Jump to content

YoungProdigy

Members
  • Posts

    116
  • Joined

  • Last visited

  • Days Won

    1

Posts posted by YoungProdigy

  1. At some point you have to realize where your samples have their limits. Sometimes a particular performance is too difficult to emulate precisely, so maybe you're, let's say, 90% there, and that may sound like a realistic-enough, yet slightly different performance.

     

    For example, I recreated the first solo for Bohemian Rhapsody two years ago. It still holds up pretty well today IMO, but it sounds like a different performance when you compare it to the original (besides the tone of the guitar and the obvious lack of vocals in mine). I think mine sounds a little too perfect, specifically in how the notes connect on the ascending scales, for example.

    Yes I agree with you that some samples just have their limits.

     

    In cases like those; should I even shoot for realism or just try to make the instrument 80-90% realistic?

  2.  

    Let's dissect the question while answering it, because I feel it's important to change your mindset before mulling over the answer. These answers are also according to personal philosophy and not professional advice, so they apply to questions of values in music, but not necessarily as a process for how to go about professionally doing music (those questions are answered by considerations of budget and time, and not always artistic ideals). I answer the question like this specifically because you're posing the question in response to your audience having issues with your realism (as opposed to your "client" having issues with your realism).

     

    "Are super realistic samples really necessary for video game music?"

     

    We can substitute "video game music" here for "music", because video game music is completely and totally equivalent to (a subset of) music and should be held to the same standard (because they are the same thing). Because we're not doing professional considerations, differences in medium between film, tv, games, etc. don't apply here.

     

    "Are super realistic samples really necessary for music?"

     

    This depends. Are we trying to use instruments in our music that are performed acoustically? If so, then yes. The purpose of super realistic virtual instruments is to emulate human performances with ease. The goal of making music is in general to sound organic and human, to sound like it was something expressed by real people at a real time and place. This doesn't apply to many electronic elements, but it does to some, and it certainly applies to all acoustic instruments. If it's a real instrument, and it is sequenced poorly, it sounds awful, and because we're culturally still accustomed to hearing what real performances are like, we do have lackluster reactions to fake performances.

     

    That being said, are "super realistic samples really necessary"? No, but they make the job, your goal, easier. So let's try this one, then.

     

    "Are realistic samples necessary for music?"

     

    I'd say realistic samples are necessary for intuitive music composition even if super realistic ones are not. The distinction here is that super realistic samples are very expensive and have lots of features. However, skilled composer-producers can work with not so super realistic samples to still make them sound like organic performances. The requisite here is that the samples at least resemble the sound of their real-world counterpart ("realistic"). Even with less realistic samples, composers still manage to get great results, but that begins to cut into your composition flow, because you're spending more time MIDI programming than you are focusing on the music.

     

    I consider intuitive workflow and minimal tinkering a necessity for music composition, so I hold that samples need at least a decent level of realism in order for me to put them in my music. Though, in actuality I always go for the biggest and best super realistic ones.

     

     

    To clarify, my response is assuming you're writing music without the intention to explicitly sound fake in the realm of sound chips or dated production eras, so Timaeus's answer also holds.

     

    Well to answer the question of "Are we trying to use instruments in our music that are performed acoustically?" the answer would be yes. But in my experience; I've come to realize that without a dedicated VST for certain instruments, that realism can be very hard to achieve. For instance; a realistic jazzy smooth saxophone like in the song in that youtube link, would be almost impossible with just one multivelocity sample. Making realistic guitar parts from just one sample is also difficult.

     

    And I agree with you that super realistic samples are expensive. I could probably get a realistic sax sound from "Mr Sax T", but that's around $164. I would say my current samples do resemble the instruments; but you can still tell they're fake. Especially when it comes to guitars and saxes. It's less of a problem with multi-articulation instruments though. But I have another question. Could I truly make something similar to the sax in that video, with just programming? And could I truly make a convincing guitar solo with a single articulation sample?

  3. Hey, Young Prodigy here. One of the biggest critiques I get is that my songs lack realism. . I use Sampletank 3 SE and most of the samples are of acceptable quality. But even with decent samples; it's still hard to make certain instruments sound realistic. Take the saxophone for example. Even if I add a lot of velocity variation, portamento, vibrato and expression; it still ends up sounding fake. It's the same situation with guitars too.

