timaeus222

If you have the newest FL already, you can just right-click an instrument channel, click Insert, and find Layer alphabetically. It doesn't noticeably save CPU, but it does make the size of your project easier to manage. That's the main thing. Serum does take up a lot of CPU (as well as about 200 MB RAM per instance); that's just how its design worked out. Haven't seen any updates about it. I don't use a bunch of instances of Serum because of that (maybe 3 or 4 instances max). I do use u-he Zebra2, though, which is much more efficient with CPU and RAM consumption, and is arguably as good or even better. I used it almost exclusively on the Rollerball remix I showed you earlier. If you start to wonder whether a song's section is too messy, just try muting each layer and seeing which one doesn't really make a difference. It's possible to get lost in the number of layers you add when you have a lot.

I didn't mind the overall loudness, actually. The main thing I'd say is that the quieter timestamps could have been a bit louder, but not the overall track. Some of the attacks (brass in particular) were a bit slow for the sequencing, so perhaps layering in some short layers (like staccato or spiccato) could have helped make up for it. Other than that, this is a pretty enjoyable, elegant and film-credits-like remix.

Drums1: The kick has a pretty good ~4000Hz click, but it could use more sub-80 Hz for a low-end "thump". Doesn't need to be metal-level, but right now it's just missing the anchor that holds the beat down. The snare decay is a bit too short. If you can give it a longer decay, I think that would help make the 230 Hz-range stand out more. I'm guessing 0:53 with the bitcrushed panned hits are supposed to be toms. They're close, but maybe dial it back (upwards) a notch on the bit depth for the floor and low toms; sounds decay more quickly at lower bit depth (at lower bit depths they shorten more), and the lower toms' tails could be a bit longer. The hi hats could use a tad less upper treble, as they sound metallic but a bit too "hard", if that makes sense. Though the hi hats may just be poorly compressed by soundcloud's 128 kbps, as they lie above 12000 Hz. The rest: The rest are pretty similar in what they lack, so basically the above covers what I would say.

Fun stuff! I love it when I end up having 10+ layers going on at once. By the way, one tip to make it easier to manage is to coordinate the patches (that is, transpose the output inside the synth) in a multi-layer sequence so that you can paste it all into one Layer instance. You have several instances where you put the same notes onto, say, three instruments, when with the Layer tool, you can turn that into a one channel deal that layers all three (or whatever number) instruments. Here's an example where I'm layering snares: For tonal sounds, all you would really have to do is to make sure the same MIDI note could play multiple patches on their intended octaves, which you could adjust within the VSTs themselves. The only drawback is that the velocities will all be the same across the instruments. Usually though, it saves time by letting you write only one set of notes instead of multiple sets. If you really think you need different velocities on each instrument for a particular moment, it makes more sense then to not use the Layer tool.

No worries, I just mean in general, as a hypothetical. It's good you enjoyed making them. And thanks with regards to the remix.

Well, one crit that doesn't seem to be addressed is that the rhythm guitars have a thin feel to them. I can hear a bass, but the guitars are completely missing low-midrange EQ (130~380 Hz, specifically, has zero guitar content), and they are overboosted in the midrange as well (~700 and ~2100 Hz, by about 2~3 dB). The bass also has almost nothing above 200 Hz, so I guess you low-shelved or scooped it that much? As a result, the track feels hollow in the low-midrange. At 0:53, you go out of the key, not in the way jazz does, but in a way that sounds like a mistake. You're using basically a whole tone scale (you play E, G#[higher], F#, D, C, E, D, G#[lower] at 0:53). I dunno, I guess it's maybe because you were playing in F# minor (F#, G#, A, B, C#, D, E, F#) at 0:39 - 0:53, which somehow doesn't jive well with the whole tone scale you used at 0:53. B and C#, though in F# minor, aren't in a whole tone scale that starts on F# (F#, G#, A#, C, D, E, F#), so if you play a chord that uses notes in F# minor, it would usually sound dissonant (especially after some distortion is added to make each note less tonal). https://app.box.com/s/ktj8z3k68qoxessbwhgx6yedsrctmym6

