Go Back   Foo Fighters Postboard > Foo Fighters Discussion > General Foo Discussion
FAQ Members List Calendar Search Today's Posts Mark Forums Read

Reply
 
Thread Tools Display Modes
  #21  
Old December 31st 2014, 12:05 AM
xz38 xz38 is offline
Visiting is pretty...
 
Join Date: Mar 2011
Posts: 5
Default Re: Pocketwatch 2014... Just... Almost Late LOL

WOW!!! Just got my old account unlocked and this is the first post I see. Thank you so much for this.
Reply With Quote
  #22  
Old December 31st 2014, 01:20 AM
Dave T.'s Avatar
Dave T. Dave T. is offline
my tits are also out for Simon
 
Join Date: Mar 2001
Location: Hamilton, ON, Canada
Posts: 8,409
Default Re: Pocketwatch 2014... Just... Almost Late LOL

Bad. Ass.
__________________
I was a young boy that had big plans. Now I'm just another shitty old man. I don't have fun and I hate everything the world owes me so fuck you.
Reply With Quote
  #23  
Old December 31st 2014, 02:39 AM
abstract_t abstract_t is offline
Visiting is pretty...
 
Join Date: Dec 2014
Posts: 2
Wink Re: Pocketwatch 2014... Just... Almost Late LOL

Newbie but not to the band or the causes. I am looking for a pocket watch for an anniversary. ......yes was the only answer.
Reply With Quote
  #24  
Old December 31st 2014, 04:52 AM
FooZealand's Avatar
FooZealand FooZealand is offline
Easy For YOU To Say
 
Join Date: Sep 2010
Posts: 3,756
Default Re: Pocketwatch 2014... Just... Almost Late LOL

Quote:
Originally Posted by Simon View Post
I'm also curious, the only reference for speed we have on this is the CD version of 'Skeeter'. When they're both trimmed to start and end at the same point the CD version is 2.00.639 and this version is 1.58.153, so a 2% difference.

Oh almost forgot Marigold on In Utero Deluxe, that one is slower than this as well.
First it's certainly not impossible it is wrong. Second, nothing is final and can't be fixed so there's no panic, and most importantly really it's just tape speed and not tape quality so again it can all be redone, will just need some tech help.

I have a fair few references, but theses are from an old school guy (and old articles)
http://www.endino.com/archive/cassettes.html
http://www.endino.com/graphs/index.html

The deck I used is a Yamaha KX-530 with 3 heads (separate tape heads for playback, recording, and erase)
http://super-audio2.nazwa.pl/allegro...kx-530rs_1.jpg
http://www.fein-hifi.de/images/produ...3_3_kx530d.jpg

"Play Trim" isn't a speed thing (memory fart) it's a feature:
http://www.tapeheads.net/showthread.php?t=20817

I have two of these decks - one was used a lot in the day for playing, the other a 100 hours or so for recording. The record deck has been pulled out and checked by a guy I know and trust.

Long story short the tape speed referenced here - (1 7/8 ips) for 1/8"
http://artsites.ucsc.edu/EMS/music/e...recorders.html

has been triple checked for the sole purpose of recording the Rough Cut demo tape. The tape speed on the capstan motor was very slightly adjusted to 100% truly pull correctly. The azimuth and all frequency responses were all fine after testing against a reference tape with test tones (as mentioned in the first link).

So the tape was transported at exactly the correct speed as it should be (not further fucked with by me except for a bit of play trim which is not speed affecting).

If the end result is agreed to be fast, then in fact it was recorded fast, which is totally possible but impossible to know without other true references (what we are now discussing) or without playing the recorded tape back on the original recording deck (utterly impossible).

There's That Song (Petrol CB) on the Simple Machines Neapolitan vinyl set is maybe the only real way to be sure so I'm gonna have to hunt and find it.

