|
|
|
Beginners question: ripping speed?
|
|
Newbie
|
26. October 2012 @ 06:24 |
Link to this message
|
HI, I just started ripping my CD collection and it goes extremely slowly. The speed is about 1.2X or so, a single half hour CD takes over 20 minutes to rip.
computer: Very state of the art SandyBridge cpu overclocked to almost 4GHx speed.
software: Cdex.
settings in Cdex: read sectors 52 (anything higher produces corrupt,silent compressed files). Overlap =7.
CDrvie: generic, not a name brand.
format: OGG (not MP3).
MY QUESTIONS : Is the drive too slow, is that probably my problem? Will buying a faster drive likely help a lot?
Does VBR mp3 rip faster than OGG (Og Vorbis)?
I think I should be seeing much faster ripping speeds, but I'm new to this, so not sure...
Drivr!
|
|
Advertisement
|
|
|
|
|
Mez
AfterDawn Addict
|
26. October 2012 @ 09:29 |
Link to this message
|
You will not see vastly faster rips no matter what you do. The process is far, far trickier than you can amagine. For one thing the CDs are not data disks. They are written to foil attemps to rip them.
It is your ripping software. However, 20 min is not all that slow. Make sure your disk surface is as clean as it can be. That will speed the process. A Lame VBR mp3 will probably be slightly superior in quality and is compatible with more applications than an ogg. Lame had massive help (thousands of users) from the audio world debugging the VBRs because they were the first to get close to a near perfect replica of a wave. At the end it was a pefect replica to the human ear. Subsiquent processes do not get 1% of that attention. This effort preceded AD. With hu8nderds of 'also rans' who really cares what artifacts each produces? If persons want to use them who cares? They are probably too ignorant to even notice the artifacts or their play back equipment is too inferior to replicate the artifacts so they can't be heard.
You are probably ripping with a secure mode. dbPowerAmp uses burst mode and will down shift to secure mode if it detects a problem. Secure mode reads the same sector many times to see if there is a difference in reads. If there is a difference, it will run a ton more reads to insure a perfect read.
For more info you can read my top sticky. There is a great deal of technology that goes into making a CD, playing a CD and ripping a CD. Playing a ripped tune from magnetic media such as an ipod or HD will be an exact replica of what was recorded or at least what was mastered. Playing a purchased CD will not be. It is more likely a burned CD will be more accurate. That is all in the top sticky.
|
Newbie
|
27. October 2012 @ 01:26 |
Link to this message
|
Thanks, Mez, lots of good stuff there, learned a lot. Maybe I'll purchase dbPowerAmp if I think I need it.
CDEX: I had downloaded the latest beta version which had no help file. I went back and got the release version and the help file showed me how to speed up the process. An CD that played an hours music was taking 48 minutes, now it's taking 9 minutes for the same CD, and am using more overlap (13 instead of 7) for better error correction (I think). I just set CD speed from 0 to 30 to accomplish the increase in speed.
Online music such as New Riders Purple Sage Filmore East 1971-04-28...
http://archive.org/details/nrps1971-04-...8176.sbeok.shnf
...has the concert music files in VBR mp3 and Ogg Vorbis. To my ears the Ogg files played at higher quality hifi, which is why I started using Vorbis last night when I first began ripping music. Maybe I'll give mp3 VBR another shot, my sample was based on this one site, after all :)
Drivr!
This message has been edited since posting. Last time this message was edited on 27. October 2012 @ 01:27
|
|
Mez
AfterDawn Addict
|
27. October 2012 @ 09:03 |
Link to this message
|
The ogg files might have been better quality. Any given format can't guarantee a certain level of quality, they can only provide an upper limit to the quality. Especially public domain recordings can be very iffy. I have heard very low quality lossless. The recording was likely made with a pocket recorder then converted to lossless. If the raw recording had a 50 br converting it to lossless with a br of 1500 doesn't improve the quality at all.
|
|
hello_hello
Junior Member
|
30. October 2012 @ 17:37 |
Link to this message
|
Originally posted by CrowDrivr:
...has the concert music files in VBR mp3 and Ogg Vorbis. To my ears the Ogg files played at higher quality hifi, which is why I started using Vorbis last night when I first began ripping music. Maybe I'll give mp3 VBR another shot, my sample was based on this one site, after all :)
Pretty much all lossy compression is "transparent" at a high enough bitrate. I can't give you specifics for every lossy format as I mainly use MP3, but information regarding the LAME MP3 encoder can be found here: http://wiki.hydrogenaudio.org/index.php?title=LAME
Personally I never rip CDs directly to a lossy format. Hard drive space is cheap these days and most people have plenty of it, so I'd suggest ripping to a lossless format such a FLAC. From there you can convert the flac files to a lossy format for playing with a portable device etc. The advantage of doing it this way is if you ever decide to change lossy formats you don't have to re-rip the CDs. Just convert the flac files to the new format. At the moment I rip to flac then convert to MP3 using the LAME encoder and the standard v2 VBR preset.
