Remix of The Curse of the Twisted Tower

"Oh god, not another reissue..."

But before you say that, please let me explain a bit as to why and how this is a bit different than AC/DC or the Eagles doing a reissue.

In the table below are audio sample comparisons between the original release and of the remix/remastering of
Twisted Tower Dire's first full length CD The Curse of the Twisted Tower .

Even though these comparisons are obvious on typical computer speakers, I strongly suggest you try and listen to these samples on a decent sounding audio system if possible.

Links to samples of the Original CD tracks:

Links to samples of the remix and remastered versions:

An explanation of the work involved

The Beginning

After mixing Netherworlds, Marc approached me with the band's wish to re-release the out-of-print CD Curse of the Twisted Tower, TTD's first full length CD effort. I recalled hearing a bit of it, and Dave mentioned to me that it sounded as though a pillow was sitting on the speakers. The idea was to only remaster the existing mixes.

I was able to take the pillow off of the original mix a bit with just the use of some of the mastering processors I have, but I mentioned that it really needed to be a remixed to make any noticeable improvement. From what I heard and not knowing anything at all about the session (what format the multitrack was, how the individual tracks actually sounded) I was leery of really pushing for any hope that it would be a significant improvement.

When it was first released, and one can verify this in the thousands of Google links about the band, it was very well received by most metalheads. But everyone seemed to complain about the production quality; reviews of the following TTD release, Isle of Hydra, seemed to never fail to mention the improvement over the production of Curse.

Days of ADAT dumping

A bit of time passed and Marc called, mentioning that he felt that any improvement would be worth the effort of remixing, especially if he could have a local studio dump the ADATs (which I learned was the format used for the multitracking). His and the band's hope was for a marginal improvement. The studio that was doing the ADAT transfers would be doing it digitally so as to avoid unnecessary D2A-A2D conversions and had multiple ADAT's sync'd for the three (and sometimes 4) tapes used for each song.

After some time, Marc called and said he had DVD data disks of the dump from his friend's studio. But he said he must being doing something wrong, for when he imported the tracks into his DAW workstation it was playing at the wrong speed.

He brought the DVD's to my place and lo and behold, indeed they were funky. It appeared there was some sort of issue with the sample rate during the transfer. After dicking around with the tracks for a few hours it was clear that we'd need to redump the ADAT's again. The studio that did the original dump was unavailable to give it another go so we decided to borrow an ADAT from Matt, the engineer that tracked Netherworlds. He only had a single machine, so that left no choice but to run each tape separately and somehow sync each of the 3 or 4 ADAT tapes within a single session.

We had no idea what that would entail. Getting Sonar to slave to the ADAT's (again using ADAT optical for a fully digital transfer) was daunting in the least; long story short, using a Cooper Datasync II (which thank god Matt happened to also have) it became apparent that even with the best of chase locking we'd end up having to do each song separately (since the drift prevented us from doing an entire tape at once). Furthermore, the cymbal tracks spanned on to the second set of ADAT's; which led to annoying phasing of the cymbals that leaked into the tom and snare tracks (gates don't help - when the gated tom is hit, cymbals invariably also pass through the open gate).

Needless to say, it took weeks to dump all the songs, a whole lot of sitting around and waiting ensued...Then came the task of Marc editing all the takes and tracks - a major effort in itself. Thank god he's got his shit together with DAW's since I had no clue what happened during tracking way back when.

Actual mixing

When we finally were able to hear the tracks assembled, it was clear that this would also be a bit more work than originally planned. The one saving grace was that even during the ADAT dumping I noticed that the tracks themselves were not as bad as the original mix seemed to dictate. But there were some hideous issues, all typical of ADAT tracking. One glaring issue we noticed was the variance in levels, even within a song, of tracking levels. Even drum tracks seemed to change levels, sometime by 6dB or more in what was obviously sub-takes or punch-ins. Other issues also surfaced, such as super hard clips on many of the tracks.

A few screen grabs below are shown to give the reader an idea of what we had to deal with and how the end result measured in relation to the original CD release and how it measures up against a major big $$$ production..

	What we got from the ADAT's Shown on the left is a screenshot of a few of the tracks. As shown, esp. the green vox2 track, it was apparent that the ADAT contents were not the most friendly to remix. Recall these are dumped digitally, we verified the tracks thru my test gear on the analog output of the ADATs - those in vox2, were indeed clipped as shown. Notice the orange bass tracks above it, to the right of the Now cursor. Significant asymmetry, and DC offset issues made this even more difficult. Even still, with a bit of processing I was able to obtain a product significantly better than the original CD audio. And yes, those of you that know, that's a V-Vocal clip. But I must mention those were few and far between. In fact, I think less than a dozen small snippets on the entire CD were corrected like this - a lot less than other mixes I've done in the past. Much less. In fact the majority of the songs had no alteration of Tony's vocals at all.
	Typical Spectral content of a modern national release - Rob Halford To the left is the screen grab from the secondary machine I use which runs spectral 3D waterfall plots and a running spectragram. This is an example of one of the more recent metal albums, shown as a reference - this is Rob Halford's Resurrection - The One You Love to Hate - about in the middle of the chorus where it's crankin'. The Halford CD was mastered by George Marino at Sterling Sound: http://www.sterling-sound.com/epk.php?e=george&c=en Note the even distribution of spectra on both the 3D plot and the spectragram.
	The original Curse Shown to the left here is the original Curse title track, captured during and immediately after the lyric "...dwell in the shadow of the tower..." where it crunches - the exact part in the above sample MP3 file: http://www.ajawamnet.com/ajawam2/ttdsamp/scurseorig.mp3 Note the uneven dispersion of the spectra, and the lack of any activity above 8K.
	The remixed Curse Now shown on the left is the exact same part, but remixed and remastered: http://www.ajawamnet.com/ajawam2/ttdsamp/scurse.mp3 Note the improvement of the spectral content as it more closely matches the Halford sample. Also note the additional effect to have the "toweeeerrrrr" hang over the altered guitar chords. I couldn't help myself... trying for the aural equivalent of of a video pull back as Tony stands next to a large castle/tower...