More often than not, making a buildup fuller increases the anticipation for the drop. So maybe, you might have some prolonged snare rolls, plus a riser, plus the melody during the buildup, and then perhaps a short fill right before the drop starts, if applicable. A recent example of mine is this: You can tell that 0:33 - 0:48 is a buildup, and that 2:36 - 2:52 is a bit of a larger buildup (due to the more intense sequencing at 2:48 - 2:52). Using the same Rollerball example above: Even though the sound palette is perceivable as consistent throughout the track, it's not so similar that you can't tell where you are in the track at a given timestamp. 0:33 - 0:48 is a buildup, due to the prolonged snare rolls, the riser, and the fuller soundscape at 0:40 - 0:48 adding more hype. 2:36 - 2:52 changes it up at 2:48 - 2:52 and adds more subdivided sequencing for even more hype than the first instance of the buildup. The repeat of 0:48 - 1:19 at 2:52 - 3:23 is noticeably different in the rhythm and melodic contour (if you have to listen back a few times to notice it, that's OK). Even though the sound palette is similar, the composition is varied. There's a low-energy chill solo section at 1:19 - 1:50, with softer half-time drums, greater focus on the e. piano than before, and a more ambient/atmospheric feel. There's a lower-energy breakdown section at 1:50 - 2:21, with no drums except cymbals, greater focus on slow-attack pads, and overall a "slower", even more atmospheric (and cosmic) feel. There's a low-energy lead-in (2:21 - 2:36), with very soft e. piano and bell arpeggios, as well as soft 4-on-the-floor high-passed drums; it leads back to the buildup variation at 2:36 - 2:52. In terms of energy, it kind of went Low (Intro), Medium (Buildup), High (Main), Low (chill solo), Lowest (Breakdown), Low (leading out of breakdown), Medium (Buildup), High (Main v2), Low (Outtro pt 1), Lowest (Outtro pt 2, without drums). If you draw it out, it's basically 1.5 sine waves. Take the time to process this. I think this demonstrates many of the things that I expect there to be in a dynamic EDM track (clear sense of intro, buildup, drop, breakdown section [optional], climax, and outtro; diverse-enough textures). Try to pick out the individual sounds that you can manage, and try to understand their role in the section. I think I mentioned this before, but writing something that you would like is probably more productive in your musical output. If you keep writing what your audience likes, then if you don't like what you're writing, I don't see why you'd be satisfied with that. Audience tastes can be very subjective, so if you keep catering to them (as selfish as this sounds), you might actually be less productive in your musical output. This is just one man's take on this, but that's what I think.

For "Sunshine", the same sidechained saws were repeated quite a bit, so it got tiring to listen to. The buildup at 0:40 didn't really connect with 0:44. I expected 0:44 to be more interesting, but it's the same textures as before plus a violin and some padding. It just doesn't feel as big as it suggests it should be. I think it was a good starting point, but it still sounds like a work in progress. Not really electro house, by the way. It sounds like house or trance. For "Daylight", 0:44 was a pretty good dropoff. It could have used a transition, though. 1:15 needed a clearer transition, one that started at around 1:12, because that was actually a good climax. This is closer to electro house, but it's mostly house. 1:45 needed another transition sound, like a snare roll or something. The ending cut off prematurely. For "Shade", immediately there's a trance-style sidechain. I like the dropoff around 0:40. This is effective. At 1:11, the saw waves confirm to me that it's trance. It's good, but it's not electro house like you had said. The climax is more like electro house, but it took a while to qualify as that. 2:36 - 2:39 gets a little cluttered in the bass. I don't think the vocal clips really added much, but they were OK.

Kinda wish this was still going; I finished my track.

