Rozovian
-
Posts
5,295 -
Joined
-
Last visited
-
Days Won
16
Content Type
Articles
Profiles
Forums
Events
Posts posted by Rozovian
-
-
Good job everyone who got something done this year. It'll be interesting to hear what you all have come up with.
21 hours ago, Dyne said:Secondly, about the updated site; part of the reason for the update is that I'm learning frontend development, well trying to at any rate, that said a site update has been long overdue.
Going forward all of the albums will be on the main page, and they will be chronologically newest to oldest. This means no more clicking around to each volume to grab the albums you want. You can do that from the home page. This also has made my life easier and increased the speed at which I can get everything done and posted. I think I just spent about 20 minutes adding all the information onto the site itself, testing it, and then uploading it all to the server. I used to have to go to each and every page to update navigation and that was a very time consuming task. So, there's that.
I've also long wanted to change the color scheme. Initially, I was going to stick with green because I love the color, but also because it featured on the original version of the site. I decided to mix it up a bit and go with a blue color scheme, and I think it's a lot nicer.
About this, the server is slow, so you should probably use smaller images on the main page. We're a bit spoiled with most big sites being fast, so a lot of them don't bother will optimizing image size. But it's pretty noticeable here.
I like the general vibe of the new colors, feels more wintry, though the background is a bit too bright for my liking.
Overall, it's an improvement, and should indeed make updating a lot easier. I've worked with php, where you can easily update an element that shows up on multiple pages. Requires a different server setup, though. In html, this is probably the best approach, though in 17 more years the page is gonna be loooong. But by then you'll have learned javascript and can do some show/hide things. And linking to anchors/bookmarks. And more.
Good job.
- Wassup Thunder, Xaleph and Dyne
-
3
-
Someone that knows this stuff well enough should do a write-up on how to release remixes properly licensed on these services. I know there's info out there, but this is where it's relevant, and not all info out there is.
-
Seems like your main problem is platform specific, so I'd do an experiment:
Upload with more headroom. Spotify recommends -1 dB True Peak. The article on their site said it's to avoid distortion or clipping when encoding to lower quality. Or something like that.
I'd even go as far as to upload it as wav, highest quality, more than 1 dB of headroom. And maybe just the hats and cymbals. If I don't hear a problem with that, I can start stepping back towards my previous upload until I replicate the problem.
--
As for samples, I wouldn't mind some samples for layering, normal drums and other percussion and weirdly recorded ones and other useful sounds. Any recommendations?
-
Sounds like someone got rejected and is trying to deflect. I've done that too, in at least one bad attempt at humor. Some good points here, but I think they get lost in an unhelpfully negative attitude.
Since OP supposedly won't respond, let's just talk about some of the points being made here. I'll ignore the ones conflating tracks passing the panel with tracks being good, because not every track is suited for ocr and vice versa.
16 hours ago, Ocean_Charity said:Volume levels should be normal compared to the average recording.
The standards could benefit from an update along the lines of OP's suggestion of a LUFS number. Useful reference. I propose a range, so people aren't making their ballads and bangers equally loud.
16 hours ago, Ocean_Charity said:Next, there should be zero exceptions for any track submitted. They should all go through the process just like any other. The profile of the submitter should have zero bearing. In fact a better way to do this would be to anonymize the submissions entirely except for the piece of music remixed. This then removes any form of personal bias based on name alone. Again it is Submission Standards. Why does it matter that it was remixed by John Smith?
Nepotism! Like hoboka and platonist and velkku and a bunch of other familiar names among the recent rejections.
I kind'a agree, actually. I would have liked to read a bunch of positive remarks from a minimum number of judges on my Frozen Rose track which bypassed the panel and got direct-posted. But from a listener, non-remixer, non-direct-posted pov, direct posts look a bit sus. No specific suggestion on how to improve here.
Regarding album tracks, especially regarding those selected for an initial remix flood, they've been discussed internally on an album eval thread... Could that thread be made public after the album is released? In part so the remixers get to read about their tracks, which is nice and/or useful, but also to add transparency to the process, even if it isn't during the process.
16 hours ago, Ocean_Charity said:Further, there should be no re-submissions permitted. The voice of the Judges Panel is what it is. A remix of Super Mario Bros. Level 1 by John Smith fails to make it through the Judges Panel? Well, that sure sucks. But that is how it goes sometimes. If John Smith wants to re-submit the track, then it must go through the entire process from beginning to end again.