Based on the others being slower (which did concern me a bit or I wouldn't have mentioned it) I'm thinking it could perhaps be a touch fast, but it's definitely not either the deck or editing doing it.
Reply With Quote
  #25  
Old December 31st 2014, 08:52 AM
FooZealand's Avatar
FooZealand FooZealand is offline
Easy For YOU To Say
 
Join Date: Sep 2010
Posts: 3,756
Default Re: Pocketwatch 2014... Just... Almost Late LOL

http://www.discogs.com/Various-Neapo...elease/1199902

http://www56.zippyshare.com/v/16045676/file.html
Needle drop flacs of There's That Song... no audio edits whatsoever. Contains two versions as below:

Vinyl All.flac
Good chunk of the first song on the side then TTS/PCB plus constant clicking at the end where my turntable does not return because it is fully manual.

Vinyl Song Only.flac
Same song just length edited to what is the actual song plus enough to prove it has finished.

Vinyl Song Only = 4.20
My 2014 = 4.40 (But really 4.35 or less as I have way too much lead on that one, not sure what happened but it's obvious)
2011 Flac = 4.44 (But really 4.40, it in turn has way too much lead out)

7 inch is 33 1/3 RMP - not going to consider trying faster, nor that my German turntable is suddenly also playing tricks on my ears - but at the very least that box set single is well out there to be timed manually by anyone that also has it. Maybe the edit is different I will listen more intently when I don't have 5 people abusing me for not drinking lol. Yeah just bugging me too much...

So right now I can not account for the big time gap against the vinyl - I'm certainly not SLOW, and as I hear it I could be a tad fast - but I think also most tapes recorded (including the 2014 flac which I think has since been the best) are at least the same amount (or more) SLOW as I might be fast.

Anyway as requested my remaster flacs. Certainly no doubt that putting aside the speed issue the content in these has a massive amount more audio data - total 414 MB vs 245. Down to the tape quality and probably also the deck.

01
http://www58.zippyshare.com/v/79198395/file.html

02
http://www76.zippyshare.com/v/56518310/file.html

03
http://www45.zippyshare.com/v/83750452/file.html

Happy New Year Fookers
Reply With Quote
  #26  
Old December 31st 2014, 02:00 PM
Simon's Avatar
Simon Simon is offline
The Sidekick
 
Join Date: Dec 2005
Location: UK
Posts: 29,001
Default Re: Pocketwatch 2014... Just... Almost Late LOL

Quote:
Originally Posted by FooZealand View Post
First it's certainly not impossible it is wrong. Second, nothing is final and can't be fixed so there's no panic, and most importantly really it's just tape speed and not tape quality so again it can all be redone, will just need some tech help.
So first of all where are my manners, thanks for sharing this Mike. It's the first tape transfer I've heard where the first part of 'Hell's Garden' isn't really rough. I can only guess that SM tape got damaged at that point pretty quickly, or some other flaw caused it pretty soon in the tape duping process. This one must be very early, which is great.

Quote:
Long story short the tape speed referenced here - (1 7/8 ips) for 1/8"
http://artsites.ucsc.edu/EMS/music/e...recorders.html
So about the speed, indeed it's nothing set in stone because it can be re-adjusted without any loss, and I agree the tapes SM made could have run or duped at the wrong speed, we know from many sources it was not exactly a pro tape duplication center, it was a few guys in a room with some tape decks.

Anyway regarding the two official sources whilst they could be wrong, I think they are the most likely to be right, especially the 2013 IU Reissue version of Marigold. That would obviously have been taken straight from the master reels professionally by Barrett. It could have gone at the wrong speed somehow but I think it's far, far less likely than cassette sources, which are far more prone to speed issues. As you say though there is no 100% correct answer so it's up to you, or I, the listener to decide. I'd personally go with the official source speed but if you or anyone else believe it to be slightly wrong, that's your opinion of course.

Quote:
Originally Posted by FooZealand View Post
http://www.discogs.com/Various-Neapo...elease/1199902

http://www56.zippyshare.com/v/16045676/file.html
Needle drop flacs of There's That Song... no audio edits whatsoever. Contains two versions as below:

Vinyl All.flac
Good chunk of the first song on the side then TTS/PCB plus constant clicking at the end where my turntable does not return because it is fully manual.