I'd recommend trying foobar2000. As well as being a good media player it's also a good ripper and converter and will rip using "secure" ripping.
I don't know if it relates to your problem but when I originally built this PC I couldn't rip a CD in secure mode. I tried different software, different drives, nothing helped. Ripping in secure mode took forever. One day I tried again and the problem was gone. To this day I don't know what caused/fixed the secure ripping issue. It may have been a BIOS update or a Windows update, but I'll probably never know.
|
|
Mez
AfterDawn Addict
|
31. October 2012 @ 09:47 |
Link to this message
|
hello_hello as I mentioned above secure ripping takes a great deal of time. As I also explain in the top sticky, dbPowerApm fully integrated with Accurip rips in bust mode. If accurip says the read was good for a block it goes to the next it doesn't read the same block twice unless there is a problem with the read.
Foobar2000 has tools you might use if you are an audiophyle. Otherwise, why use it? It is not as good as the 3 top rippers so why would anyone use a third rate tool? It has nothing to do with quality either. That is codex dependant.
I rip to lossy just because I archive the CD. If it is a borrowed CD I might do things differently.
|
|
hello_hello
Junior Member
|
1. November 2012 @ 04:45 |
Link to this message
|
Originally posted by Mez:
hello_hello as I mentioned above secure ripping takes a great deal of time.
I never said it isn't slower. I was talking about something else.
Originally posted by Mez:
As I also explain in the top sticky, dbPowerApm fully integrated with Accurip rips in bust mode. If accurip says the read was good for a block it goes to the next it doesn't read the same block twice unless there is a problem with the read.
That's certainly not how it's described on the dbpoweramp site. The way I read it accuratrip is used to verify a rip of an entire track, not on a "block by block" basis.
Originally posted by Mez:
Foobar2000 has tools you might use if you are an audiophyle.
For example?
Originally posted by Mez:
Otherwise, why use it?
Foobar2000 is free. dbpoweramp is not. Does dbpoweramp even support replaygain? I couldn't see it mentioned anywhere on their web site.
Originally posted by Mez:
It is not as good as the 3 top rippers so why would anyone use a third rate tool? It has nothing to do with quality either. That is codex dependant.
It's also nonsense. Foobar2000 rips using non-secure ripping (burst mode), it has two secure ripping modes and it also supports accuraterip. http://wiki.hydrogenaudio.org/index.php...000:Ripping_CDs
Originally posted by Mez:
I rip to lossy just because I archive the CD. If it is a borrowed CD I might do things differently.
I rip to lossless so I only ever need to rip a CD once. If I wanted to rip a CD a second time, I might do it differently.
As you mentioned it, I had a read through your sticky. What's the "density" of lossy audio you refer to in post #10? The post doesn't make sense to me.
Post #11 is pretty much nonsense by the way. Dynamic range compression of music has nothing to do with increasing the level of low frequencies while decreasing the level of high frequencies as such, it's purpose is to decrease the volume difference between the quietest and loudest parts of an audio track. And the "On the soap box" paragraph is basically wrong from start to finish.
A sampling rate of 44.1k is used for CDs because it's required to accurately sample frequencies up to slightly over 20khz. There's no empty "ultrasonics" on CDs.
http://en.wikipedia.org/wiki/Sampling_rate#Sampling_theorem
The belief that you must be losing something between lossy and lossless is 100% true, and it's not just "ultrasonics". You mightn't be able to hear the difference, but by definition "lossy" compression throws information away. http://en.wikipedia.org/wiki/Data_compr...dio_compression
And.....
78 cartridges don't eliminate the RIAA equalisation found in 33 cartridges. There is no RIAA equalisation in cartridges as such. LPs use a particular EQ curve when they're mastered which is compensated for (reversed) by the pre-amplifier. Old 78 records use a different EQ curve which must be compensated for when sampling 78s, even if a 78 cartridge is used.
It's easy to verify. Plug a tape deck or CD player into the tape or CD input of an amp and listen to a recording. Now plug the tape deck or CD player into the phono input. There'll probably be a volume difference, but there will be an EQ difference. That's because the pre-amp is compensating for the different EQ curve it expects via the phono input. It'll compensate for the same EQ curve whether you're plugging a tape deck, a CD player or a turntable into the phono input. In the same way, software which records LPs would normally apply the EQ curve which 33s and 45s use. Even when recording a 78 using a 78 cartridge, that EQ curve needs to be removed or reversed.
