Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: CD de-emphasis? (Read 25490 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

CD de-emphasis?

Reply #26 –
Quote from: greynol on
This makes me wonder why EAC shows this on a per-track basis in its GUI.  I guess it's therefore impossible for it to ever say that only some tracks have pre-emphasis.



This is what made me wonder.  Has anyone ever seen an EAC GUI output showing some tracks on a disc having PE and some not?


CD de-emphasis?

Reply #27 –
Quote from: dnewhous on
Quote from: krabapple on
Is there a site anywhere that compiles a list of CDs with pre-emphasis?

Your ears should tell you.  There should be some sort of marker.  On early Japanese CDs it's the "TO" marker after the matrix number.



Just to straighten out some info, the bolded part is incorrect. The 'TO' in the matrix number means it was pressed at a Toshiba plant, has nothing to do with the pre-emphasis flag.

I have the Pink Floyd - Dark Side of the Moon CP35-3017 black face cd that is made in Japan and has the TO in the matrix number but does not have the PE flag set.

CD de-emphasis?

Reply #28 –
Quote from: Kees de Visser on
I've never tried it but I suppose that if you can find a silent part of the audio with only noise, it might be possible to find out if emphasis was used because the noise will have a 10 dB lift towards 20 kHz.

This seems to be true for the CD that I think has pre-emphasis with no flags. Does the emphasis only go up to 20kHz?

CD de-emphasis?

Reply #29 –
Quote from: Zilog Jones on
Quote from: Kees de Visser on
I've never tried it but I suppose that if you can find a silent part of the audio with only noise, it might be possible to find out if emphasis was used because the noise will have a 10 dB lift towards 20 kHz.

This seems to be true for the CD that I think has pre-emphasis with no flags. Does the emphasis only go up to 20kHz?


How exactly do you use this method? I have Audacity and have just tried plotting spectrums for the silent parts of a song known to have PE. I don't see anything unusual from the graph.

CD de-emphasis?

Reply #30 –
Quote from: Zilog Jones on
One thing I've noticed with SoX's de-emphasis is that the output is very quiet - the album peak is 0.615814 now instead of 0.977264. I guess this is normal when de-emphasis is applied without normalization?

I'd say this is normal, because "de-emphasis" lowers the levels of certain frequencies.

While comparing SoX-Deemph and WaveEmph results I noticed (like many others in an older thread) that the SoX output sounded a little muffled. I am now convinced that this difference is because WaveEmph compensates partially for the volume loss you mention. I now normalize files before(*) applying SoX-Deemph on them, because my 80's CDs with pre-emphasis are mastered at a low volume. The result is as good as WaveEmph's to my ears. Not that I'd say that meddling with the audio data two times is audiophile, but if de-emphasis is required you heavily alter the data once anyway.

What I'm not sure about though is, if instead of RMS normalize, peak normalize or some other type of loudness adjustment would be more appropriate concerning the original dynamic range of the music. (Because of practical issues I don't want replay gain adjustment through meta data.)

(*) More volume means more used bits per sample means better algorithm effectivity, or am I wrong?

CD de-emphasis?

Reply #31 –
Hi,

I was wondering if anybody had a copy of WaveEmph they could put up on the web for me somewhere. The creator's site is down and I can't find anywhere else on the web to download it. I tried sox, but found the results to sound a little muffled compared to de-emphasised files I've heard from other people using unknown software.  Would be VERY grateful if anyone could help.

Thanks!

CD de-emphasis?

Reply #32 –
Quote from: royphil345 on
I was wondering if anybody had a copy of WaveEmph they could put up on the web for me somewhere. The creator's site is down and I can't find anywhere else on the web to download it. I tried sox, but found the results to sound a little muffled compared to de-emphasised files I've heard from other people using unknown software.  Would be VERY grateful if anyone could help.


I played with different possibilities of de-emphasis. I more or less sumarized my findings here
http://forums.slimdevices.com/showpost.php...mp;postcount=37

That sox de-emphsais sounds muffled i believe is nonsense. I wonder why even 2 persons post that. Pleas try to find a way to prove that!
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

CD de-emphasis?

Reply #33 –
Quote from: Wombat on
That sox de-emphsais sounds muffled i believe is nonsense. I wonder why even 2 persons post that.


Post #31 from instaud on this page:
Quote
While comparing SoX-Deemph and WaveEmph results I noticed (like many others in an older thread) that the SoX output sounded a little muffled. I am now convinced that this difference is because WaveEmph compensates partially for the volume loss you mention.

emphasis mine (no pun intended).

CD de-emphasis?

Reply #34 –
Quote from: lvqcl on
Quote from: Wombat on
That sox de-emphsais sounds muffled i believe is nonsense. I wonder why even 2 persons post that.


Post #31 from instaud on this page:
Quote
While comparing SoX-Deemph and WaveEmph results I noticed (like many others in an older thread) that the SoX output sounded a little muffled. I am now convinced that this difference is because WaveEmph compensates partially for the volume loss you mention.

emphasis mine (no pun intended).