     

    Take something like this:

     

    I could never achieve a sax sound anywhere near that with Sampletank 3 SE.

     

    I realize there are realistic vsts dedicated to saxophone and guitar; but they are expensive and I can't afford them right now.

     

    So back to my original question; are super realistic samples really necessary for video game music?

  4. The rhythm on this feels strange. Everything sounds like it's at a different amount of swing with something else. The shaker seems to not have much swing, the bass has a touch of swing, and the drums just have some off timing ("sloppy", but not in a mean way, and both late and early timings are present). For me it just distracts me from enjoying the mix otherwise. In general I think you got the vibe you were going for, and the instrument choice made sense, but the rhythm on everything should be more in sync with each other.

    Thanks I'll try to make the rhythm more in sync.

  5. I like the chords you're using. The instruments though have little to no expression and reverb; the synth lead has no vibrato or variation in velocities, the piano has no variation in velocities or rhythm, the drums are on autopilot, and the bass pattern sounds copied and pasted. It's a good start, but there should be more attention to detail on the sequencing to make it more realistic and personalized, and the lack of reverb makes the piano, especially, sound more stiff.

     

    (I'm not recommending, however, to put reverb on the bass)

     

    I think it would help to re-read that reverb article I showed you earlier, and see if you can put it to use while practicing with a reverb plugin---putting something into practice pretty much requires actually using it, so... yeah.

    Thanks for the critique. I agree with you that the synth and piano melodies could probably use more reverb. I'll also try to add some variation to the bassline and fills to the drums to make them less stiff.

  6. The real question you have to ask yourself is what are you missing that you feel the professional tracks have. Is your low end completely out of control and lacking that cemented feeling? Are you lacking that certain intangible notion of depth and space? Are you lacking the stereo width? There are seriously dozens of questions to ask yourself of what you might be lacking, but first have to identify what it is you're missing or at least what you feel you're missing.

     

    A lot of times mixing is really performed in the composition without even realizing it. A great mix stems from a great composition. There are things that can be done in mixing to force things around but this is fixing something that should have been identified in the composition. Start mixing from the get go in the composition. Getting the spacing of instruments and macro dynamics there sorted out. Then when it comes to mixing these can be enhanced. 

     

    Having said all of that without an example I'd be hard pressed to give you more than general/philosophical advice about mixing. I can say that for sure the biggest two things for me have been to listen lots of different material ranging from Jazz to Heavy Metal to Trance to Pop to really most anything. This has really helped me in coming up with a sort of mental palette of different kinds of sounds a track could possibly have. The other big thing is practice. Lots and lots of practice. Every technique you read about in reality is more of a tool in an ever growing toolbox. It isn't often that you really should have to break out EQ notching and then using that to feed into distortion to make a sound usable. However, when you need that kind of technique for something then it certainly comes in handy to know it and how to use it. Ways to create width, space, depth, control low end, prevent masking, use masking to your advantage, de-essers on cymbals, etc . . .. Just techniques in what should be an ever growing repertoire of skills. But you have to practice them. 

     

    I can say this for certain. I mix my own material quite differently than I do someone else's material. It is really important to try understand what you want the track to sound like at the end. Having a clear goal in mind of what you want it to sound like is a very important thing to keep in mind when mixing.

    On the subject of listening to professional tracks; the problem is that I just don't know how to achieve the same sound. If I listen to a professional track; I don't know how to achieve the clarity of the drums, the low, but not muddy bass, ect. As for my composition skills; I think they're pretty good. The main problem I face is mixing. And it seems no matter how much I practice; I just can't figure out how to get that perfect mix.

     

    As someone who struggles with awful mixing, I'd be very tempted to just blame it on my system (AKG k240 + Bose PC Speakers) to make it seem like it's not my fault. However, there are issues I can address regardless of my system by having references. I can distinctly hear the difference between my stuff and better stuff, I just don't put in the time and sweat into hashing out how to do it (usually because I'm mixing on a deadline and don't have time). I don't gain stage effectively, and I overcompensate in EQ (literally all the time, my EQ curves are always between +/- 6-12 dB cuts and boosts).