This message has been edited since posting. Last time this message was edited on 1. November 2012 @ 11:06
|
|
Mez
AfterDawn Addict
|
1. November 2012 @ 09:35 |
Link to this message
|
That's certainly not how it's described on the dbpoweramp site. The way I read it accuratrip is used to verify a rip of an entire track, not on a "block by block" basis.
That may be out of date. That is that way things were when I wrote the article. I also have not updated my ripper in years since I let my license laps.
Originally posted by Mez:
--------------------------------------------------------------------------------
Foobar2000 has tools you might use if you are an audiophyle.
--------------------------------------------------------------------------------
For example?
A/B testing
Transcoding
normalization
Foobar2000 is free. dbpoweramp is not. Does dbpoweramp even support replaygain? I couldn't see it mentioned anywhere on their web site.
No dbpoweramp is a ripper. As I see it normalization really doesn't have anything to do with ripping.
It's also nonsense. Foobar2000 rips using non-secure ripping (burst mode), it has two secure ripping modes and it also supports accuraterip. http://wiki.hydrogenaudio.org/index.php...000:Ripping_CDs
OK I am out of date - sorry
Originally posted by Mez:
--------------------------------------------------------------------------------
I rip to lossy just because I archive the CD. If it is a borrowed CD I might do things differently.
--------------------------------------------------------------------------------
I rip to lossless so I only ever need to rip a CD once. If I wanted to rip a CD a second time, I might do it differently.
Good for you!
As you mentioned it, I had a read through your sticky. What's the "density" of lossy audio you refer to in post #10? The post doesn't make sense to me.
Post #11 is pretty much nonsense by the way. Dynamic range compression of music has nothing to do with increasing the level of low frequencies while decreasing the level of high frequencies as such, it's purpose is to decrease the volume difference between the quietest and loudest parts of an audio track. And the "On the soap box" paragraph is basically wrong from start to finish.
A sampling rate of 44.1k is used for CDs because it's required to accurately sample frequencies up to slightly over 20Hz. There's no empty "ultrasonics" on CDs.
You may not have understood where I was comming from. Let me explain though I am sure you will dissagree. An hour of silence as a wave is the same size as an hour of symphony because the bit rate is fixed. The bit rate was sized to allow for untrasonic frequencies.
http://en.wikipedia.org/wiki/Sampling_rate#Sampling_theorem
The belief that you must be losing something between lossy and lossless is 100% true, and it's not just "ultrasonics". You mightn't be able to hear the difference, but by definition "lossy" compression throws information away. http://en.wikipedia.org/wiki/Data_compr...dio_compression
Agreed, depending on the codex type of compression and version. Elimination of high frequencies is common to all because it is an easy process and removes a great deal of data.
And.....
78 cartridges don't eliminate the RIAA equalisation found in 33 cartridges. There is no RIAA equalisation in cartridges as such. LPs use a particular EQ curve when they're mastered which is compensated for (reversed) by the pre-amplifier. Old 78 records use a different EQ curve which must be compensated for when sampling 78s, even if a 78 cartridge is used.
Interesting, I read an article about such equalizations you claim do not exist and provided a 'tool' to compensate for over a dozen brands that DID equalize. That was a few yrs ago and I can't remember the details or even how many brands were covered. In fact now that I think on it it may have been the in the mastering. I do not remember.
It's easy to verify. Plug a tape deck or CD player into the tape or CD input of an amp and listen to a recording. Now plug the tape deck or CD player into the phono input. There'll probably be a volume difference, but there will be an EQ difference. That's because the pre-amp is compensating for the different EQ curve it expects via the phono input. It'll compensate for the same EQ curve whether you're plugging a tape deck, a CD player or a turntable into the phono input. In the same way, software which records LPs would normally apply the EQ curve which 33s and 45s use. Even when recording a 78 using a 78 cartridge, that EQ curve needs to be removed or reversed.
I suggest you put you comments into the sticky. It is opened for comments. We rairly have a poster of your caliber on this forum and some of my information seems to be out of date and some could be out right wrong. To tell you the truth, so few persons read that information, I am guessing less than 1 a month, I don't have the inclination to spend the hrs researching all the areas and correcting the errors. I say this because if you review this yrs queries at least 1 in 5 queries would have been answered if they read that info. Who knows maybe it is read more than I guess. I do not remember how long it has been up but you are the first person to complain. Since you will have the last postings maybe someone reading your posts a few yrs from now will find issue with what you posted. I feel lucky you didn't find more issues since the volume of information is fairly large. Maybe you just didn't read it all. I don't claim to be the most technical preson that posted to this board. Many yrs ago there were many far more technical than I but I am the only one that has stayed on.