Ok  If loudness=muffle then sox should be used together with --norm to un-muffle
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

CD de-emphasis?

Reply #35 –
My experience with Sox is alright, and it seems that the deemph switch does the job.

Take an album like "Pet Shop Boys Please" which has preemphasis on all original pressings. If you deemphasize with Sox, it will sound muffled. That was my impression too. Then I compared with a vinyl rip, and it sounded 99% the same. I say 99% because I could notice that the vinyl was still a bit brighter after all, but that must be inherent from the CD source.

CD de-emphasis?

Reply #36 –
WaveEmph isn't available via web.archive.org and I couldn't find any restrictions on redistribution, so I put WaveEmph 1.1.1 online temporarily here (click) (the filename is waveemph110.zip, but it contains v1.1.1). After unzipping, right-click on the .exe and set Win98 compatibility before running it the first time.

CD de-emphasis?

Reply #37 –
Quote
The program can't start because MSVBVM50.DLL is missing from your computer. Try reinstalling the program to fix this problem.

CD de-emphasis?

Reply #38 –
Quote from: mjb2006 on
WaveEmph isn't available via web.archive.org and I couldn't find any restrictions on redistribution, so I put WaveEmph 1.1.1 online temporarily here (click) (the filename is waveemph110.zip, but it contains v1.1.1). After unzipping, right-click on the .exe and set Win98 compatibility before running it the first time.


Thank you SO much mjb!!!  I didn't think I had much hope of finding this program! Awesome!!!

To the guy who had problems with MSVBVM50.DLL being missing... I think I read you're supposed to download that separately and put it in the WaveEmph folder or your system folder (not sure which). I actually have that downloaded, but I couldn't find WaveEmph. You should be able to Google for that file and download it.

EDIT: I tried it out. I still use Windows XP and I didn't have to do anything for compatibility. I popped the program in a folder with that MSVBVM50.DLL file just in case...

It worked fine right off the bat and I definitely think it produces better sounding results at the default settings than sox. The highs are a bit more open and the bass is noticeably deeper, probably just due to the gain or level of the recording remaining higher when using this program as some other posters pondered. But, it works better than sox for me without having to mess with anything else and I like it!

SINCERE THANKS!!!  It's exactly what I was hoping for and I never thought I'd get a reply to this post.

CD de-emphasis?

Reply #39 –
Quote from: royphil345 on
To the guy who had problems with MSVBVM50.DLL being missing... I think I read you're supposed to download that separately

Actually, I found it already on Windows XP CD-ROM

CD de-emphasis?

Reply #40 –
Quote from: royphil345 on
It worked fine right off the bat and I definitely think it produces better sounding results at the default settings than sox. The highs are a bit more open and the bass is noticeably deeper, probably just due to the gain or level of the recording remaining higher when using this program as some other posters pondered. But, it works better than sox for me without having to mess with anything else and I like it!


With that you most likely broke at least 1 TOS of Hydrogenaudio. I bet it doesn´t matter to you cause you have the download you wanted...
Nonetheless you may be interested in not fooling yourself with false conclusions in the future and give sox one more try.
Using:
"input.wav" "output.wav" --norm deemph
will create you an de-emphasised file that should be as loud as the waveemph file. The internal mathematical precision behind sox should be easily better as the one in waveemph and it uses dither.
Yu may check with a replaygained abx in foobar and report here again.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

CD de-emphasis?

Reply #41 –
Quote from: Wombat on
Quote from: royphil345 on
It worked fine right off the bat and I definitely think it produces better sounding results at the default settings than sox. The highs are a bit more open and the bass is noticeably deeper, probably just due to the gain or level of the recording remaining higher when using this program as some other posters pondered. But, it works better than sox for me without having to mess with anything else and I like it!


With that you most likely broke at least 1 TOS of Hydrogenaudio. I bet it doesn´t matter to you cause you have the download you wanted...
Nonetheless you may be interested in not fooling yourself with false conclusions in the future and give sox one more try.
Using:
"input.wav" "output.wav" --norm deemph
will create you an de-emphasised file that should be as loud as the waveemph file. The internal mathematical precision behind sox should be easily better as the one in waveemph and it uses dither.
Yu may check with a replaygained abx in foobar and report here again.


I truly doubt if stating my opinion as just my opinion violates even 1 TOS... lol

But, I have no problem with giving it another try...  I really have no idea how to use it. Last time, I copied a line of text I found in another forum to create a batch file with and simply ran the program with the audio files I wanted to de-emph in the folder. Maybe there was something wrong with my batch file? Can you tell me how to create another to try it again?

Does that "norm" you put into the code mean normalize? If so, I don't want my files normalized, just de-emphasised. If sox does that perfectly, couldn't normalization only damage the perfection? Would normalizing separate tracks in an album be a good idea?