     

    For instance, as a YEAH HEAVY POUNDING fan, it's very tempting to jack the shit out of the 100-200 Hz range, and I often do, unnecessarily. Or I won't mix hot enough, and my frequency graph just looks like a downward curve. I got better about managing sub (usually cutting it out since my system can't actually tell me what's there, so better safe than sorry).

     

    If I had to point to a singular reason why my mixing sucks, it's because I don't listen to my own mixes. That's the truth; when I make something, it's usually in a ridiculously small timespan. I just make it, send it to zircon or someone to get feedback, then it's done. I'm much better with doing this in an orchestral setting than in metal or electronic. Even so, the problem is that when you start out, your ears are still in "omg this is so cool" mode when you make stuff, and you block out the mixing flaws because your ears instead perceive what you want it to sound like. You are *actually* not hearing it the way it physically comes out of your speakers, it's not that you're crazy. Perception is a deep, unintuitive science.

     

    The way to fix this is to sit on your mix for at least 4-7 days without hearing it at all. When you come back to it and listen to it again, the self-love perception with fade off, and you'll hear the mix again for what it actually is. Then you can address its flaws. Keep doing this as time goes on, and then as you become more experienced, you start to just be able to properly mix from the get go, without the sitting period.

     

    And yeah, getting a better system will help properly mixing from the get go as well. I'm going to buy actual monitors starting in June. That'll be fun!

    Hmm, I might try that; mixing and then coming back to it later. But I think my main flaw is when I have a bad mix; I just don't know how to fix it.

     

    Ultimately I can sum up the sources of some of my past mixing issues to be (pre-2015):

    1. Overboosting EQ in general

    Sometimes it's slight, sometimes it's super obvious. Listen closely.

    2. Not choosing good sounds from the get-go

    Pick bad stuff, and you won't have a cohesive palette to work with. Furthermore, if the EQ on the sample is good as-is, you don't have to EQ as much to get it to sound good in context.

    3. Volume imbalance between instruments

    This depends on your listening levels, but ultimately, it's hard to get volumes just right sometimes, especially when the complexity of the layering is high.

    4. Not-so-good audio system

    Goal: flat frequency response, ideally. That way, you're not biased towards certain boosts or cuts to compensate (or overcompensate).

    5. Compensating for loud mixing with poor-quality compressors

    Yeah. Think before you put on compressors. Why would you want to use something if you don't know why you're using it?

     

    Hardest thing to fix for me: 1

    Easiest thing to fix for me: 5

     

    If you get a good audio system, you'll be on track to addressing most of those. That's not to say it won't be hard. The rest is constant practice, patience, and critical listening. Good luck.

    Thanks I might try to look at my mixes and see if I'm making any of those mistakes.

     

    The big thing I struggled with for a long time was mud, which is frequencies that detract from the clarity of your mix in the mid-lows. Mud typically exists between 200-500 Hz. To get rid of it, on each track of my mix, I use a parametric EQ with a narrow Q (usually I set the Q around 7 oct/db) and sweep across those frequencies with it boosted until I hear something that doesn't sound too good, and then cut that frequency. I often have to broaden the Q a bit for the cut to make it work. I do that 2-3 times on each track to clear up mud. Just don't get super aggressive with it and it will help a lot.

     

    I also do a low cut up to 175-200 Hz on pretty much every track except for the bass and the kick, and that gives a ton of breathing room for the kick and bass. 

     

    Drums is another tough one, you want them to sound "phat" (which is a whole other walkthrough itself) and it helps too to understand what frequencies each drum (snare, kick, toms, etc) needs in order to sound the way it's supposed to. This guide helped me a ton: http://www.benvesco.com/blog/mixing/2008/mix-recipes-tom-eq-and-compression/ It's for toms but there's links in the first paragraph for kick and snare too.

     

     

     

    EDIT: I know we've all posted a ton of info, but honestly, it would probably best help us figure out how to help you if you could post an example of something of yours that you're not satisfied with.

    Well I've tried filtering out at 250hz but my mixes still seem to be a little muddy.