With all the things I was coming across, I was beginning to think maybe Scott was right, let's just run the old mixes through some linear phase EQ's, a few multibands and a good peak limiter. But the nagging voice in my head kept going back to the fact that hey, there's some really great sounding stuff in these tracks, let's see if we can't fix the bad stuff as best as possible and give it a chance.

And as shown in the last illustration above, we were able to get a very modern sounding release that brought out the best in this sophomore effort of the band.

Glad we did all this work

The first mix was Land of Illusions, figuring it was the lead in to the CD. Marc recalled that the original hell-sounding, keeps-religious-door-to-door-people-away-from-the-house-during-mixing intro didn't exist on the ADAT's, since it was flown in during the original CD's mastering in SAW. We bantered a bit about this and even considered dropping it from the re-release.

But there's no way I was gonna give up that bible-thumping repellent.... I grabbed the original rip from Wavelab, did some stuff to it and added some delays to make it carry in to the actual song - right up to the next Tony scream, which was not on the original mix.

As I went through each song, it would start with me thinking, "what the hell is this shit?" as I listened to the bare tracks. Not so much from the music, which is killer for a first release, but more from a technical standpoint, looking for the intent of the composition and ignoring what I recalled hearing on the original CD. As I navigated around each session, looking for obvious transfer errors, it became a quagmire of "what's this supposed to be doing?" and how should it have been mixed.

Technically, as to how each part would sit in the mix, a lot of experimentation was required, some songs took well over 3 days of constant pissing around with to get to a point where the mix "popped" - anyone that's done mixing knows what I mean by that. Typically a song starts out sounding like shit and stays like that until 80% of the parts are processed, placed, and edited correctly.

That's when it "pops" and you start saying, "Hey - this ain't too bad" or in the case of this CD, "Wow - so that's what it was supposed to be..."

And everyone of them popped.... even ones that the band mentioned, "uh, that one kinda sucks - you'll have a hell of a time with it..." In fact, some of the ones I was warned about are my favorites.

But one does remain clear in my head, and was actually the opposite - The Walkyrie Death Squadrons - was one that, as I sent other mixes to the band, they mentioned as being their favorite. To this point I really never listened to it much, being a long epic. I recall Scott saying, "...can't wait 'till you get to Valk..."

The next day I started Valk.... OMG, I thought.... this sucks. What the hell is this shit? Man, I hope he doesn't expect too much. Again, anyone that's mixing blind (ie... didn't track it) usually ends up hearing it for the first time in a not so flattering mix; things that are supposed to be subtle are blaring and the main themes are usually muted or way off spatially/level wise.

I then began dissecting each section, again I always try to imagine what the composer(s) had in mind. As I broke each section down, got the levels and imaging correct, it began to emerge. It took about 35 hours over two days to get it there, but I began noticing, "OK... Pink Floyd.... OK.... Procol Harem.... ok...." and so on and so forth, keeping in mind the interviews and such I read about the band, talking to them, and their influences mentioned. It then began to emerge as I envisioned they'd like it to be remembered, sort of as an aural adventure of the female deities in mythology that determined a warrior's fate. It also became apparent after days of hearing just each part over and over again as I set processing, that Wagner was correct in forbidding Die Walkyrie to be played out of context with the rest of Der Ring des Nibelungen.

In conclusion

Even though it was to be just a re-release of the out-of-print first CD, as I went through the tracks and became familiar with the band's intent, it dawned on me that this CD should be something that even owners of the original release would want to own and cherish. With the advent of modern techniques in mixing and especially mastering, my desire (and hope) was to provide a CD that would not require the listener to aurally have to adjust to it, even when played back-to-back with the latest Metallica record. One that would not only stand strong with more modern recordings, but also provide insight into what the band had envisioned.

I hope I was able to do that.

A disclaimer - Please realize that I too am not into the loudness wars; my thought is that since most consumer systems nowadays are typical 5:1 tiny satellite "full range" with subwoofer-type systems, and as Bob Katz mentions in his great book, Mastering Audio, a threshold shift in the consumer's perception, that most releases utilize significant processing in order to get the crest factor ratio smaller than in previous decades of recording technology.

I will say that on my far-field monitors, the unmastered mix is fucking amazing... Dave and Marc stopped by during this project and I gave them a taste of what 1000 watts per channel through huge ass monitors sounded like, especially on the middle part of Curse. But again, this is on a large full range system that has the headroom to reproduce the peaks even when cranked to 11.

But most consumer systems don't have that kind of headroom. Speaker farting, and possibly amplifier or driver damage would occur on a lot of this stuff with these consumer systems. In reality, on smaller speakers to get the cranking that most metal consumers look for (and believe me I've been through wars with labels on this) and will invariably do with their volume control (which does basically even worse limiting/clipping) I can see the point in it.

Sort of...

Now I'm going back into my humble mixing abode (next to my drill press and CNC machine) and fucking crank it!!!

WAM
http://www.ajawamnet.com