Correcting HAECO-CSG Processing

Metoo

Member
Do any of you know if the 2006 CD of "Ye-Me-Le" contains the HAECO-CSG treatment as is the case with "Fool On The Hill"? I ask this because I have been able to eliminate HAECO-CSG from a copy of my CD and would like to know if applying the same treatment to "Ye-Me-Le" would result in a better-sounding CD. Or, perhaps, it was corrected on this album before publishing it on the remastered CD.

BTW, after the correction of the HAECO-CSG on "Fool On The Hill" the result is 'revealing' to say the least. The piano is more present on the right channel and some other musical details are clearer, not to mention the voices themselves.
 
I believe all CDs of YE-ME-LE also have the HAECO-CSG processing.

You must tell us how you managed to defeat the CSG processing. Hopefully it's not too complicated.

Harry
 
Hi Harry,

Well, the 'booklet' that comes with the CD says it is a HAECO-CSG album, but it seems to sound better to me as-is than applying the treatment. Perhaps they corrected it before publishing that album. I just wanted to know if someone here had this information.

As to how to correct HAECO-CSG it is not too complicated a process (at leaste not to me):
You need audio processing software that can tweak the phase of the sound waves. We are talking a program such as Adobe Audition (Check out its 'Graphic Phase Shifter' option in the stereo imagery menu).

You start by changing the bit depth of the CD-quality file to 24 or 32 bits (preferably the latter) and upsample the file to 96/24 (this, in my experience, yields the best results). Then, what you have to do is turn the waveform on the right channel +90º. You will immediately notice the difference. Then you must downsample the file back to 44.1kHz (hopefully with the use of a transparent SRC algorithm such as SoX resampler for Foobar 2000 if you are on a PC) and bring down the bit depth to 16 again so that you can burn the resulting files back to CD. If not, you can truncate the bit depth to 24 bits and play the files like that on your computer or using a hard disk-based media player or on your iPhone/iPad/iPod.

Here's a demo clip for you to hear the difference (first half is the album as-is on the CD, second half is with the HAECO-CSG treatment eliminated). Can you hear the difference (probably not the best track for this perhaps because it's an mp3, but still noticeable): http://bit.ly/iTiJpP

Hope this is of help.
 
Hi Harry,

Well, the 'booklet' that comes with the CD says it is a HAECO-CSG album, but it seems to sound better to me as-is than applying the treatment. Perhaps they corrected it before publishing that album. I just wanted to know if someone here had this information.

The 2006 Verve re-issue of YE-ME-LE on CD does indeed still possess the HAECO-CSG system. The booklet includes the mention probably only because it features a reprint of the back cover of the LP. I somehow don't think the compilers of the actual CD are even aware of the processing, let alone to actually mention it or to attempt to correct it.

As to how to correct HAECO-CSG it is not too complicated a process (at leaste not to me):
You need audio processing software that can tweak the phase of the sound waves. We are talking a program such as Adobe Audition (Check out its 'Graphic Phase Shifter' option in the stereo imagery menu).

You start by changing the bit depth of the CD-quality file to 24 or 32 bits (preferably the latter) and upsample the file to 96/24 (this, in my experience, yields the best results). Then, what you have to do is turn the waveform on the right channel +90º. You will immediately notice the difference. Then you must downsample the file back to 44.1kHz (hopefully with the use of a transparent SRC algorithm such as SoX resampler for Foobar 2000 if you are on a PC) and bring down the bit depth to 16 again so that you can burn the resulting files back to CD. If not, you can truncate the bit depth to 24 bits and play the files like that on your computer or using a hard disk-based media player or on your iPhone/iPad/iPod.

Here's a demo clip for you to hear the difference (first half is the album as-is on the CD, second half is with the HAECO-CSG treatment eliminated). Can you hear the difference (probably not the best track for this perhaps because it's an mp3, but still noticeable): http://bit.ly/iTiJpP

Hope this is of help.

All of that is a bit beyond my scope at present, though I understand the philosophy of "tilting" the phase by 90°. I listened to your sample and it does indeed sound much better. I wish I could get the whole album to sound that way.