I think it could be made more clear what a resub is. Have it in the submissions standards/guidelines/instructions/recipe and/or in a sticky in the judges' decisions subforum.
For those unaware, a resub is a re-submitted track, one that has already been rejected by the judges. To have the improved version go through the entire judging process is a waste of time and energy, when all it takes is for the same judges to revisit their previous points and see if the new version has solved the problems without introducing any big new ones. Hence why resubs get fast-tracked to the panel.
I'm not sure a resub has to be given a "no, resub" vote to get to be a resub. That too could be clarified.
16 hours ago, Ocean_Charity said:Another thing to consider is actually when the topic or thread for a decision is opened even if it is hidden, is to let the whole community see it. This not only lets submitters see where their submission is or if it has even been listened to by the Judges Panel beyond an initial quality check. It opens up the process a bit. By including that original song link too, then actual real proper discourse can take its proper course and provide timely and critical feedback for the Judges Panel to consider.
The waiting time. Argh.
An automated system would be nice, one that takes in the tracks and creates the appropriate judging threads and updates some page or forum thread about the progress of them all. Initial eval, views, posts, maybe also votes. But that's a big coding challenge. Not to mention the initial eval might be a form rejection, simply a statement that the track clearly doesn't meet the standards, vague as they might be.
I suppose a possible solution for this is to give people viewing access to the subforum overview or whatever, the names of threads and the number of replies on each. Might be what OP suggested, just using the wrong words.
Having the entire community be the judge negates the need for a panel in the first place, and lets every beginner with a posse overrule more experienced folks. Not a good idea. And it would create an imbalance in attention even if the crowd doesn't get to vote. There'd be a lot of attention on Zelda and Final Fantasy remixes, while other games would be largely ignored. And that assumes the remixers are cool with their possibly rejected tracks being scrutinized by everyone.
I don't think the remixers usually want rejected remixes to be available to people, so them being removed by default is reasonable. Remixers can always opt-out of this default, opt-in to having the link left in. When submitting something that's more of an experiment and not being sure it's fit for ocr but wanting to try submitting anyway, that makes sense. When planning on resubbing if rejected, link removal makes a lot more sense.
16 hours ago, Ocean_Charity said:But due to the lack of any actual formal definition for any of the parts the Submission Standards fail at their job of setting a standard. Instead they form a guideline of how a submission should maybe be in order to pass the evaluation process.
Submissions standards might be a misnomer. Unless standards can work in reverse, and the standards list things that are cause for quick and easy rejections, while tracks that don't fail those standards are more closely evaluated. That's how I'd interpret how they work. Is standards the best term for that? Dunno.
Finally:
16 hours ago, Ocean_Charity said:The Judges Panel is only as good as the feedback it receives. Which is often in the form of long and confused posts from submitters that are frustrated with the process as a whole.
Quoted for irony. Possibly intentional.
I do think OP's greater point about transparency and feedback is a valid one. I've been staff, I've seen some of the staff subforums, I'm on and very occasionally check staff discord. So I have some amount of insight here, something not everyone has. And I remember my frustration with a community I greatly enjoyed but with a staff that walls itself off to make a lot of decisions in private. I've ranted about the feedback checklist and a bunch of things, often because there was little involvement with the community before those things were brought out. The things themselves weren't necessarily bad, but presented in a decree-from-on-high kind of way. I think there's a failure or unwillingness to leverage the strengths of the community, which is that it's actually a community.
So maybe there's a few points here worth talking about.
edit: forgot to mention mastering. do we count preparing a mix for ocr as mastering? because if so, it's all fine. otherwise maybe a bit of pedantry on that point is in order.
-
-
Oh look there's a trackball option.
Thumbball trackball, 3-finger hold. Had to go and check, on laptop atm.
-
Grab yourself a well-mixed track in roughly the style you're going for, and aim to mix similarly. Compare. Especially check that your lows aren't too loud, as they'll eat up a lot of headroom, and that your leads are bright enough, ie have enough high frequency content. It's okay that yours isn't as loud, mixing is more important than loudness, and loudness is easier when things are well mixed. Turn down your reference track accordingly.