     

    But listen to this mix guys:

    https://soundcloud.com/youngprodigymusic/title-music

     

    In this mix; I picked decent samples, I tried to balance the volume levels and I tried to EQ it. But to me; it just doesn't sound good. The drums are buried and not in your face. The bass is also a bit muddy. Overall it's too boomy. It's just missing something that I can't put my finger on.

  7. Pretty nice theme. I liked the guitar; it was very realistic. However; I have to admit that the drums sounds kind of fake. Particularly the hi-hat; it has that machine gun effect. I think varying up the velocities on the drum samples will help them sound more realistic.

     

    Other than that; the piano samples are a bit dry and could benefit from some reverb.

     

    But composition-wise; this song is pretty great. Keep it up.

  8. Well, right off the bat, the guitar just sounds tonally and notationally fake, quite literally like a computer is playing it. :/

     

    The writing isn't bad, but the fake guitar is really bringing this down. It's kind of inherent in rock and metal that you need realistic guitars. They don't have to be super convincing, but they aren't up to snuff here IMO.

     

    You're hitting point #2 on this post:

    http://ocremix.org/community/topic/40776-mixing-what-am-i-missing/?p=785497

     

    and points #4 and 8 in the last section of this post:

    http://ocremix.org/community/topic/25003-post-most-common-mixingmusic-problemsmistakes-you-see/?p=784385

     

    Basically, you need to train your ear to pick out good sounds. I honestly think this is more of a writing issue than a mixing issue. This can't be mixed to sound like a convincing rock track because the specific samples chosen here are pre-distorted and lack round robins. An example of a normal rock track:

    https://www.youtube.com/watch?v=B5nMWTR7wb0

    If you compare the two, it should be noticeably different on the tonal realism.

    Well yes the lead guitar was made by putting a dry Sampletank 3 guitar sample through the free version of Amplitube 3. Should I perhaps try a different amp simulator to get a less nasally tone? Or maybe I should try to mess around with EQ until I get a realistic tone?

     

    I've also heard RealStrat is pretty good; but as of now, I don't have the money for it.

     

    But thanks for the critique I appreciate it. I'll definitely try to focus on improving on those points you made.

     

    Honestly the composition of this is pretty rad! Cool riff, cool lead, a little repetitive at first, but I mean that's pretty chill considering how you use the key change and pads to change things up. 

    Yea to be honest the biggest issue is that the composition is sweet but the sounds are terrifyingly off for this sort of piece. The distortion is really overtone-ridden. I'm not sure what sort of samples you are using but since they sound like they are pre-distorted. If you could take some time and program some "errors" in the playing it'd help, but also some velocity changes.

     

    I would suggest trying to get the guitar sound through an amp sim, and try a clean guitar sound that is really "dead* sounding without a lot of overtones... it helps the distortion. I mean if it sounds a little mechanical that can still be fine, but the biggest problem with sampled guitars is that it ends up sounding super fake because it's just like... straight distortion and the overtones don't get smoothed out by some kind of speaker sim. 

    Thanks for the critique man.

     

    What I did to get the tone of that lead was I put a clean guitar sample from Sampletank 3 through Amplitube 3.

     

    I may try to use a different clean guitar sample and see if I get better results.

  9. dude, you're posting on OCReMix. There's a bunch of super high-quality music here, why are you still looking? :P

     

    In all seriousness, you can use Spotify, Amazon Music, Google Play, Pandora, whatever. Just find music that's done well and then listen to that, then mentally compare your music to it. That's basically it. 

     

    If you want some recommendations as to well-done music, I'm sure a bunch of us can chime in--what genre are you looking at right now?

    True there are a lot of good tracks on this site. But there are a lot of songs so it takes me a while to find the perfect reference.

     

    But I'll probably check Amazon Music and Google play too.

     

    I do pretty much all genres by the way.

  10. I've been trying to improve my mixing skills lately. I've read almost every guide on EQ, compression, panning, frequencies and the like. But in FL Studio my mixes still don't sound that professional. Am I missing something? Or is there something I'm doing wrong?

    I use Sampletank 3 which has effects on the instrument already. Would mixing with those effects off improve my mixes?

×
×
  • Create New...