Star Sword

Agreed! I don't always comment but I read a whole bunch of the stuff that people post here and a great deal of it is positive, constructive and applicable advise. I'm really happy with this hype free place!

 

--
1 - On that note, I personally tend to add the tiniest amount of noise in my exported tracks. The biggest reason for that is so I can leave about 0.5-1 second of "silence" (very low noise) at the start. It's just a personal preference, but it helps with making any start feel less abrupt. Again, just personal preference. Do as you see fit.

Note that you can also take the opposite route. @Aaron Venture mentioned he missed a backbone, which I take to mean (let me know if I'm misinterpreting what you were saying Aaron) that he's suggesting you go the opposite direction and give a little bottom to your opening. I think Aaron and I are both getting at the same thing, just proposing different solutions. Take your pick

2 - Yes, I think you're right. I had somewhat forgotten what your goal with this piece was. It's probably best right now to stay focused on the fundamentals and see what you can do in terms of tempo and instrumentation. You can always jazz up the chords later. Since the goal is to make a piece that's easy to follow for people, it's probably best to focus on other stuff first

3 - Don't let me tell you how to use the sleighbell. Having it consistently go over the track is a perfectly valid thing to do, I think. I can't come up with direct examples, but I am pretty sure I've heard other christmass-y tracks that do that. The only remark I have is that I don't really hear the sleighbells anymore after a while. Just like I don't hear the whirring of my computer fans because it's a consistent sound. Personally, I'd sooner use sleighbells as a means to accent a particular phrase or not, but that's just my take

 

Sorry if it was unclear, I meant that the high strings are too loud and soaring and the whole texture is missing the lower end. If soaring strings are the opening statement, I'd fill out the whole section. Hence why I suggested playing the melody in lower dynamics with the harp and/or lower strings pizz (can really be anything you want) playing the chords, to help accentuate these lower notes without taking over the sound or making it overbearing.

 

Here's the thing. If your music sounds colored, all music will sound colored. Thus, by knowing the weaknesses of your setup, you can get around them (to a degree, I'll come back to it in a sec). It's how I used to roll before I got all serious and built a calibrated control room. I did have pretty good pair of headphones that I made all my final decisions on and it worked well before I went pro. And while my results weren't as good before I upgraded, I know of people who've made careers with €70 Trust speakers or even just headphones, pulling off pretty good mixes.

Your best bet is constant AB-ing between yours and other people's mixes. This is what a filled out texture that is similar to yours should sound like. Note the low end. If you can't hear the difference on your speakers, try your headphones.

Sample libraries that are sections, like strings, brass and woods should be balanced within themselves. So it's either your orchestration or you just have to bring out the lower on your Double basses, low brass and low woods. The low end sounds okay during the choir part and the final part, I'm not sure if it's because the drums are there as well. But your overall loudness doesn't change that much throughout the piece. And that's definitely mf in the high strings in the beginning. For a first melody statement at that dynamic, I don't think you should hold out on double basses. Again, just listen to Across The Stars above.

Composition-wise, on another listen, I'm not hearing much (any?) counterpoint. Counterpoint is great device that can be implemented very quickly and painlessly, and can serve as a very effective developmental tool. In your final part, even the drums are playing on the melody. Have them play a counter line.



Or something similar, whatever you like. There are a million other ways you can do counterpoint. You can have another instrument or a section play another counter line on top of this. In short, it's what goes into the holes of your melody, to quote Mike. Mike has a fantastic class on counterpoint which sums it all up very efficiently and clearly (like all his classes) in a very short amount of time, and he writes a melody in his DAW, and demonstrates various counterpoint examples and types with different instruments and orchestrations, talks in length about their uses and applications as well as effects, and of course, throws in a rant or two (aren't his rants what we all take the classes for anyway?)

 

Unless you happen to have a major peak at the bass frequencies in your opening statement leading you to believe that you have enough low end, or your speakers simply aren't reproducing it loudly enough, you should hear a pretty clear difference between your mix and Across The Stars at the linked timestamp. The idea is to just try and match it as close as you can. Either play your basses louder, bring them up in volume or boost the low end of the basses (or the section, depends) with your EQ. Your drums sound okay in the low end.

You can try transposing the entire thing up a semitone or two and listen then. Does it change?

As for the speakers, if they can't reproduce it, they can't. Headphones are much better at providing full range and flat response, but they have other downsides. But generally, if you can't afford to spend €1000 on speakers and €500-1500 on room treatment, quality headphones are your best bet and will get you very far.