Vinyl Song Only.flac
Same song just length edited to what is the actual song plus enough to prove it has finished.

Vinyl Song Only = 4.20
My 2014 = 4.40 (But really 4.35 or less as I have way too much lead on that one, not sure what happened but it's obvious)
2011 Flac = 4.44 (But really 4.40, it in turn has way too much lead out)

7 inch is 33 1/3 RMP - not going to consider trying faster, nor that my German turntable is suddenly also playing tricks on my ears - but at the very least that box set single is well out there to be timed manually by anyone that also has it. Maybe the edit is different I will listen more intently when I don't have 5 people abusing me for not drinking lol. Yeah just bugging me too much...

So right now I can not account for the big time gap against the vinyl - I'm certainly not SLOW, and as I hear it I could be a tad fast - but I think also most tapes recorded (including the 2014 flac which I think has since been the best) are at least the same amount (or more) SLOW as I might be fast.
So I took a look at these, I lined up 1) This Vinyl 2) One of my older transfers of the tape and 3) Your transfer of the tape in this thread. They were all at different speeds, unsurprisingly. First thing I did though is adjust them all to the same (not necessarily correct) speed so that I could see where the length differences were, if any remained. Essentially, they cut the Vinyl off before they cut the tapes off, for whatever reason. Again, DIY nature I guess.

Here is an image of the three tracks layed next to each other, speed adjusted. Vinyl on top, old tape transfer middle, your tape transfer bottom.

This is how they looked originally.


So as you can see the old transfer was slowest, the Vinyl a little faster and the new transfer faster still. I adjusted to the middle one just for ease of lining up and the differences were -2.1% on top and -4.275% on the bottom. As I say though by no means is the middle one the correct speed.

Quote:
Anyway as requested my remaster flacs. Certainly no doubt that putting aside the speed issue the content in these has a massive amount more audio data - total 414 MB vs 245. Down to the tape quality and probably also the deck.
Well the reason they're bigger is simply because they're 24-Bit rather than 16-bit , whether there is any audible difference is the big question. Some tapes it can be beneficial transfer at 44/24, some not. IIRC Pierre who transferred the tape in 2011 decided there was nothing to gain on that tape above 16.

Cheers for the FLAC. So what is the total work done on these post transfer?

When I have some time I'm going to line up the various transfers and the official sources, see how different they are looking on speed and see if somehow we can come to an agreement what is 'correct'.

As for the sources overall, I think we're going to probably end up with no definitive source for all 10 tracks (discounting the official releases). One thing we have to bear in mind is that we now know these 10 tracks come from three different sessions, not two as long thought and written on the tapes. That means three different setups, two different studios and lots of other variables to take into account. Then of course we don't know how SM ended up with the compilation of those tracks, who mixed them into one, which tapes were used, how they were dubbed etc. There is a lot to consider.

More later.
__________________

--
Can't post? Can't sign in? Email me if you have a technical issue related to the board.
Reply With Quote
  #27  
Old January 3rd 2015, 04:38 AM
FooZealand's Avatar
FooZealand FooZealand is offline
Easy For YOU To Say
 
Join Date: Sep 2010
Posts: 3,756
Default Re: Pocketwatch 2014... Just... Almost Late LOL

Quote:
Originally Posted by Simon View Post
As for the sources overall, I think we're going to probably end up with no definitive source for all 10 tracks (discounting the official releases). One thing we have to bear in mind is that we now know these 10 tracks come from three different sessions, not two as long thought and written on the tapes. That means three different setups, two different studios and lots of other variables to take into account. Then of course we don't know how SM ended up with the compilation of those tracks, who mixed them into one, which tapes were used, how they were dubbed etc. There is a lot to consider.

