Haven't done the music part yet, but I figured I'd take this entry out of the way first in case I'm too tired later.
Perspective for today: first, check vocals' validity. I think they will do. Fingers crossed.
Second, a lot of tiny confusing things. I'll try to be diligent and maybe finish the song today (I think I'm being fooled by Hofstater already...). Meaning, finishing the tracking part, after which it will be so good and healthy for my psyche to work in some different song and come back later for the production part with fresh ears...
The main hurdle I see is how to make QTractor's midi output be received by Ardour as several tracks (one per Drumgizmo instrument). I've researched this thing several times and never found the way to do it, or even (yikes) if it can be done. Maybe QTractor is more in beta than I thought? I hope it's just a case of the confusing documentation. Worse case scenario, the battery goes untouched, as a one track thing. Given that it sounds so good, the loss wouldn't be terrible (although there would indeed be loss. For example, the kick gets lost in the mix during the final crescendo, bummer).
In addition to that, there's more amusing stuff, like adding sweeteners (which given my genre I've gone in the tongue-in-cheek habit of calling "bitterners"...)
The whole day I've felt a bit melancholic and confused ("clearly confused", as David Allen says). I don't feel like doing much with music, but I hope it becomes one of those cases when after the first resistance time starts to fly. If not, at least I get to check the box...
Recorded vocals today. Tomorrow I'll listen to them and if they sound authentic that's it.
I've sung really comfortably, in terms of sound. Just a little bit of delay effect in the headphones, and everything has melted together nicely.
Info I've gathered from the internets on singing raspy:
1) It's better to start with a clean voice (aaaaa) and then add raspiness gradually (ajjjjjjj). Otherwise you might strain your voice --I forgot to do this, and, being new to this style, after the third take I was gasping for my life.
2) A good starting point is thinking you're holding your breath.
3) A definition that I loved: "you use less air and more power".
The "redux" system has also contributed to make the session a short, tremendous blast. After the recording session I was in a state of "eustress", a word I once read online, ugly but very useful: it is the kind of stress you get for a good reason, when you're euphoric. Also, I've lost a lot of fluids and I never had yearned so much for a yerba mate in my life... Hmm...
In the physical arrangements I have also upped my DIY game with the pop filter, which now is detachable and simpler (for the paper I tried using a coffee filter instead of toilet paper). I could have maybe added some kind of anti-reflection on my back, but it will have to wait for upcoming iterations.
Feeling kinda chewed so I'm going to give it a rest today. I could get down to it and get some stuff done, but I think it's better to just rest and get to it tomorrow more renewed.
Finished the "lightning mode" software setup. Funny how the process is similar to what you do in audio production; you fire it up, see where it derails, fix it, go back to the beginning, fire it up again, see where it derails now... Everything is a process. I've corralled it to 3 clicks+3 keypresses between boot and being singing. Everything documented so that the improvement stays.
Another innovation today that is a keeper is what I'm going to call "redux version"; as the song has long instrumental interludes, I've created a version that skips them gracefully; that way my verses will be more consistent among them, my voice won't cool down with the waiting, and also I'll finish every take quicker. Editing was a bit boring (the song has 2 tempo changes; that's what I call hitting the road running with Drumgizmo...), but the result is well worth it.
Additionally, pops are being filtered, room reflections are being killed, and today I even got to the fun part, the simplicity that must result out of all these intricacies; I recorded a few goofy takes, experimenting with my voice's degree of raspiness, the intonation... I'm aiming for something like "you make me more bored than angry, but don't push your luck". I always thought that being a singer is a lot like being an actor...
Prep work for the vocal tracks. Last summer I instituted a concept called "lightning mode", which seems to stick. It is simply setting the software so that the singer (me but with a different hat) can get to the session, to be recording voice, in as few clicks as possible.
That way you take the software factor and its horrible stops out of the equation. After that is set, I also have a list for the hardware setup, and everything is orchestrated so that I can go in and out as quickly as possible (here I have to explain that I don't have a recording space of my own where I could leave my stuff in a stable manner...)