CD de-emphasis?

Reply #42 –
Quote from: royphil345 on
I truly doubt if stating my opinion as just my opinion violates even 1 TOS... lol

You cannot weasel out of TOS #8 by saying it's your opinion, sorry.

CD de-emphasis?

Reply #43 –
Quote from: royphil345 on
But, I have no problem with giving it another try...  I really have no idea how to use it. Last time, I copied a line of text I found in another forum to create a batch file with and simply ran the program with the audio files I wanted to de-emph in the folder. Maybe there was something wrong with my batch file? Can you tell me how to create another to try it again?

Does that "norm" you put into the code mean normalize? If so, I don't want my files normalized, just de-emphasised. If sox does that perfectly, couldn't normalization only damage the perfection? Would normalizing separate tracks in an album be a good idea?


First you claim waveemph doesn´t muffle cause of being louder and now you think it may hurt!? Waveemph doesn´t do anything different as normalizing when maximising the volume.
For easy use i have added sox deemphasis to an old frontend. Just download frontah, copy the sox.exe to its folder and use De-Emphasis under Other.
You can use --norm as additional parameter.
Here is a link. Just copy the inis to the frontah folder like mentioned over there.
http://forums.slimdevices.com/showthread.p...8011#post458011

Don´t forget to specify an output folder.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

CD de-emphasis?

Reply #44 –
WaveEmph doesn't do any normalization unless you tell it to. It estimates how much of a volume change you could do without clipping, but leaves it to you to specify whether to do it, and the amount (in 0.1% increments). The default is no change. So no more speculating about WaveEmph normalizing or what SoX settings would achieve the same result.

FWIW, I can't hear any difference between files de-emphasized with WaveEmph and files de-emphasized with SoX. I tested with the first tracks on both Please and Dark Side of the Moon.

CD de-emphasis?

Reply #45 –
Quote from: mjb2006 on
WaveEmph doesn't do any normalization unless you tell it to. It estimates how much of a volume change you could do without clipping, but leaves it to you to specify whether to do it, and the amount (in 0.1% increments). The default is no change. So no more speculating about WaveEmph normalizing or what SoX settings would achieve the same result.

FWIW, I can't hear any difference between files de-emphasized with WaveEmph and files de-emphasized with SoX. I tested with the first tracks on both Please and Dark Side of the Moon.


I'm working on trying sox again and doing a comparison...

I believe the default on WaveEmph is that it keeps 100% of the same peak levels as before some of the highs were removed? This would be different than normalization because it wouldn't nomalize all the tracks on an album to the same peak level. I suspect this is the reason I liked the sound and it seems to me that would kind of duplicate what would happen if the EQ were done in analog. I really do not want my tracks normalized, but I will include sox with normalization in my comparison anyway.

Am I allowed to post short snippets of songs in a comparison if I still hear differences?

CD de-emphasis?

Reply #46 –
Quote from: royphil345 on
Am I allowed to post short snippets of songs in a comparison if I still hear differences?

Of cause, if you really did serious abx and not only think the bass is farting deeper.
After all no one really did abx the different de-emphasis methods. As in the thread i linked earlier all implementations differ a bit. I talk about tiny differences like +/- 0.1dB in frequency response (besides itunes -1dB)
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

CD de-emphasis?

Reply #48 –
Well... Here's my snippet comparison...

You guys were 100% correct in saying I shouldn't have referred to one result as sounding "better" than another. I don't have the technology to test and support that theory. I like the fact you guys discard claims like that in this forum. Very refreshing. In many forums, if a certain trusted member posts a claim like that, it soon becomes part of the gospel of the message board, regardless of if it's right or wrong or based on any actual facts and everybody on the board will then defend the idea to the death for awhile while it's in vogue.

But, I will say IMO the results are different enough a person could prefer one over another, depending on what sounds best to them personally. Maybe I'm wrong... I would also say none of the results sound "perfect" to me. I came across a post somewhere in my research where somebody said they wished they could find a de-emphasis program that produced results having certain qualities of the results from both programs. I think I would like to find that too. So, what am I looking for?... LOL!

Here are the test snippets...

http://www.4shared.com/file/IiFK0ugg/snippet_comparison.html

Personally, I'm still digging the little WaveEmph program and again, MANY thanks to mjb for hooking me up! It's too bad it seems like this little program will become extinct. I searched long and hard to try and download it and came up with nothing.
 

CD de-emphasis?

Reply #49 –
Quote from: Porcus on
Quote from: Wombat on
I talk about tiny differences like +/- 0.1dB in frequency response (besides itunes -1dB)

But, since you need >16 bits at an intermediate stage: do they all dither properly?


Convolver plugin in foobar: foobar dither
sox deemph: sox dither
waveemph: no clue, most likely no dither at all
iTunes: no clue about dither but worst response
NCL EQ VST: dither by the host program
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!