Jump to content

Rozovian

Members
  • Posts

    5,295
  • Joined

  • Last visited

  • Days Won

    16

Posts posted by Rozovian

  1. On 9/29/2019 at 7:07 PM, Dezorianguy said:

    Or is there a way I could do it myself with some practice and the right kind of program to make a simple rework on the amiga chiptune originals? 

    Amiga chiptunes huh? Probably .mod files. You could import them into a more modern tracker (specialty music program for this type of music notation) and swap out the simple samples for more complex ones. That won't necessarily sound good, and it's nothing like making an orchestral track (whether properly arranged for orchestra or not) in modern DAWs (also music programs). Those require arrangement, humanization, sound libraries, mixing.

    If you have some skill with music and good ears for mixing, you can probably do it yourself (to a passable degree anyway), as those tend to be the more important things to learn when learning how to make music yourself. If not, you'd best make friends with some music people that know the game and its soundtrack. Not to discourage you from trying it yourself, making music is fun and worthwhile, but it does take time.

  2. That bass is humongous. For headroom reasons, you should probably ease up on the lows a little, just don't overdo it.

    There's a really loud couple of notes around 2:13. Watch out for stuff like that.

    I like how there's this noisy thing being side-chain-bounced at some parts. And I like that riser at 1:40. And the drum break at 3:35. A lot of little things to like, besides the overall sound.

  3. Good news, bad news. Let's start with the bad.

    On 1/17/2017 at 6:41 AM, Wiesty said:

    Often I find it's helpful to know the limitations of your samples. If you have string samples that sound really bad playing fast passages, it might be better to re-write the part for an instrument which can handle the music you've written for it. 

    This needs repeating.

    The piano-strings-woodwind interplay in the beginning could use some work. I think it's the offbeat start of the piano couple with the slow attack of the strings that's throwing me off. Better strings might solve it, though a single low piano note might also do it. Either way, a good downbeat would anchor the rhythm and help orientate the listener. The strings are a bit of a problem here. Not their writing but the sound itself, hence the quote.

    You should probably redo the mixing entirely. It's salvageable as it is, but starting from scratch (levels, eq, reverb, compression) with what you've learned since you did it last time can help a lot. Any sound-design-related effects can probably stay. I'd mute all the tracks that aren't the most important for the tracks, and make the most important ones sound good on their own, then bring in the secondary tracks one by one and adjust their levels and effects accordingly. Maybe even do two passes of primary tracks, one for the melodic instrumentation and one for the rhythm parts, and then mix them together before bringing in tertiary things.

    It sounds like you're fading out at the end, which you don't have to. You could end on that 2:28 note (or one earlier) and just let it ring out. I can't tell if that's a fadeout or just low-velocity notes on a piano that doesn't sound right at low velocities. But I don't think you need to soften it that much anyway, just dropping out the other instruments does a lot for the track already.

    Your piano humanization is successful in that it sounds human, but it's not a great performance. From what I can tell, there's a lot of timing adjustments in here, but they sound more like random timing imperfections than a performance. If you can record midi, I suggest you play the parts yourself, even if you only play on a single note, just to get the timing and velocity right, and then create the melody from that. If you have some soft humanization tools, you can use those too. This is less important for instruments with slow attack.

    The frequent breaks in the beat are a little strange. You could mitigate that with a more percussion-oriented track that doesn't do the breaks, or that leave cymbals and things ringing out over the breaks. Or maybe a heavily filtered copy of the drums you've got. These might not give you the sound or style you're going for, it's just what I'd do. (Beware the "what I would do"-type suggestions, they might not work for what you want to do.)

    I don't think humanization (or arrangement) is the main problem here, sound design and mixing is. You could improve the track a lot with just a few changes to the sound design and a mixing overhaul.

    I like the arrangement. I like the bass glide thing at 0:27, and the glide effect at 1:29. I like the sound of the snare and those highest little string things. I like the muffled drums sound nearing 2:00. And I like how this is from a source from a game I haven't even heard of.

  4. On 10/2/2019 at 1:21 PM, BloomingLate said:

    This is an area I definitely need some direction in. With reverb, what would I need to do to make an instrument sound farther away without creating a ton of unnecessary "echoing". When applied to the drumkit it tends to really boost the kick and snare sound beyond what is desirable. Maybe I'm using too big a "room"? And as for panning, is it advisable to pan a drumkit or bass guitar or should they stay centered (and thus keeping that foreground feel)? Would it help to draw up a "plan" for a "stage" as it were, to determine where each instrument should go and pan accordingly?