More later.
Definitely lots to consider but thanks very much for all that. I've been away a couple of days but it bugged me the whole time lol. It's very clear (in editing) there are several different sessions even if you didn't already know. Which does make editing even harder (though I wasn't deliberately editing for consistent sound, my changes were for consistent to the entire tape 95 + %)... and the differences that do exist will be constant throughout each cassette, leaving quality and speed. ie whatever SM ended up with, however it was put together, we have whatever we can get from the best quality tape. I think I have one so now it's just how to get the best from it - and no I don't think we are there yet.

I see where the vinyl is cut now, and I'm confident my masters are fast. The only real upside is I'm also confident mine are all fast by whatever they are out, as the deck can't really be phasing on the speed. I can only suspect they were recorded fast (my copy) early on, then there was a fix for speed (or just a better play deck used) and maybe that happened before the master was "damaged" (HG) and maybe it didn't. So go buy every copy and find a good one in the middle (incredibly difficult or impossible if it does even exist) or be serious and work with what we have.

Editing - I did about 20 edit versions (plus/minus) of each song hence the 15 odd hours I alluded to. Minor adjustments all the way each time back and forth. Mostly in the top end, also in targeting the bass and guitar range. That said even the totally raw tape copy had higher bass than any other copy and the drums were much more distinct. But the balance wasnt there. But actually the balance isnt there at all - even the vinyl version sounds muddy but then I haven't actually ever even heard Barrett's IU version...

So I'm really not doing anyone major favours in guessing the edit levels either. So let's start there - fire me that IU edit as that has to be a master reference for this don't you think, because there's really zero chance his tapes play anything other than exact, nor that the sound levels aren't as meant to be at least for that session.

Then we can work on a speed difference and have another try I have to say though personally even though I agree it's a tad faster than it almost certainly should be I've been listening to it a lot and do like how it sounds. But us perfectionists

EDIT
Also reasonably confident your tape version is probably slow - it happens all the time on any deck which is why you do the check before restoring anything important. Maybe just by the rough amount my deck was adjusted by for utter tracking accuracy. But that's not helping here as the change said it should play faster when it was already playing too fast.

Last edited by FooZealand : January 3rd 2015 at 04:42 AM.
Reply With Quote
  #28  
Old January 3rd 2015, 01:43 PM
wicked_gravity's Avatar
wicked_gravity wicked_gravity is offline
Have you found your way around
 
Join Date: Jan 2005
Location: Earth
Posts: 4,566
Default Re: Pocketwatch 2014... Just... Almost Late LOL

Quote:
Originally Posted by FooZealand View Post
Definitely lots to consider ...
I'm blown away by all this. Thanks so much for your time. I downloaded but I am waiting on listening, I am hoping there is an agreement on the speed that will deemed official/correct.
__________________
-------------------------
Gonna find my way to heaven
'Cause I did my time in hell
Reply With Quote
  #29  
Old January 3rd 2015, 04:09 PM
Simon's Avatar
Simon Simon is offline
The Sidekick
 
Join Date: Dec 2005
Location: UK
Posts: 29,001
Default Re: Pocketwatch 2014... Just... Almost Late LOL

Alright so I lined up the 2013 new mix/master official source, my copy and your copy.

The new mix and my copy were extremely close in speed, but my copy was marginally slow. I sped it up by +0.7% to get it to match. Your copy needed slowing down to match the official source, but the offset (3.575%) of what I adjusted Petrol CB to match my copy was too much, so either on my source or your source or both the speed fluctuates, which is to be expected I suppose. The correct amount was -2.75%.

So where do we go from here? Well you could adjust your whole tape by -2.75% but of course with no 'correct' reference for other songs we have no idea how correct it is for those. Skeeter is the only other reference of course, so next I will check that. Fingers crossed the difference is the same or very similar.

Feel free to have a play yourself though, I've sent you the necessary files.

EDIT: Lined up Skeeter and yeah, not the same. My copy had to be speeded up again but by more this time, 1.8%, and conversely your copy needed slowing down less, only 2.25%. So, do Skeeter and Marigold on the official releases run at the same speed? I've no idea if you can work that out, but I'd guess not.