You get used to the sound of your track as you're working on it, so make a habit of using a reference track to reset your ears so you can tell when you have too much reverb, too much compression, too much bass, too much of anything.
You can also check your mix by turning down the listening volume and mixing on super-quiet, because then you'll only hear what's most important in the mix. Use this to check that background things stay in the background and the important things are the loudest. Compare with your reference track like this too.
Use decent mixing headphones.
-
I can't count. Thirteenth.
-
How do I put this...
I would love to do a Christmas tune for a twelfth year in a row but that seems like an awfully big lift for me.
We shall see.
-
VBR has to do with data compression, not related to volume, so that's not an option.
iTunes on desktop has (at least has had; my version is old) a setting where it does some analysis and tries to do exactly what you're asking for, with varied results. Streaming platforms like youtube does this too, at least for uploads deemed too loud by its system. iTunes also let you adjust the volume of tracks and save this in the metadata. Depending on where you're listening to them from, you've got different tools available.
It's far from convenient, but you can also process batches of remixes with, uh, some audio tools out there, probably. There should be negligible effect on sound quality with a single conversion, if you convert to a high-quality format rather than convert to a different mp3 encoding (that's a xerox of a xerox). This is fine so long as you don't distribute these copies.
If there are no other tools available, you can always put compression (the audio kind) on the output from your device. This might require some special software to hijack and process the output, or a hardware compressor, either of which will likely cost you a little. And figuring out the right settings might take a bit of work, too. It will screw a bit with the dynamics of the track, and might make the start of a soft track extra soft... Not the best solution, obviously, but it's an option.
I guess the most convenient solution is to use a music player with this feature built-in.
-
Yeah, the VSTs themselves aren't all that big, but any sample libraries they use often are. These you might want to have on a separate drive, be this internal or external, USB or something else. Depending on their size and number, you might not need to, especially if you're just starting out. The kind of drive and connector doesn't really matter, so long as it's fast enough, and USB drives are more than enough for most cases.
Reinstalling is the safer option compared to just copying things, though I haven't had much trouble (on mac) copying libraries and setting a new library directory. The few VSTs (well, AUs in my case, because mac) that wouldn't let me do that I could just trick with symlinks. Symlinks, or symbolic links, are a special kind of alias on unix systems like mac, not sure if there's an equivalent option on windows. Moving the actual AU components (the VSTs, basically) would probably also work using symlinks, but I don't think it's worth the trouble.
-
On 7/4/2021 at 6:51 PM, WillRock said:
The interesting thing about the E chord is that over a C root, that creates a C7 chord
Cmaj7. A C7 is an Edim over that C. Flat 5th. Tritone for the E. And that REALLY really wants to resolve somewhere, probably to an F or C# chord.
- Garpocalypse and WillRock
-
2
-
-
I'm no lawyer. This is uneducated opinion.
btw remixing the song, without licensing it, might not be covered by fair use, depending on... things. If you want to be safe when it comes to all this, get a license. Game music is usually fine to remix (probably not ok to sell the remix), but commercially released music is different.
1. Purpose and character of the use
The first factor is "the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes." To justify the use as fair, one must demonstrate how it either advances knowledge or the progress of the arts through the addition of something new.
Quoth Wikipedia. It's transformative, in that it's not the original audio of neither the trailers nor the original song. And your own cut of the video material. Can't say if that's enough, but I would think so. Educational purpose though? Nope. But it might "advance ... the progress of the arts" though.
2. Nature of the copyrighted work
Fact vs. fiction, ideas vs. expression. This kind of stuff. You're not copying an idea, you're copying the expression itself, the video material, but re-cut. The material is fiction. The material is published. What else might matter?
3. Amount and substantiality
You're presumably using a lot of the material you can find. A trailer itself is not the whole game, but it is a complete video that's been cut, mixed, released.
4. Effect upon work's value
The fourth factor measures the effect that the allegedly infringing use has had on the copyright owner's ability to exploit his original work.
In this case, a positive effect, if any, on the trailer. Maybe also on the song.
So as far as I, non-lawyer, can tell: You're probably OK on the first and fourth factor, maybe not on the middle two. All are supposed to be considered.
- Master Mi and Garpocalypse
-
2
-
Sorry it took so long to give this a listen. Bad year.
Listened to the most recent version in the thread, 9a. I'm not familiar with the sources so I'll focus on the sound.