    Reverb - Big halls don't work for all tracks, so yeah, you might want to use a smaller size. The parameter might claim it's 40m, but don't believe it. Use your ears, adjust to taste. And you don't have to have a long reverb. Length can be fairly short. The dry/wet ratio lets you adjust how clear a track is, so less wet means more foreground-y. You can filter and eq the reverb too, and the reverb plugin might have some options for that, like low ratio or something. At least filter out the lows. If you can set early reflections separate from reverb, you can give the reflections of your leads a longer pre-delay, so their attacks are clear, while the attacks of background instruments blends into the wet signal.

    Of course, you can do a single reverb bus for the whole thing, or multiple (e.g. foreground, background, distant), or give each track its own reverb. Or some combination. Different methods give you different options, like full control over a reverb bus with eq and side-chaining, for example. Reverb levels per track matters, more reverb means more background-y sound. But track level is ultimately determines foreground-y-ness, reverb is an addition/enhancement to that. As is panning and eq.

    Panning - Our ears easily tell where high frequencies come from, not so for low frequencies. Center is usually best. Usually. A stage-like plan can work, depending on the music, but I find the better way of thinking is to spread out frequencies, to spread instruments depending on their roles. Kick, bass and snare middle. The rest of the drumkit mimics what a drummer hears (so stage, but mirrored) with the amount of panning adjusted to taste. With the hihat panned left, other high-frequency percussion can go right, eg shaker. If a guitar goes left, another guitar (or any instrument occupying roughly the same frequency range at the same time) can go right. For this track, I wouldn't hardpan anything, I'd go for a kind of jazz club thing, with some instruments panned a bit, others not at all.

    There are different schools of thought when it comes to panning. I can think of a few:

    -No panning (stereo is just for stereo-recorded tracks and for effects)
    -Listener-like panning, with variations:
      -Drums from drummer or listener POV
      -Drumkit and bass centered, or placed according to band
    -Center and hardpan only
    -Frequency balancing (works well for my tracks)

  5. First, I like this. Especially that percussion thing. Really good sound.

    Humanization is tricky. Too stiff and it's mechanical. Too loose and it's just poorly "performed". This sounds human enough to me. Except the stickerbrush ostinato. Its timbre and the mix makes it stand out, and the timbre is difficult to humanize anyway. I tend to cheat by recording (poorly) and then quantizing to 70% or so. But I make more electronic sound stuff anyway.

    I'd spend some time working out which instruments you want as foreground and which ones you don't, and mix accordingly. The percussion (which sounds pretty nice), sounds very foreground-y, and the guitar which sometimes functions as a lead, doesn't. You adjust foreground-y-ness with track level, eq and reverb. EQ down the tracks that aren't supposed to be foreground, muffle them slightly, make them softer than the foreground tracks. If your foreground tracks come out of their synths/samplers/recordings already muffled, there are tools for adding higher frequencies (energizers/exciters, sure, but you can also make your own with a bus with distortion and a filter).

    You can also consider adding some higher-range percussion to the earlier parts of the arrangement. I've been using various shakers since GSlicers recommended that. There are many tools for that too. It might smoothen those parts a bit. If you find a nice shaker loop, that's good. If it's for a background part, you can probably record a box of rice on your phone, too, just filter out any room/fan/clock/other noises. Touches like that add a lot of human feel to a track.

    This is all advice applicable to this, if you change your mind about it being completed, and to future mixes.

  6. Mixflood selection sounds like a lot of fun, but when you're actually doing it it's so much harder. For me it was about narrowing it down to categories and also knowing a few artists that had to be represented in the mixflood. I went for collabs to try to cover as many artists as possible, and also ones where I could explain something about the album, not just where the tracks themselves were representative. But that too. It's a big dumb puzzle to figure out.

  7. Nice change in style. Much more dramatic. A simple arrangement, but it works well.

    A piece of constructive criticism though: I noticed you had a bit of a swing thing on the drums where the other instrumentation was straight (before 0:28). It's usually a good idea to try to keep things lined up more for a tighter feel, whether that's in a swung rhythm or not.

    Kudos on the video production, too.

  8. Heroic with a flat fifth might work. Chromatic movement in the bass. Phrygian mode. Some ideas.

    I'm noticing a lot of your examples have aggressive rhythms, be it GoW's staccatos or the metal-inspired Shadow stuff. Starting from a rhythm might be a good approach.

    Heroic but screwed up, whether you go dark, overly aggressive or creepily off-scale depending on what you're after.

  9. Just the other day I saw the track was in the to-be-posted queue. And now here it is. :D Uh, merry... late summer?

    (also Ophilia best girl indeed)

    For Rexy's benefit: I was playing around with the 1:48 part of source from 3:18 to 4:18, followed by some minor scale variants of melody A until 4:48. (insert appropriate sneaky fox meme here)

    It's worth saying I really enjoy Rexy's analyses of remixes. She's got a crazy in-depth one of Dragonfood, and I keep coming back to that one to remind myself I've made something that someone has listened _that_ closely to.

