DIGITAL The Alan Parsons Project - The Turn of a Friendly Card (2023 Blu-Ray 5.1 mix)

[JulesRules]

Listened a bit after applying the EQ profiles.
It's definitely better. But, for instance, at 1.47 in "Nothing Left to Lose"... what was AP thinking to have our ears pierced by that synth sound? I can't believe my ears... it's not like that in 2004 Classic Records HDAD's version.

Just as an aside, there's no synth on that track I think, I guess you mean the accordion?

 

[gorman]

Having said all that, however, there is one problem with this remaster - the left-right balance skewed slightly to one side (sorry, can't remember which way now, but it's obvious in the waveforms) for some reason, by somewhere between 1.5 and 2dB. Not enough to be catastrophic, but it pushes the phantom center (lead vocals in particular) slightly to one side of 12 o'clock. So if you have this, it's worth fixing as it really improves the phantom center.

Also, many thanks for pointing this out. I never realized it. Better, I did realize it but since a long time passes between listening to the album, I was led to believe that I had a momentary problem with my ear, due to a cold, etc.

The left channel is quieter. Not always by the same amount. The last track of the album is in balance, from my experimentations... the other nine need adjusting.

 

[gorman]

Just as an aside, there's no synth on that track I think, I guess you mean the accordion?

I honestly cannot tell what the instrument is. You can 100% be right and it's an accordion. But the piercing notes, in those 15-20 seconds, all come from that instrument.

 

[gorman]

I just listened to that section on my 192/24 stereo version (from the HDAD DVD, which was the basis for these adjustments) and then on my remastered 5.1 and aside from the fact that the 5.1 is a bit more compressed than the stereo, they sound almost identical to my ears in terms of tonality.

It's hard for me to diagnose because I don't use Audacity and I don't know exactly how you've converted my EQ curves into Audacity EQ curves, but you may want to re-check your calculations if it sounds that objectionable. The only other thing I can think of is that if you haven't reduced your preamp/input volume prior to the EQ, then your adjustments are pushing the peaks above 0dB and causing clipping. As you can see from my list of adjustments I had to reduce the volume of Nothing Left to Lose (Track 9) by 3.9dB before feeding it to the EQ plugin to avoid clipping.

I dunno if anyone else has undertaken these adjustments (@MrSmithers mentioned to me that he was going to try, but I'm not sure if he did or not) but maybe someone more well versed with Audacity could chime in.

Hi! This album is very dear to me too, as such I'm spending an inordinate amount of time trying to understand this multichannel edition.

I think that, on top of different equalization choices, there is a problem of speed (and consequently of pitch). Obviously it's impossible for us to *know* what the right speed might be. It could well be that this last edition is the most correct one...

The HDAD DVD edition has a total duration of 41:01:554, the multichannel mix has a total duration of 40:11:413. Now, some of this difference may come from longer pauses between one track and the other, but not to the extent of 50 seconds.
I took some time to try and analyze Nothing Left to Lose and the multichannel version (taking into account the half second longer segue from the previous track, compared to the 192/24), has to be sped up by 1.257% to get to the same lenght and pitch as the 192/24 version.

For extra comparison, the old 1987 Arista CD that I originally bought back then lasts 40:30:707.

Quick grab to compare the different durations. From top to bottom: HDAD DVD, 5.1 Mix, original Arista CD. I don't know if this might negate the need for ReEQing but, for sure, sped up playback raises the pitch.
Interested in your thoughts on the matter.

 

[MrSmithers]

I dunno if anyone else has undertaken these adjustments (@MrSmithers mentioned to me that he was going to try, but I'm not sure if he did or not) but maybe someone more well versed with Audacity could chime in.

Yes sorry Dave for not getting back with the results. I did the remix using the CurveEQ plugin on Logic Pro and cleaning things up with Izotope RX.

The results were remarkable. There was extra clarity, smoothness and room to breathe in the sonics that’s totally missing in the overly ‘hot’ original master. It sounded like a brand new recording. The only thing that I encountered was some of the levels with certain channels were lower or higher on different songs. It was my intention to fine tune this by going back and doing it again - seeing where I might have gone wrong. But due to just setting it aside for another day for too long I recently decided to just go rogue and adjust all the channels by sight so they matched up. Hence my lack of reporting. But I’ll post some pics of my rogue waveforms when I get a mo…

It makes such a difference hearing this new master. I like that expression of “you can’t un-cook burnt toast, but you can put butter on it…”. It totally works here! Would love it if you can work some more of your magic on the other AP mixes that are too hot…

 

[Albert Dagger]

The HDAD is not the best version to start with. It runs slower than all other releases and is a tad on the warm side. I think better results can be achieved with the 2015 DSD Download. It’s now sold under a different name, if I remember correctly via Qobuz. Still the best option to date in my view. And I’ve compared them all extensively!