Man, the sound design just nails an unnerving feel perfectly. Sure, it's dissonant, but it's done tastefully. Pushes the boundary of pleasant-enough sound design (in terms of abrasive timbre), but for me it's just about right. Today, I'd probably define music as "the logical sequence and combination of tonal and atonal sounds to generate a harmonic experience." Yeah, this counts. I honestly found this to fit more as a track in an OST than as a remix, since you generally have more freedom in an OST, but this can be some fun listening in the dark.

Well, if you move non-muted notes around in the piano roll, when you don't highlight them, they're going to make sound anyways. An additional thing you could do is Ctrl+Left-click-drag and highlight the notes you want to move, and that will make the note movement silent. So, if you're looking to work in the same project file and have trouble with lag, try that while you haven't yet turned off the convolution reverb. And indeed, Kontakt's convolution reverb can get quite CPU-intensive; I would agree that you could opt to use external reverb instead.

Minor EQ things: There seem to be two slightly conflicting "leads" at 0:51 - 1:24. I know the funk guitar (with the phaser) isn't supposed to be a lead, but it sounds like it because it has a fair bit of midrange. It contributes to the overall uneven (overboosted) EQ of the mix as a whole. If you do a wide dip on the EQ for the funk guitar in the midrange a bit (about 1~3 dB, at about 400 ± 50 Hz), that should help make it feel more like accompaniment than a conflicting lead. Could just be soundcloud's poor compression for the upper-treble at 128 kbps, but the treble sounds overboosted to me, specifically in the drum loop, and maybe a tad in the funk guitar due to the bitcrushing. This isn't a big deal, but I thought I'd mention it anyway. Repetition (major concern): Speaking of the drum loop, it's on autopilot, and it really could use more variation. The same 9-second drum loop gets stale pretty quickly, and I can tell that (or, it feels like) it's copy-pasted all the way through with no variation because the kick drum follows the same pattern each time. It would really help to add drum fills and other drum variations to differentiate between sections, to keep the whole mix from sounding too same-y and to introduce dynamics to the overall arrangement. This is actually a fairly significant issue, because it makes the pacing plodding and invites people to skip through the mix to see if the soundscape changes. Random loud instances (semi-major concern): The reverse at 2:12 - 2:14 and similar repeat places is too loud; it could be about 2~4 dB quieter if you want it to be a transition sound. At this loudness, it's distracting IMO. Not a huge deal, but an easy fix. This does add to the repetition, because you do use it multiple times. Miscellaneous (semi-minor concern): As far as I'm concerned, a sufficient ending was at 3:21. The ending at 3:22 doesn't make any sense. It sounds accidental, even though it wasn't. Honestly, I don't think it fits at all. It's not going to make me troll the mix all over the interwebs, but I don't see any justifiable reason why it needs to be there. If not having there "makes it boring as hell", then having it there doesn't do it much good either. ----- Overall, it's an enjoyable groove, but the copy-pasted drum loop and the lack of dynamic variation would be significant issues to address. Even a funk mix that depends on groove doesn't have to have a bunch of copy-pasted measures. You could add a breakdown section with a lower energy, or maybe even add a short jazz solo. Anything to intersperse a different energy amidst a continuous groove. One way you could check for repetition is to skip 15 seconds at a time through your mix with your screen off (such as with WinAmp and the D-pad). If you can honestly tell where you are in your mix (intro, 'verse', breakdown section, climax, outtro), then you have enough variation and dynamics. If not, keep working on it. This is an example of a funk remix with sufficient variation, while being a decent match to your style: https://www.youtube.com/watch?v=fsaBsI0_viE Or, this one is simply a funk remix that I found to still hold up quite well today: https://www.youtube.com/watch?v=xn8zj1--_bg

No, not me; I did use 16-bit int for a while, but I actually have been using 24-bit int for a few years. Any differences I may or may not have heard are really minor (w.r.t. 16-bit int/24-bit int/32-bit float). You'd sooner hear quality issues if you decrease the bit rate (e.g. 192 kbps vs. 224 kbps) upon encoding than if you decrease the bit depth (16-bit int, 24-bit int, 32-bit float; basically bit resolution, analogous to pixel resolution) upon render. ----- However, whenever I have rendered 32-bit float WAV files, encoding into VBR1 MP3 via WinLAME gives me a silent MP3. My 24-bit int WAV files don't run into that issue.