Preparing this lightning mode has taken more than I expected because it has forced me to advance in the computer configuration, something that I've been procrastinating as much as possible, giving preference to creating stuff and also to this nasty stuff I've recently had with the guitar tone.
Funny how you find this same dichotomy no matter what you're doing; cranking out stuff with what you already have, vs improving the machine, for the promise of better results, but while nothing gets through the door. In my Tony Stark life I have the same challenges of allocation of resources...
I will probably still have one prep session tomorrow to get the lightning mode stable + hardware setup, and then the next day I'd be recording the vocals.
Listening to the song today in a casual manner, "zooming out from the microscope", so to speak, I've suddenly surprised myself by going "holy shit" (to be more precise, the expression's equivalent in my native tongue) by how the thing rocked. The multilayered guitars quite do the trick, although the song still sounds a bit too demo-ish (unproduced, and using new programs in a new computer, what do you expect... But the song comes through, and this makes me so happy...)
Today ends a new deadline for these posts. I'll do a bit of a stretch and extend to #39, in honor of Queen's song of the same name.
Feel so lost right now. Recording guitar within Ardour also gets this "harsher than while recording" effect. And, after so many takes IT CANNOT BE ONLY PSYCHOACUSTICS. For worse my mouse just died in the middle of a session (he couldn't stand the power).
I hate this situation (when stable? when?), but I think I'll be able to patch up something decent with the different guitar layers I have recorded. Today in fact I've made the first arrangements for the vocals session (as yesterday,, too early, but I guess it's normal that I want to speed up the process).
My "goals" (another word for "wet dreams") of releasing one album in October and another in November have gone down the drain with this delay. But I still can aim to launch album #1 somewhere in November and #2 somewhere in December.
New tests with guitar tones. I estimate I have like 2-3 more days of tests until I have everything tamed. But I'm happy because I can jam along the way, so it's not all boring theory and parameters.
Looking back in perspective, today I've been a bit too confident, a bit too quick. I've tried to go back to the "one pass of drums then tracking", but it was too soon, too much work. I should have focused only in consolidating the new guitar stuff, that I just discovered yesterday. There is an energy cost to the change of activity; although there is also a cost of boredom and therefore morale effects on doing only one task for too long... A dynamic balance.
I promise myself that once I have a decent guitar tone in place, I won't shake things for at least two albums. Unless, of course, a certain song demands something special that requires innovation. But other than that, if it ain't broke don't fix it.
Today I've tried a lot of stuff, under the bracket of "fail fast", and I think I've finally found something I can work with.
I've been doing experiments regarding the guitar tone. It's difficult for me to summarize here the chaos of tests in a linear manner; the result, or at least the most plausible hypothesis as of now, is: my audio interface, a Samson G-Track, works better with buffer size=3, but according something I read somewhere, Guitarix acts funny under buffer sizes that are not a multiple of 2 (perhaps this is true for Rakarrack too?). So that's why historically all my tests with buffer size=3 did weird stuff. Today I have tried recording the guitar within Ardour, no Guitarix, just adding a distortion plugin to the track and monitoring via software, and the sound has finally become something I feel confident, not that sensation that someone or something is taking a dump on you whenever you try to play and the latency is "slimey".
So I'm very happy, as I have the sensation that I've finally gathered everything I need to start cranking out stuff seriously. To be honest I get this sensation periodically, but this time what's undeniable is that I've taken a giant step forward, with Drum mutherfucking Gizmo and this song that is starting to sound really kick ass...
The distortion in Ardour is primary and raw, so maybe it will need some more audio polishing, but I'm not going to obsess about it because in the end I'm a musician, a punk musician for worse, I just want to put it out there, especially in this age when so many hide that they have nothing to say under 300 plugins...
Today I was going to work on the drums, but in a funny, organic way, I've finally done some prep work for the vocals too.