Harry
 
In reality the only more complicated part, provided you have access to the audio editing software, is the up and downsampling and the changes in bit depth. The changing of the phase in Audition is as simple as choosing the right channel and lifting a liner to +90º.

The difference after solving the HAECO-CSG fiasco on the Verve "Fool On The Hill" 96/24 remaster is significant. I had tried it before with a plugin, but the end result did not sound so clear and the right channel had much less presence than what I got this time. This is what lets you hear the piano better on the tweaked version of the file.

I have to check "Ye-Me-Le" closer because I have not been able to identify the signs that make it clear to me that "Fool On The Hill" has been 'de-HAECO-CSG'd'. As I said, it seems to sound clearer as is, but I will try other songs because I've only checked it with the first track.
 
We are talking a program such as Adobe Audition (Check out its 'Graphic Phase Shifter' option in the stereo imagery menu).

Which version of Audition are you using? I have access to 3.0 or CS 5.5 (and I've heard the latter actually removed some features, including a phase analysis window). I can install whichever one has the Graphic Phase Shifter.
 
Hi Rudy,

I've got version 3.0, so if you can get a hold of it you'll have the Graphic Phase Shifter.
 
Great, I will grab that version and install it. Thanks!
 
OK, I have it installed and found the graphic phase shifter. I tried once and it sounded more out of phase until I adjusted things a bit more.

Which steps did you use in your sample? What I did was open up the WAV file, drag the ends of the blue line UP to 90 degrees, and chose "Left Only" for the channel. When I tried "Right Only" it sounded even more out of phase, and the bass largely disappeared.

I am listening over Grado in-ear monitors. Listening to HAECO-CSG over headphones is painful, so I felt this was a good test.

One think I am noticing, and I'm hoping this is not difficult to describe. The easy part is to say the sounds are more localized and easy to pinpoint. But when you listen to Sergio's voice before processing, the upper frequencies of his voice do NOT match where the lower range of his voice (the fundamentals) comes from in the soundstage. In other words, the high frequencies come from the middle, but the bass part of his range was coming from somewhere to the right; it's like he is "smeared" across the soundstage. After it is fixed, his vocals come clear out of dead center, spot on, and correct.

To be honest, both the HAECO and the "fixed" sound different in phase, but the more I listen to it fixed, everything just seems to settle in nicely once you've listened awhile.

Now I need to make up a list of the HAECO-ized albums I can fix. :D The Butch Cassidy soundtrack is one of the first I'd tackle--the bass on that one is totally screwy!
 
Last post, and I'm done for the night.

WHOA!!! I just tried the first track from the "Butch Cassidy" soundtrack... :bigeek: This time, choosing "right-only" just locked everything in where it should be; left-only threw it further out of phase. Maybe the phase depends on each album? (Maybe Fool is 180 out of phase compared to Butch? You probably know what I'm referring to.) This means...maybe I can actually listen to this over headphones now? (This is actually where I notice phase problems the most.)

I see a busy few days ahead. :D
 
"Metoo" mentioned YE-ME-LE as not adhering to the formula, so I'm guessing that different albums are shifted in phase differently from each other. It may take some experimenting to find the formula. You're right about Burt's BUTCH soundtrack - it's always sounded among the worst - Herb's BRASS ARE COMIN' didn't sound quite as bad.

Harry
 
What's interesting is that in the phase shifter, you can choose to shift varying amounts by frequency. They could have phase-shifted only certain parts of the frequency range. They are probably all done at 90 degrees, but if they used any kind of filter to shift only certain frequencies, anything we would try would be a crap shoot. The main thing was to create stereo recordings compatible with mono playback equipment, and bass frequencies are the ones which would cause the most problems.

I wonder if there was an actual piece of encoder hardware that went with the HAECO-CSG system. That would give a clue as to what may have been adjustable. I actually know a good person to ask.
 
I love the Baden Powell penned song Lapinha, the story about a cemetery in Rio de Janeirowhere a villain wants to be buried .