Still, that doesn't help us much. My copy runs slow by 0.7-1.8% compared to official sources, yours runs fast by 2.25-2.75%.

I suppose I can compare all of your tracks to mine and see how they compare, if there is much fluctuation.
__________________

--
Can't post? Can't sign in? Email me if you have a technical issue related to the board.

Last edited by Simon : January 3rd 2015 at 04:41 PM.
Reply With Quote
  #30  
Old January 4th 2015, 11:09 PM
FooZealand's Avatar
FooZealand FooZealand is offline
Easy For YOU To Say
 
Join Date: Sep 2010
Posts: 3,756
Default Re: Pocketwatch 2014... Just... Almost Late LOL

Quote:
Originally Posted by Simon View Post
Alright so I lined up the 2013 new mix/master official source, my copy and your copy.

The new mix and my copy were extremely close in speed, but my copy was marginally slow. I sped it up by +0.7% to get it to match. Your copy needed slowing down to match the official source, but the offset (3.575%) of what I adjusted Petrol CB to match my copy was too much, so either on my source or your source or both the speed fluctuates, which is to be expected I suppose. The correct amount was -2.75%.

So where do we go from here? Well you could adjust your whole tape by -2.75% but of course with no 'correct' reference for other songs we have no idea how correct it is for those. Skeeter is the only other reference of course, so next I will check that. Fingers crossed the difference is the same or very similar.

Feel free to have a play yourself though, I've sent you the necessary files.

EDIT: Lined up Skeeter and yeah, not the same. My copy had to be speeded up again but by more this time, 1.8%, and conversely your copy needed slowing down less, only 2.25%. So, do Skeeter and Marigold on the official releases run at the same speed? I've no idea if you can work that out, but I'd guess not.

Still, that doesn't help us much. My copy runs slow by 0.7-1.8% compared to official sources, yours runs fast by 2.25-2.75%.

I suppose I can compare all of your tracks to mine and see how they compare, if there is much fluctuation.
My next plan is a new tape master on a Nakamichi CR-7A. That will say categorically how fast this tape was recorded, ie it will give an absolute constant speed master. If there are speed variances and track compare speed variances they will all have occurred when the tape was recorded.

http://s218.photobucket.com/user/eld...307a1.jpg.html

http://www.audioreview.com/cat/other...8_1595crx.aspx

Should have that done hopefully by the end of next weekend. It will give a good compare on this tape at least.
Reply With Quote
  #31  
Old January 4th 2015, 11:48 PM
mj mj is offline
math is fun!
 
Join Date: Nov 2006
Location: NoVa
Posts: 1,239
Default Re: Pocketwatch 2014... Just... Almost Late LOL

Quote:
Originally Posted by FooZealand View Post
My next plan is a new tape master on a Nakamichi CR-7A. That will say categorically how fast this tape was recorded, ie it will give an absolute constant speed master. If there are speed variances and track compare speed variances they will all have occurred when the tape was recorded.

http://s218.photobucket.com/user/eld...307a1.jpg.html

http://www.audioreview.com/cat/other...8_1595crx.aspx

Should have that done hopefully by the end of next weekend. It will give a good compare on this tape at least.
This is all so fascinating, I love that you and Simon are such geeks about this. Thank you both for sharing your time, expertise and passion!
Reply With Quote
  #32  
Old January 5th 2015, 02:10 AM
FooZealand's Avatar
FooZealand FooZealand is offline
Easy For YOU To Say
 
Join Date: Sep 2010
Posts: 3,756
Default Re: Pocketwatch 2014... Just... Almost Late LOL

Quote:
Originally Posted by Simon View Post

EDIT: Lined up Skeeter and yeah, not the same. My copy had to be speeded up again but by more this time, 1.8%, and conversely your copy needed slowing down less, only 2.25%. So, do Skeeter and Marigold on the official releases run at the same speed? I've no idea if you can work that out, but I'd guess not.