For any general questions about room treatment and stuff that might help someone else, yeah, open a thread in Gear Talk and I'll post there. Any personal stuff, feel free to PM.

My room isn't perfect. It's nearly perfectly square which makes it nearly the worst type of room you can choose to monitor in, but it's the only room I've got. You can solve any room's acoustics by treating every single surface; that's very expensive and anechoic chambers aren't very pleasant spaces. Ideally you want to reduce decay (most importantly bass decay), get rid of early reflections and diffuse whatever remains of the room sound. I use Sonarworks calibration for both headphones and speakers.

This is the measurement before correction (screenshot from the Sonarworks app):



And once correction is applied (not actual measurement, just a resultant curve):



I did check in RoomEQWizard and it was very, very close to this resultant curve.

As you can see from the graph, without correction there's quite an unpleasant difference of 10db at 40-100Hz, as well as many other peaks, as well as differences between left and right speakers (which I'll attribute to minor asymmetry of the room and the differences in the absorber quality on the back of left and right walls). These bumps, especially in the low end, caused by an exorbitant amount of bass decay in the room creating nulls and peaks, can badly influence decision-making in the mixing (or production) process. Some bass notes will sound louder and you will automatically perceive them as better-sounding, and come to favor certain keys more because of this effect. It's the same for midrange, but to a lesser extent.

There's also the direct opposite - you get a mixing gig and the bass note happens to be playing right at the null point. So of course, you boost it up (and vice-versa). These are common traps everyone can easily fall into if they're not aware of them. However, if you are aware of them, you can fight them easily and learn to trust metrics (latest Ozone has Tonal Balance Control - fantastic little plug, I wish I had it back in the day). But there's always the question: how much bass is too much, and how little is too little? Comparing with other well-mixed music and knowing what notes are playing and what they sound like in your room will work - but since it's not natural and you can't just listen, but have to instead think about it, you're prone to making mistakes in this area and the whole process becomes tedious, tiring, annoying, frustrating.

That's why calibration is important - so that the only thing you have to do is listen.

As for the classes, I haven't seen Structure. I have seen Counterpoint, and it's a fantastic class. In fact, I might just write up a review for it like Paul did for Orch 3. Not sure if Mike will do a Christmas sale, but that's a whole month away, so the decision between $12 and a month of waiting is yours

 

Ahh..... Claude has yet to post a single comment. Again !! What's going on with you ... ) Just jivin ya.

 

Oh, okay. Do try.

Regarding the room, I'm not sure if you have a seriously skewed response or it's just the meter that's different, probably the latter since you have a sub. But yeah, you happen to have serious nulls in lower midrange. For the higher range, comb filtering is everywhere, you can't avoid it. I'm sure that even in an anechoic chamber you could pick it up between two speakers, unless you put the mic perfectly in the middle and perfectly align the speakers. As for the speaker upgrade, I'm not sure how much of a step up it would be. If you're gonna take the next step, you could push just a little bit more. Adam A7X has become something of an industry standard recently, beloved by many people due to its low price for what it offers. And you already have a sub, so they'll sit in just fine.

Corner traps would definitely help a lot. See, the most common mistake when one decides to start treating their room is that they go and buy acoustic foam and place it on early reflection spots on the side walls. This achieves little to nothing, and you've just spent ~$100 on 2,1m2 of acoustic treatment. Room corners are the most dangerous spots. It's where three (or two) walls, and thus modes, meet. You can slap treatment on your early reflection spots on the side, in the corners between the walls and the ceiling and on the ceiling itself, and you'll still be left with a metric fuckton of bass decay and room sound.Not only is foam vastly overpriced, it's inefficient. You've just covered 2 of your early reflection points with a material that only absorbs up to 75% of frequencies above 500 Hz (up to 85 if you fork out even more), and it comes down below 10% at ~100hz. And most of it starts falling off towards top end too. Here's a link a certified test for a foam panel, 6 of which are in a package that totals up to 2.1m2 (60x60 per piece) and costs ~$250. It looks cool, but that's about it.