Especially wonderful singing by Garcinha Leporace and the guitar by Oscar Castro Neves.

alll the best Aqua do Brasil AKA Sjef Nix
 
I love the Baden Powell penned song Lapinha, the story about a cemetery in Rio de Janeirowhere a villain wants to be buried .

Especially wonderful singing by Garcinha Leporace and the guitar by Oscar Castro Neves.

alll the best Aqua do Brasil AKA Sjef Nix

Of course Gracinha Leporace!!!
 
Sorry to post as crappy MP3s (320kbps), but here are 12 samples of files I've processed. If requested, I can rip the equivalent tracks straight from the CDs and post here also, for comparison. (Most here probably have these CDs already.)

http://min.us/mcZYjNK
 
Hi Rudy,

Sorry for replying so late.

My anti-HAECO-CSG tweak on "Fool on the Hill" was only applied to the right channel. I have just rechecked the process and in this case you should shift the right channel to -90º. What you will notice, apart from the clearer voice location, is that the piano on the right channel will sound much clearer.

As far as my experience goes you should only work on the right channel. If, +90º does not sound right, try -90º and that should do the trick.

As you can see it's easy-peasy. Enjoy! :)

On another note, it is interesting that you have pointed out the "Butch Cassidy" soundtrack. I am going to rip that and get to work on it to see what I get.
 
Hi Harry,

The thing with "Ye-Me-Le" is that I do not notice a difference when I applied +90º to the right channel. It just seemed to sound its best as-is. I'll try it again and apply -90º to see if I hear any 'good' change.
 
Hi Rudy,
My anti-HAECO-CSG tweak on "Fool on the Hill" was only applied to the right channel. I have just rechecked the process and in this case you should shift the right channel to -90º. What you will notice, apart from the clearer voice location, is that the piano on the right channel will sound much clearer.

It would be interesting to compare both. In theory, it's the phase relationship between both channels that is off, so if you were to adjust one channel +90 or the other -90, it should come out to the same net effect, as you are putting both channels back in sync, phase-wise. Haven't tried it though, and it could be that the absolute polarity of the end result would be 180 degrees off. But, I've never been able to hear that anyway. :D

On another note, it is interesting that you have pointed out the "Butch Cassidy" soundtrack. I am going to rip that and get to work on it to see what I get.

Huge difference! You'll be pleasantly surprised. It really brings everything into focus in the soundstage.
 
I'm checking out the first tune in "Butch Cassidy" and it seems that the best setting for it in Audition is +90º. It seems that the choice between +90º and -90º is not only recording dependent, but also software dependent. I have another piece of software that works the same wonders but with the opposite settings (as in, -90º instead of the +90º in Audition).

The basic sign of having gotten it right is that the instruments and voices sound much clearer and are easier to locate in the stereo soundstage.

Just for comparison I suggest that you preview both the +90º and -90º settings with any particular recording you like.

BTW, do not forget to upsample and increase the bit depth to 24 or, better, 32-bit float before doing the tweak. In this way you will avoid getting a harsher sound in your final file. Remember to use the best SRC and dithering you can to bring it al back to 44.1/16.
 
Hey, this "Butch Cassidy" is sounding much better. Have you noticed the low-level, low-frequency thumps (perhaps a bass drum) at the beginning of the "Raindrops Keep Falling On My Head" instrumental (left channel)?
 
Here's an example file of what the correction of the HAECO-CSG system does to the music of "Butch Cassidy". The first half is the CD as-is. At the 2:46 mark the HAECO-CSG correction kicks in and you should be able to clearly hear the differences: http://min.us/mdpM454

Do you hear them?
 
Harry, which of Herb's albums used HAECO-CSG?
The Brass Are Comin'. Big improvement on that one also, although there is no processing on the beginning of the track "Sunny."
 
BTW, the five albums I did were the two Brasil '66 albums, the lone Bossa Rio album, Butch Cassidy and the TJB album. I have to check and see if the Edu Lobo Sergio Mendes Presents album is OK or not--I'm pretty sure there is no CSG on it.
 
Does anyone here know if the HAWECO-CSG fiasco found its way unto any of the Sergio Mendes and Brasil '66 Greatest Hits albums?
 
Back
Top Bottom