  10. Some albums don't have their own sites, afaik. Every so often, once a year maybe, I grab all the new albums. On more than one occasion, I've found albums without a dedicated site, just the ocr page. Which is fine, though it could be more stylized with album artwork elements or something.

  11. Transmission, for OSX, lets me pick folders or files to include or exclude, and prioritize some over others. Dunno if that's standard for torrent clients. There's apparently an early build version of the Windows version of Transmission, but I can't speak for how well it works.

    I don't share the concern about missing out in case of more selective listening. I started out listening exclusively to Zelda mixes, then started listening to Metroid mixes as well, and now I've got probably every ocr release, including many since removed. Some of my favorites are from games I haven't even played. Small selections are bite-sized explorations. Consider a work-out selection, a winter holiday selection, a Halloween selection, a night-time driving selection, a romantic evening selection, a chill background groove selection... 3k mixes can be overwhelming. A few hundred is already a lot.

  12. No need to be sorry. We'd rather have people show up to want to help correct what we get wrong that people not care about what we've created. And I fully understand the confusion over the track names. They _are_ a mess. I think we found three different sets of names. It's mostly the battle tracks that get shuffled around, but some others as well, like Falcon or Legend or whatever that track really is. I think Usa checked this stuff with the official soundtrack, so it should be correct, and the tracks from the archives we couldn't find on the original soundtrack, we still tried to cover as well. Poor Sixto accidentally got his Obsession Nocturnal posted on ocr normally when it was thought the project had died, and another of his tracks we weren't even sure it was used in the game.

    Names, man. Them names.

    Where's that link at?

  13. I tested a little bit: https://ocremix.org/remix/OCR03894

    I noticed the blueblue.fr link having some issue for this track. I didn't get "waiting for network" but the track never loaded.

    I tried an older mix, and no problem there for that mirror: https://ocremix.org/remix/OCR01219 So it might be something with recent uploads on blueblue.fr.

    But you say all three links have this problem for you? Which remixes did you try this with? Do you have the same problem with old remixes and new ones?

    What phone OS, what browser? iOS, android? Safari, Firefox, Chrome, something else?

  14. [This is an automatically generated message]

    I've reviewed your remix and have returned it to Work-in-Progress status, indicating that I think there are some things you still need to work on. After you work on your track and feel that it's ready for submission to OCR, please change the prefix back to Ready for Review and someone will review it again. Good luck!

  15. Late eval, we apologize for that.

    And not much of an eval, either. No source link, no source comment.

    It seems to be mostly voice clips from the game, random vocal samples, big edm synths and a nice beat. I'm not sure what melodies are relevant to an eval.

    There are also some weird key changes between parts. Jarring, but not a dealbreaker.

    Look, you know better than anyone else if this contains source material from GoW4's music and to what extent you've used it. I'll mark this as wip for now, link a source and change it back to ready for review if you want it evaluated for being posted on ocr.

    Fun track though.

  16. eval

    Still sounds like a midi mockup. Decent piano sound, no problem there, but the performance isn't very convincing. I can hear variation in velocity, but that's mostly between individual melody lines or voices, not within phrases. There's also no timing variance, which is also something used (unavoidably) by actual performers. I don't think there's any use of pedal, either. Additionally, this seems to be a three-handed piece, as best as I can figure. Not a huge problem, but deliberately four-handed (as in for two pianists) or downscaled to two-handed would make more sense. That's far from the biggest concern, though.

    Good choice of source, and I think it works quite well as a piano piece. It seems to be quite conservative, but I'm not terribly familiar with the sources and what makes each unique and different. As an adaptation for piano it's a lot more about the performance and the embellishments rather than overall structural transformations. So I can't say whether it's too conservative. Transitions are rather abrupt at times, e.g. at 1:27 and 3:32. Dramatic changes between sections aren't a problem, but it has to flow better than that.

    A solo piano track needs a good performance. A more thorough arrangement-focused review could tell you if the arrangement itself is passable after transitions are dealt with, in which case you could ask a pianist on the site to play it. They'd also be able to provide better feedback on the arrangement in regards to it being playable, something that's well beyond what I can say. What I can tell you is that in its current state it won't be posted.

    Yeah, I find the stiff performance the biggest problem here. It's something you seem to struggle with, a recurring thing with your wips. I suggest you either work towards improving your performance skills (either actual performance, or faking it by manually humanizing sequenced material), or shift to a genre where the sense of performance comes from sound design more than from the notes themselves, or both. Whatever you do, good luck.

  17. [This is an automatically generated message]

    I've reviewed your remix and have returned it to Work-in-Progress status, indicating that I think there are some things you still need to work on. After you work on your track and feel that it's ready for submission to OCR, please change the prefix back to Ready for Review and someone will review it again. Good luck!

×
×
  • Create New...