Still, that doesn't help us much. My copy runs slow by 0.7-1.8% compared to official sources, yours runs fast by 2.25-2.75%.

I suppose I can compare all of your tracks to mine and see how they compare, if there is much fluctuation.
So wait on any more compares until I have a new version - I'm also going to try and do it as one straight flac file not two.

Meanwhile I've done some note reading and comparing.

Do we have any absolute guarantee Marigold was not tempo adjusted for 2013? I'm guessing not but I just wondered whether there had been any actual comment on what was done to it?

This is a composite of the (applicable) general notes on cassette speed that I have:
"As a cautionary note here, you can never adjust the speed correctly for the entire cassette - the tape will always run a little faster at the beginning and a little slower at the end on account of the changing tape take-up force (back tension is significantly higher at the end than at the beginning,) Also the largest variance is at the extremes of the tape ends, where supply back tension is at its extremes, typically about the first 3 minutes and last 5 minutes on a 90 min tape."

So based on the above it can be expected that when making absolute speed comparisons of even a perfectly correctly recorded and played back transfer that there could (actually should be) speed differences when compared to a confirmed third party source. Recordings get slower over the course of a cassette, with that change being more or less linear over the cassette length EXCEPT for the first and last few minutes where change is somewhat more rapid.

So what's very important with what data we do now have is where the songs are positioned on the tape. Marigold is B1, Skeeter is A5.

Marigold B1
Simon 0.7 Slow
Mike 2.75 Fast

Skeeter A5
Simon 1.8 Slow
Mike 2.25 Fast

It's a shame they aren't on the same side, but nevertheless it should be pretty safe to assume the same playback machine was used for both sides in each case. So this shows that while on different sides, in both cases you have exactly what you would expect to see - a COMPARATIVE speed decrease from the first to the fifth track. In your case it slows by 1.1 percent over five tracks, and in mine 0.5 percent.

Based on the above what can actually reasonable safely be assumed is that your source recording is slow throughout, and mine is fast throughout. And in fact quite usefully because of where the two songs lie in their respective sides the speed difference we have should actually be close to boundary differences for each recording because they are first and last tracks.

Yours 0.7 to 1.8 percent slow
Mine 2.25 to 2.75 percent fast

Mathematically each track of each recording would get slower. So actually after looking at my notes I'm starting to think we have quite a lot there to work with at least from a let's try and see perspective. But again I want to see how another pass of my tape goes on that other deck.

As an aside I have this other note in my stuff too:
"Considering tape duplication machines run at 15-20 times the normal speed, even a slight mis-adjustment on a duplicator can result in a large difference."
Reply With Quote
  #33  
Old January 5th 2015, 02:20 AM
FooZealand's Avatar
FooZealand FooZealand is offline
Easy For YOU To Say
 
Join Date: Sep 2010
Posts: 3,756
Default Re: Pocketwatch 2014... Just... Almost Late LOL

Quote:
Originally Posted by mj View Post
This is all so fascinating, I love that you and Simon are such geeks about this. Thank you both for sharing your time, expertise and passion!
I'm a software test manager - probably about the only profession I could have found where I'd be paid to be such an absolute dorky geek
Reply With Quote
  #34  
Old January 5th 2015, 11:43 AM
FooZealand's Avatar
FooZealand FooZealand is offline
Easy For YOU To Say
 
Join Date: Sep 2010
Posts: 3,756
Default Re: Pocketwatch 2014... Just... Almost Late LOL

Quote:
Originally Posted by Simon View Post
Your copy needed slowing down to match the official source, but the offset (3.575%) of what I adjusted Petrol CB to match my copy was too much, so either on my source or your source or both the speed fluctuates, which is to be expected I suppose. The correct amount was -2.75%.
Thinking more... you're going to say - well then how did I need 3.575% for PCB if the range is under 3% and I'm going to go shit that doesn't totally work either... though the maths looks right.

Anyway I've been talking some more off line. Another test is match entire utterly known multi track digital source with best possible cassette source of exactly the same material.