 

[gorman]

It runs slower than all other releases

Thanks for confirming this. I've already spent a couple of hours testing stuff...


The DSD version lasts 40:43:360, though. Which, although faster than the DVD-A 192/24 version, is still 15 seconds slower than this new 2023 multichannel mix. And almost 23 seconds slower than the 2023 stereo mix. What a mess...

 

[ar surround]

Again, thanks to all of you for the hard work you've done correcting the EQ on this title. I will try to take a shot at it...one day (LOL.) @J. PUPSTER and I spent an inordinate amount of time early on trying to fix Games People Play using EQ adjustments and DeMix Pro. That effort kind of wore me down.

 

[gorman]

Hmmm... ok, after spending a good part of today experimenting I conclude that there is no way to simply apply an EQ curve to the whole song to fix all that is wrong with the master. I agree that the DSD version sounds probably the best. EQ matching it to the multichannel release ends up producing a muddled sound (I take the average of the whole song, I don't know how else to proceed with Voxengo tool, if you only pick a portion, the results are of course skewed) in some portion of songs...

I'll probably end up with applying a bass shelf raise (still have to decide how much but around 3dB should be okish) and maybe intervene on specific songs (such as Nothing Left to Lose, the accordion note between 1.46 and 1.47 is the main candidate).

It's an interesting endeavour but one we're left doing without everything needed. Because I'm pretty sure EQ is often applied differently to different portions of a song, when mixing (correct me if I'm wrong of course), and at that point it all starts flying well over my head.

 

[JulesRules]

I've never quite figured out what the deal was with the speed differences. Once I read that the 2007 remaster was "corrected", then it said that all versions prior to the 2015 ran at the wrong speed... Quite confusing.

 

[bookofsaturdays]

Hi! This album is very dear to me too, as such I'm spending an inordinate amount of time trying to understand this multichannel edition.

I think that, on top of different equalization choices, there is a problem of speed (and consequently of pitch). Obviously it's impossible for us to *know* what the right speed might be. It could well be that this last edition is the most correct one...

The HDAD DVD edition has a total duration of 41:01:554, the multichannel mix has a total duration of 40:11:413. Now, some of this difference may come from longer pauses between one track and the other, but not to the extent of 50 seconds.
I took some time to try and analyze Nothing Left to Lose and the multichannel version (taking into account the half second longer segue from the previous track, compared to the 192/24), has to be sped up by 1.257% to get to the same lenght and pitch as the 192/24 version.

For extra comparison, the old 1987 Arista CD that I originally bought back then lasts 40:30:707.

Quick grab to compare the different durations. From top to bottom: HDAD DVD, 5.1 Mix, original Arista CD. I don't know if this might negate the need for ReEQing but, for sure, sped up playback raises the pitch.
Interested in your thoughts on the matter.

View attachment 111887

I actually found this out just a few days ago, I’m glad I get to share what I now know!

Apparently, all previous CD releases prior to the 2015 Deluxe Edition are at the incorrect speed

 

[SeeMoreDigital]

Apparently, all previous CD releases prior to the 2015 Deluxe Edition are at the incorrect speed

I'm glad I never got around to buying this on CD now... I still have it on vinyl somewhere, stored within one of several blue plastic boxes!

 

[gorman]

I actually found this out just a few days ago, I’m glad I get to share what I now know!

Apparently, all previous CD releases prior to the 2015 Deluxe Edition are at the incorrect speed

Thanks for confirming. At least that's a firm point we have in our quest. The DVD-Audio 24/192 is no go for comparison. That is, unless one likes it slightly slowed down. And I'm not saying this ironically. At the end of the day it's music and one has to like it as one has to like it.

 

[Albert Dagger]

I actually found this out just a few days ago, I’m glad I get to share what I now know!

Apparently, all previous CD releases prior to the 2015 Deluxe Edition are at the incorrect speed

View attachment 111901

Both the 2015 Deluxe Edition and the DSD download run pretty much at the same speed. Also the recent stereo remaster, but I don’t like the tonality.

 

[gorman]

(With the clipped peaks restored, and probably a lot like the track looked before it was originally mastered, with the louder parts allowed to 'breathe' properly.)

I wanted to inquire on how you go about restoring clipped peaks. Because I was looking at the waveforms, zooming in, and there are no samples actually clipped. It's more about the overall "shape" of the waveform that gives the idea it's clipped. But maybe that's what compression is all about. So I'm curious about this reconstruction process. You didn't go in any kind of detail about it.