It was so because I started my musical day by trying to find some songs in a similar style to mine, and listen consciously how they deal with the crash cymbal. This is a part of my song that has been bugging me for a long time; the song is fast and repetitive, noisy as hell, with a crash on every beat. In the session, however, this continuous crash ended up sounding very unrealistic, and sometimes generating subpatterns that become very distracting because each beat should be evenly accentuated.
I thought this issue had to be dealt with during production; compression to tame the excess of noise. But I wasn't totally clear about that, so today I listened to how other bands do it in these "noisy" scenarios.
Firstly, I had difficulties finding a song that did this stuff. Probably because I never listen to a song in terms of audio production unless I deliberately set myself to do it (and even then, I often shift unconsciously at some point to just following the song), none of the examples I chose (Black Flag, Minor Threat, Hansen Brothers) had that kind of noise I needed. So then I turned to the main influence of my song, Motörhead.
Again, no luck with an alike example, but before I noticed it my audition session had turned into simply listening to some of my favorite Motörhead songs. "Ace of Spades", "On parole" -I'd love to cover this one one day-, "Killed by death", "Sympathy for the devil" (really curious about this one that I hadn't ever listened before; a song like this didn't fit in my concept of Motörhead, but oh boy does Lemmy own it). And, in case there was any need for further proof that Mr. Kilmster was a wonderful human being, "Suicide", whose lyrics I had never read. When I noticed it I was singing aloud, which makes good practice for the upcoming vocal tracking (although I don't know if I will try to imitate Lemmy's "broken style"; my register is like one octave higher, and I don't have enough whisky in blood...)
Then I moved to investigate and fool around with Drumgizmo, in case there was something I was doing wrong, and I think I got it (I ran out of time right then). The humanizer parameter has by its side attack and release controls. I erroneously assumed that they were for some kind of compression effect on the drum sound (as I had only heard of those parameters in such context), but they affect the humanization; when Drumgizmo detects a repeated note, it applies them to the velocity, preventing the "ratata" effect. Armed with this information, I modified those parameters, and the crash cymbal has started to sound more decent.
In addition to that, I'm thinking maybe with this humanization active, the additional work I did on velocities becomes redundant, contributing also to the weird unnatural sound.
I still have this sadness+anger re: the guitar tracks. The sound is very mediocre. Why's that? With the features of this computer I should be getting a better sound. I'll try to live with it and otherwise I'll make some more tests.
I just come from listening "Now you're talking". I think I have decent guitars. It bothers me not having reached the bottom of this matter, but I've made great learnings that will help me getting rid of problems in the future, for sure (for example, I now understand the connections screen in qjackctl and can change stuff there directly, which is very handy).
What happened today was that, out of the blue, Guitarix didn't work with the settings that I've been using all these days. After revising the whole chain, it turned out that for some reason, "capture 2" was now "capture 1", and viceversa. Strange. Never had this kind of inconsistency before.
Through the intensive rehearsals of these days I took out of the equation the poor playing element, which blurred the other possible causes for the problem. I also learned more about Guitarix settings, which helped me discard the possibility of a poorly set tone. The 5.8ms latency is tolerable, but I'm dying to take distortion out of the box. The DAW should be nothing more than a glorified recorder.
The tutorials I've seen these days say completely different things. It's hard to differentiate actual information from noise, but at least they gave me ideas that then I could try through trial and error.
Anyways, I recorded the guitar track twice, hard panned one left and the other right, and after the tracking session I was able to listen to the whole thing without feeling repelled by it, which is always a good sign. In fact I was this close to overlay a few mockup vocals on top, to improve the currents ones, which are very rudimentary, and just for the fun of it. But recording vocals is always the most noisy part of the process and I have to be discrete.
Again, after the advances in the layering, I've discovered the blatantly obvious things that I have to improve in the drums next. So tomorrow it will be voice or drums, head or feet...