I made mp3`s, and burnt them in wav format, then played em in car.

128kbps gives u bass which is "different" I cant explain why exactly, it seems exagerrated and boomy compared to the orginal. (my mates get in and marvel at how much bassier it is if i have 128k mp3`s playin, i dont like it, it sounds fake)
The top end becomes swirrly and swishy, imaging is poor (excuse the techy terms :))

192kbps is good, bass is pretty accurate and top end swishyness goes, something is different about the sound, but unless listening back to back with a wav i couldnt noitce it. Imaging is nearly perfect

HQ Lame VBR (about 180-230kbps) awesome, high end is perfect, bass stays unchanged. Soundstage perhaps "lifts" slightly, imaging is spot on, imo this is what eveyone should use, its almost perfect.

256kbps, Huge files, perfect sound, same soundstage points as above.

I used either lame or Opticom Mp3 Producer 2.1 (ffraunhofer (sp?))

its the classic quality v filesize thing, i would always use 160k or more, and i`d never ever even if i had no other enocder, use xing :)
Xing gave me swisshy sounds and big bass at any bitrate, cant stand it.

What settings and what version of lame are you using?

Sean