I suggest you frequent this board more often.
Thanks for your interest,
Mez
|
|
hello_hello
Junior Member
|
1. November 2012 @ 14:40 |
Link to this message
|
Originally posted by Mez:
For example?
A/B testing
Transcoding
normalization
Foobar2000 is free. dbpoweramp is not. Does dbpoweramp even support replaygain? I couldn't see it mentioned anywhere on their web site.
No dbpoweramp is a ripper. As I see it normalization really doesn't have anything to do with ripping.
dbpoweramp was originally called "dbpoweramp Music Converter" and it's purpose was to encode/transcode audio files. The conversion part of the program still uses that name.
http://www.dbpoweramp.com/dmc.htm
Back in it's early days the "ripping" part of the program was called "dbpoweramp Audio CD Input" and it allowed you to convert audio directly from CD. Like many programs it gradually evolved to become the "dbpoweramp CD Ripper" it is today.
dbpoweramp Music Converter has a DSP for "normalising" audio. http://www.dbpoweramp.com/dbpoweramp-dsp.htm I can't say I've even used it myself. I'm not aware of a similar plugin for foobar2000. It appears I was wrong though. dbpoweramp does support ReplayGain via a DSP.
Originally posted by Mez:
You may not have understood where I was comming from. Let me explain though I am sure you will dissagree. An hour of silence as a wave is the same size as an hour of symphony because the bit rate is fixed. The bit rate was sized to allow for untrasonic frequencies.
I don't really understand what you mean by "ultrasonic" frequencies.
Yes when it comes to lossless LPCM audio the bitrate is fixed and it's high enough to ensure accurate reproduction from 20hz to 20khz. The bitrate is a combination of the sample rate (44.1k) multiplied by the bit depth (16 bit) which gives you the bit rate. Multiply that by 2 for stereo and you end up with 44100 x 16 x 2 = 1411200 bits per second, or 1411 kbps.
But yes, the sample rate and bit depth are fixed, so a second of silence will use the same number of bits as a second of techno.
Even lossy audio can use a fixed/constant bitrate though, so it's just as capable of wasting the same number of bits on a second of silence as it'd use for a second of techno, which indicates it's not a fixed bitrate as such which explains the difference between lossy and lossless audio file sizes. I don't really understand it all that well myself but lossy audio doesn't have a fixed bit depth as LPCM audio does (16 bit etc) and it uses algorithms to encode the audio, and the algorithms use psychoacoustics in order to compress the audio along with other techniques such as joint stereo.
http://en.wikipedia.org/wiki/Psychoacoustics#Software
http://en.wikipedia.org/wiki/Joint_%28a...29#Joint_stereo
Most if it is over my head, but I guess the point I was trying to make was that lossy compression can in fact change the original audio quite considerably (although obviously the objective is to do so without you being able to hear a difference) and that it's not as simple as lossy audio saving bits by not wasting them on silence or unnecessary ultrasonic frequencies.
This message has been edited since posting. Last time this message was edited on 1. November 2012 @ 14:44
|
|
Mez
AfterDawn Addict
|
2. November 2012 @ 09:45 |
Link to this message
|
Anything above 20khz is ultrasonic. I forget what the cut off is for a CD is but I thought it was conciderably higher than 20K. However, math does not lie and I learned something big today. I probably picked that 'fact' on the internet somewhere. Any fixed rate waists space and since VBR has a higher ceiling than 320 fixed BR and very close to lossless (19.8K if I remember correctly), since it usually averages at about 200 is a wise choice.
You are right about dbPA it does transcode as well as rip. I put normalization into a different catigory while you don't. By the way the trial version continues to work in the extention to windows explorer ( or did a few yrs ago) after the experation. You just can't add any codex. I wouldn't use any normalization process unless it was highly rated and I read up on the process to make sure it was what I want. I would shy away from dbPA because I usually try to use the best not an also ran.
|
|
Oceanskey11
Newbie
|
9. November 2012 @ 15:21 |
Link to this message
|
Use the best Ripping Program that exists, Exact Audio Copy. It is easy to find on the web, is completely free, and by far the best ripping software. When you use it to rip, rip them uncompressed in WAV format, then convert the WAV's to VBR MP3's for storage. That is the way to go in my opinion.
|
|
Advertisement
|
|
|
|
Mez
AfterDawn Addict
|
14. November 2012 @ 13:30 |
Link to this message
|
The original problem was the member complained about speed. Althogh EAC produces a quality rip it is one of the slowest rippers you can get. It is one of the best but not the best. It has problems ripping certain configurations/structures. It is also slow and produces no better quality than the other great rippers.
|
|