Et voila... Next Year Japan promo cassette versus Next Year Japan CD Single. Not only both exact same source and material but Japanese made CD/Cassette in mint cond. That will pull out all the variances for either deck about as well as anything I can think of.

So that's added now to the flac job sheet
Reply With Quote
  #35  
Old January 5th 2015, 04:11 PM
Dug's Avatar
Dug Dug is offline
Have you found your way around
 
Join Date: Apr 2011
Posts: 2,352
Default Re: Pocketwatch 2014... Just... Almost Late LOL

Granted, all of this is above my head, but if it's sped up or slowed down, shouldn't it affect the pitch? If so, can you check a known note or chord from each song with it's pitch (ie A = 440mhz) and adjust to the "correct" speed from there?
Reply With Quote
  #36  
Old January 5th 2015, 11:16 PM
Let It Die's Avatar
Let It Die Let It Die is offline
I'd rather leave than suffer this
 
Join Date: Nov 2007
Location: Charlie Sheen's basement
Posts: 518
Default Re: Pocketwatch 2014... Just... Almost Late LOL

Thanks for the flac share!
__________________
Beautiful veins and bloodshot eyes
Why'd you have to go and let it die?
Reply With Quote
  #37  
Old January 5th 2015, 11:34 PM
Chicago Chicago is offline
Have you found your way around
 
Join Date: Mar 2008
Location: duh
Posts: 2,860
Default Re: Pocketwatch 2014... Just... Almost Late LOL

Wow! I echo what MJ said
Reply With Quote
  #38  
Old January 6th 2015, 12:33 AM
joshurt3's Avatar
joshurt3 joshurt3 is offline
I'll Stick Around
 
Join Date: Dec 2005
Location: Lima - Peru
Posts: 235
Default Re: Pocketwatch 2014... Just... Almost Late LOL

thanks!!
Reply With Quote
  #39  
Old January 6th 2015, 12:52 AM
FooZealand's Avatar
FooZealand FooZealand is offline
Easy For YOU To Say
 
Join Date: Sep 2010
Posts: 3,756
Default Re: Pocketwatch 2014... Just... Almost Late LOL

Quote:
Originally Posted by Dug View Post
Granted, all of this is above my head, but if it's sped up or slowed down, shouldn't it affect the pitch? If so, can you check a known note or chord from each song with it's pitch (ie A = 440mhz) and adjust to the "correct" speed from there?
Yes it affects the pitch, but it's such a small amount out it's very hard to tell without going to this sort of level...

http://janburke.de/index.php/music/1...te-tape-speed-

Tape versus CD matching is the way to go next but since I'm going to get access to a really high end cassette deck there's no point right now. If that deck says the recording is just as fast then it either really is, or my cassette was played back too fast on the duplication side when it was made (or poss recorded too slow which has the same end effect).
Reply With Quote
  #40  
Old January 7th 2015, 10:37 AM
FooZealand's Avatar
FooZealand FooZealand is offline
Easy For YOU To Say
 
Join Date: Sep 2010
Posts: 3,756
Default Re: Pocketwatch 2014... Just... Almost Late LOL

This was going to be just for Simon but I'm sharing so everyone can see what I've done - and even participate in a test if you feel so inclined. You will need Audacity (100% free and stable) if you don"t have it:
http://audacity.sourceforge.net/download/

And these 3 sources files (160ish each):
http://www43.zippyshare.com/v/14460788/file.html
http://www8.zippyshare.com/v/88033347/file.html
http://www75.zippyshare.com/v/11317277/file.html

When you unzip them just note you will get the same EDIT FILES folder 3 times - in the end all the files in each zip should be in one EDIT FILES folder (with 13 files total), plus one MASTER FILES folder (2 files) plus one NFO file.

How YOU test:
Just do steps 4F) to 4J) with the different edit files and the same 4F) file and say which one is closest.

Note - this is all JUST about matching to the version with the closest speed - any sound editing comes next - though it's already pretty good for a tape. But open the ones you like to hear them individually as well of course.
.................