Well, the main thing I can say is to vary your velocities a bit more, because the instruments sound quantized, so they sound like a robot played them. If your velocities are even a little bit deviated from the grid, the instruments should sound a little bit more human and less lifeless. Other than that, there isn't much else I can say right now, because there isn't much content there. The best I can say is, "I need more to work with", because due to the repetition, about 20% of each songs' length introduces unique content.

Despite the limited source material, it doesn't mean you can't come up with your own melodies to contribute to the remix. Also, I gave you several examples to compare, and mentioned that the drums should be more rhythmically interesting. Did you try adding something more than a kick drum? Any timpanis, bass drums, snare drums, congas, bongos, darbukas, djembes, taikos, or anything else?

I'm not really getting a "battle theme" feel here. "Bond of Fire" feels like a standby theme, and "Chidora Omega" vaguely sounds like a boss is talking, but not a "battle theme". In both, there really isn't much in the way of drumwork. It's pretty much a kick drum. Battle themes hinge on rhythmically interesting drums to convey excitement and tension. Furthermore, the leads are "too nice"; they don't make me feel like I'm battling. More like, I'm relaxing. The melodic writing is also very repetitive; in both songs, I hear at most two short repeated melodies, and little compositional development. As a result, the dynamics don't change. No amount of mixing can fix this kind of issue. If you try skipping 15 seconds at a time through your songs, can you honestly tell me that you can tell the difference between the beginning, climax, and the loop point? That would be amazing, because I couldn't. Here are some examples of battle themes: https://soundcloud.com/overclocked-records/zircon-dungeonmans-ost-05 https://soundcloud.com/overclocked-records/zircon-dungeonmans-ost-07 https://soundcloud.com/overclocked-records/zircon-dungeonmans-ost-14 https://soundcloud.com/overclocked-records/zircon-dungeonmans-ost-22

From what I can tell, it uses the highest note as the upper bound and lowest note as the lower bound, calculates the midpoint and uses that as the reflection axis, and reflects each note in the highlighted set over that axis.

Actually, a lot of these entries are really good! It's a very interesting idea to write something based on a line from a book... I may try it in general (not for a contest, just for fun).

Oh yeah, this one! Yeah, this was a pretty fun remix to see on Round 3. The production is fairly clean. I did have a few nitpicks, which were that the left-panned wobble was a bit too resonant for me, and the lyrics weren't really intelligible. That crit about the lyrics turned out to help you though, I think; can't violate the standards with unintelligible lyrics (?)! Other than that though, the meshing of the sources was fairly seamless, and I believe I had voted this highly. EDIT: yup! I put (2).

This one was used:

Pretty much exactly what I do, except I use different orchestral libraries.

Well, that's not true, dude. I'm obviously part of the community, and I contributed to the preset creation. Hence, someone from the community has contributed to Super Audio Cart. It doesn't automatically discharge me from the community, nor does it make my association with the community a mere claim. --- And well, let's say this too: I'm very familiar with Zebra2 and that has been my primary, mainstay synth for 4 years. I will always recommend that to anyone who can afford it. After having the privilege of being a part of the beta testing for Super Audio Cart, I can say that it will be a second mainstay synth, and I will similarly recommend this to anyone [with the money to spare].

Great work! I thought the lead strings attack was a tad weird when they were more exposed (0:22, for instance, though elsewhere they sounded clean), but I found the brass to be particularly excellent.

Yeah, I'm kinda 2:30 into the track... :/ If you want to go with your track, then just tell me, and I'll treat mine as a separate endeavor.

Huh, well, I found time/inspiration to do a Pokken Tournament track this morning, and I'm about halfway through! :3 If anyone else can write something in 3 days, by all means, give it a shot! Shariq wants to get finished tracks by the end of this weekend.