So far, I've settled with some spectral editing on Nothing Left to Lose while maintaining the rest of your suggested EQ. In the small section between 1.32.500 and 1.57.500 I lowered by about 2.8dB frequencies centered around 2,800Hz with an octave of extension. It kind of "tames" the accordion, making it sound more similar to what one hears on the DSD version.
I understand this is probably akin to blasphemy to some... but those high pitched notes really were terrible to my ears. And the DSD version does not exhibit them.

 

[steelydave]

While this remains valid in any case and above all else, I notice in your 9.csv file the following:

1847.34,0.31
2344.20,1.56
2974.70,0.82
3774.78,1.97
4790.05,0.94

It's not very heavy EQing, as far as values are concerned, but it's probably emphasizing what I'm hearing there, considering the frequencies involved.

Yes that boost is minimal to the point of negligible, really what that curve is doing is subtracting tons of upper bass/low midrange (the stuff between 150Hz and 1kHz) bloat, and taming the egregious ski-ramp boost on the treble (5kHz through 20kHz) so it's really just revealing what's already there rather than emphasizing it. It could just be that you don't like the sound of an accordion which I can't blame you for, but there are much bigger sonic offenders on this album, even in my fixed version the hi-hats on Games People Play, for example, are still a bit sizzly.

I wanted to inquire on how you go about restoring clipped peaks. Because I was looking at the waveforms, zooming in, and there are no samples actually clipped. It's more about the overall "shape" of the waveform that gives the idea it's clipped. But maybe that's what compression is all about. So I'm curious about this reconstruction process. You didn't go in any kind of detail about it.

So far, I've settled with some spectral editing on Nothing Left to Lose while maintaining the rest of your suggested EQ. In the small section between 1.32.500 and 1.57.500 I lowered by about 2.8dB frequencies centered around 2,800Hz with an octave of extension. It kind of "tames" the accordion, making it sound more similar to what one hears on the DSD version.
I understand this is probably akin to blasphemy to some... but those high pitched notes really were terrible to my ears. And the DSD version does not exhibit them.

I used iZotope RX10 - the peaks aren't clipped in the strict sense of the word where looking at them visually you'd see flat tops, because that would produce square-wave distortion. The compressor (or limiter) rounds them off, turning peaks that originally looked like mountains into something resembling more of a speed bump.

As for what the best version is, I didn't buy (and haven't listened to) the DSD download, but given that Parsons seemingly recently-acquired love for terrible EQ choices is the whole reason I had to undertake this endeavour in the first place, I'd probably prefer a 25 years ago mastering over a 9 years ago mastering. Of all the ones I listened to, the HDAD DVD was the only one that had a nice full sound in the bottom end, and while Parsons himself might prefer this mix to sound thin and wimpy in that department, I did this whole thing to please myself, and not him.

 

[Albert Dagger]

Yes that boost is minimal to the point of negligible, really what that curve is doing is subtracting tons of upper bass/low midrange (the stuff between 150Hz and 1kHz) bloat, and taming the egregious ski-ramp boost on the treble (5kHz through 20kHz) so it's really just revealing what's already there rather than emphasizing it. It could just be that you don't like the sound of an accordion which I can't blame you for, but there are much bigger sonic offenders on this album, even in my fixed version the hi-hats on Games People Play, for example, are still a bit sizzly.



I used iZotope RX10 - the peaks aren't clipped in the strict sense of the word where looking at them visually you'd see flat tops, because that would produce square-wave distortion. The compressor (or limiter) rounds them off, turning peaks that originally looked like mountains into something resembling more of a speed bump.

As for what the best version is, I didn't buy (and haven't listened to) the DSD download, but given that Parsons seemingly recently-acquired love for terrible EQ choices is the whole reason I had to undertake this endeavour in the first place, I'd probably prefer a 25 years ago mastering over a 9 years ago mastering. Of all the ones I listened to, the HDAD DVD was the only one that had a nice full sound in the bottom end, and while Parsons himself might prefer this mix to sound thin and wimpy in that department, I did this whole thing to please myself, and not him.

The DSD download is now available for download as a PCM album from Qobuz, and I think the hi-res version on tidal is also the same. I did comparisons when they were released. It’s so much better than the HDAD DVD because of the speed issue.
Also the 2015 Deluxe is good, but has some minor compression applied. Still a decent mastering that is available for cheap!

Don’t know why the recent blu-ray is so much brighter, both the stereo and surround streams. This wasn’t the case for Eye in the Sky, and also Pyramid in Atmos sounds great.