The NAK deck was delivered last night on loan. I got help setting it up. Ran a copy and checked the levels against the 2013 IU Marigold. Adjusted and ran again. Just matching levels and ignoring speed.

Second master was very close and advice was work it don't fuck with it - lots yet to do. Good advice in raw matching on PC it's freaking close.

What was done:

1) Cut a direct section from the one file complete tape master flac and created one Marigold tape master flac

2) Opened your (Simon's) DaveGrohl1991-02-16t01.flac file

3) Cut both files in Audacity so that they started and ended at the same points within as close as was possible to tell digitally and audibly (several edits to get there - including just the end result). Obv also totally ignoring any duration. No edits whatsoever other than trimming the orig files. 2013 IU obv had leadin not on PW etc. Named those the master files (in MASTER FILES folder):

06 Marigold - 02A Master Raw Cut TAPE MASTER.flac
LENGTH 3.05:635

06 DaveGrohl 1991-02-16t01 Marigold (Demo) In Utero - 20th Anniversary Super Deluxe 2013 Cut MASTER.flac
LENGTH 3.08:688

4) The hard part: Can the tape master Marigold possibly ever be made to be a matched length?

Attack Method:

4A) NO edits EVER in ANY WAY to:
06 DaveGrohl 1991-02-16t01 Marigold (Demo) In Utero - 20th Anniversary Super Deluxe 2013 Cut MASTER.flac

This file is the source of truth - you can not fuck with it. How "Everlong" it takes to line up, this should/will ALWAYS match Simon's original source flac (not included though I can add) exactly or I have fucked up.

4B) Load 06 Marigold - 02A Master Raw Cut TAPE MASTER.flac into Audacity

4C) Select Effect Menu > Change Tempo (Without Changing Pitch)

4D) Test a change that calculates to get the right difference in length. (I started with -1.550). Click OK, let it make the change. Click File > Export. Save the new file.

4E) SHUT down Audacity completely and reopen. (This is the biggest trick people don't know about as almost all the bugs that stop this program being close to perfect lie in the way it caches data when multiple files are open - files open can "sometimes" effect results on other files etc.)

4F) Load 06 DaveGrohl 1991-02-16t01 Marigold (Demo) In Utero - 20th Anniversary Super Deluxe 2013 Cut MASTER.flac

4G) In the SAME window load the edit file you just saved. It will load underneath the first file.

4H) Select any part of the TOP file by clicking on it somewhere in the top pane.

4I) Press the HOME keyboard key to get to the very start of the file.

4J) Press the green play button in the Audacity. BOTH files will now start playing simultaneously.

4K) Sadly there is no way the masters match all the way to the end so:

5) Start again at step 4B). Change the value you used in 4D). Repeat the rest to 4J) until you have the best match.

In theory it's that simple. In practice I had to cut the 2013 Master a couple of times as well and repeat everything. Note that's just CUT never change. The one that was closest is the file included).

All the final match results are in the EDIT FILES folder.

Load the 2013 master and each other file in turn and play together to see how things progressed. The BEST ROUGH RANGE file told me I was close. I then went to a bunch of really fine changes. In the end there are three BEST FINE RANGE files with settings -1693, -16935 and -1694. C1-C3 - C for Candidate.

It's then "just" a question of other ears saying which one feels the best. If you load them all you'd be very very hard pressed to believe they would ever even get close - but those final three are almost perfect for half the song or even more. And really that's more than enough to prove the tape master is as spot on as I think a Simple Machine one could be. Shame about the speed issue, but the sound even directly raw and whatever generation certainly doesn't destroy the 2013 when overdubbed - they sound pretty damned complimentary to me.

FWIW C1 was my pick - but I've just lost all perspective after so many listens. None match - its a futile exercise, but considering what you get if you load the two masters, it's a hell of a lot closer than I ever expected.
Reply With Quote
Reply


Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is Off
HTML code is Off
Forum Jump


All times are GMT. The time now is 11:25 AM.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.