$300 is what I paid for 60m2 of 5cm thick industrial Rockwool slabs, that I then cut, fit into frames, and had dressed for a total of $500, which covered 3 out of 4 corners of my room top to bottom (20cm thick) (damn door placement), entire back wall (15cm) and early spots on the walls (10cm). Generally, the thicker you go, the more bass absorption it has. But the lower you got, the harder it is, as you would need a thickness equaling at least a quarter of the wavelength to completely absorb it 100% (at 40Hz the wavelength is 7.5 meteres, so ~1.9m is what you would need, at the least) and super thick isn't something you're going to do unless you're building the place from scratch as a dedicated mixing studio. If you want to make your own, Rockwool RW3 or similar if you're in EU, Owens Corning 703 (for early) or 705 (for rear and corners) if you're in the US. Others will do as well, but this is the best bang for buck.

You can buy slightly better made (although with the same efficiency) panels from Gik Acoustics for roughly 3x that. You don't have to bother with anything other than mounting them. No mess, no need to store and transport materials, no need to own or handle tools or spend time working on it. I personally enjoy me some woodworking now and then and have all the tools and space anyway, so for me this was a better option.

For comparison, here's a test result of Gik's 242 (10cm thick, like mine on early spots). Here's one for the fat corner traps. Even one corner trap will make a noticeable difference. Mine aren't triangular, but standard rectangular. Talking about air gaps, if you decide to go for framed absorbers, do space them off the wall a bit; 70%-100% of their thickness, and they'll be more effective than if they were up against a wall - they'll have a better downwards range.

This is small summary of all the stuff I've learned over the few weeks I've spent studying acoustics online when I decided to proof my room. Hopefully someone finds something useful here.

AKG701 are just fine. I have cans in the same range (ATH-M50x) and I've got to know them pretty well and can pull off a good mix on them. I think you're good to go as it is. If you're gonna step it up, do step it up in all places - if you get better speakers, go to the next level or you're not achieving much, and at the same time improve your space. When you do the math on the final price here, it's more than what you planned on spending for a small speaker upgrade (I'll guess with A7X this would be ~$1700 with homemade absorbers compared to $800 for Eves), but the results are so significantly better that when you put moving to Eves against A7X and a proofed room, the Eve route seems to have you paying $300 to much.

So like I said, I think you're good to go for now, it's definitely possible to achieve good mixes on these headphones. For AB-ing, I would drop a piece or pieces that I want to compare to in separate tracks, send these separate tracks straight to speaker output instead of master, have them muted, and then just Solo/Un-solo them for comparison.

Whew, that's my essay for the day.

 

I play whatever is in focus, then align everything else to it. Rubato can't really be taught, or explained. I think if you gave 20 players a melody, you would hear 20 iterations of it. Everyone perceives it differently. That said, you just gotta feel it.

Developing that feeling, now, I've never given that much thought. Maybe transcribe solo works and then play them with your VIs? Play until get a feeling for when you can steal time and breathe, and of course, record all your performances (MIDI should do) and review them later. Maybe someone else can chime in here too.

I too work at 1024 samples. ~44ms lag. But I've learned to live with it. I have a relatively small total PDC while writing. But Reaper has this great function (maybe others do as well, I know Ableton didn't have it two years ago) that while a track is armed and being monitored, it will only delay the output as influenced by plugins only on that single track, so recording is easy. Altiverb is realtime, but it eats at my CPU, especially when I drop 13 instances of it onto the 13 Sample modeling instruments (doesn't work in realtime) I have 1 return track which has LXPHall on it (500 samples PDC maybe, very light on CPU) that I send strings, brass and woods to, and SM Brass additionally has VSS2 to position it and give it some semblance of ER and room sound, which definitely helps while writing. It could also be done with the Delay trick on top of ER from the SM Instrument, but for me this is easier as it's a all-in-one solution for a small price of 1024 samples. So in total the only delay I have is the one cause by buffer size in the driver.

It definitely sounds different once all processing is turned on (but I can't play it realtime if I have at least one Sample Modeling instrument playing). Other stuff not so much as it only gains a bit more room, but SM Brass definitely. I've so far learned to trust my instincts with it. My whole template is balanced and if everything else is playing MF and I want the kind of sound that you get when Trombones play MF as well, I just play it in and even though it sounds wonky and weird (I sometimes turn on the processing just to check) I know it will work out.

Later when I'm done, I just flip the switches, export it and it's ready for mixing. When setting it up, I've mocked up a few excerpts from various scores to check, and once it was in balance, I just write like I would on paper - I internalize it. Sometimes I mess up and have to go back to it, but it's getting better and better.