Let's talk about the first 5 seconds. The sounds used are all good enough, you shouldn't need to replace any of them. But pick a lead and make that the most prominent instrument. The pad swells into being rather loud. The piano plays the most identifiable melody, but it's rather muffled. The thing in the high range is bright and loud. Which is the lead? From how it's mixed, I'd say it's the high-pitch thing. From how it's written, the piano.
The piano sounds like it could use a high pass (low cut) filter to get rid of (or with a low shelf, mitigate) some low frequencies that don't add anything musically. Don't cut too high up, but see if you can reduce that low-frequency part of the sound. That should make the piano compete less with the pad for the low range, gives you a bit of headroom, and takes away some of the proximity effect that doesn't seem to match the distant pads or the hard-panned high-pitch chirps. You might want to do something similar to that chirp thing, too.
The next 5 second is in a different key/scale. There's a natural shift within those 5, but moving from the first 5 to the second 5 doesn't sound right. I'd experiment with transposing those 5 first seconds until the transition makes more sense.
Your synth pluck lead, despite being doubled at times by some background harpsichord-sounding thing, is weak. Your drum+bass combo is strong and aggressive. That's a difficult combo to work with. The accompanying pads are rather loud. You've put a lot of things into the background here. I'm not sure what to suggest to improve things here, besides softer pads. Lead down an octave? Play octaves? Different backing doubling? Different lead entirely? Make it more about the drum+bass groove than about melody? The lead sound isn't bad, but I don't think it works as a lead here.
There are some weird harmonies happening here and there. Some of it sounds intentional and unsettling, some of it more newby. I'm not gonna go over it all. Some of it might come from long releases and echoes on the instruments, some from mashing the sources together.
The middle section or whatever, from around 1:40 is a welcome change. It goes back into a rather weak main section after that. Dynamically doesn't make much sense. I'd try either building up the middle section, or kicking off the main section with more energy. The middle section also has the loud pad problem (which btw might be a loudness or just a frequency range problem; many ways to look at it and solve it), and I'm wondering if the accompaniment there should be an octave lower, as there are a lot of instruments competing for the high range. I can hear a lot of high-range shimmer, wondering if that's an fm synth you're using for some of these sounds. In any case, the sounds aren't bad (except maybe the lead), but they end up competing, and there's a bit of a gap in the middle of the frequency range.
There's a lot of good stuff in here, but a lot of things to improve too. Those simple, short synth blips here and there, the more rhythmic ones, I like. Very atmospheric. I also really like what's happening at 2:38-2:43. The lead coming in again isn't necessary, and if it just did that first note there, I wouldn't mind. But it's not a sound that works as a standalone. I'd try using the piano for this melody, or leaving it to just one note on the pluck lead. Or having found a different lead, maybe that'd work better. I'd try stuff.
TL;DR: Good stuff, bad stuff, nice sounds, lead doesn't work, instruments competing... wait, opera?
-
https://ocremix.org/album/96/golden-sun-a-world-reignited
I should be in there, among arrangers I guess. Just noticed I'm not.
Same on vgmdb.
-
Yeah, it inadvertently turned into a bit of a time capsule. Clem, dude, mad props for your series of album reviews. :D
-
I wrote a bit on feedback in my remixing guide, back when. It's less about giving good feedback and more about identifying it among the feedback you get, but it might still be useful to you.
Quote-good feedback-
I'm sure I've said it a few times already, but let me reiterate: listening to other ppl's music and putting into words what you think about it, the issues you hear, and the solutions you can come up with is great, both for your own artistic growth and for that of your fellow artists. By providing useful feedback you can make them feel familiar with you, even indebted to you so they'll respond in kind with feedback on your remixes.
So what is good feedback? The ocr staff has put together a checklist of common problems with mixes submitted to the site. The checklist lets you mark what issues you hear in a mix. This means you have some help in finding (or avoiding) those things, and suggests things to listen for (eg clipping or noticeable compression). As you use it, you can learn to hear these things even when not prompted by the checklist, and to avoid those problems in your own mixes.
Good feedback points out the problems. It can be offered in a very nice tone, or... a not very nice tone. It can be detailed and elaborate, or succinct. It can address everything, or just one thing you need to hear. It can encourage you, it can challenge you, it can even offend you. How you feel about it isn't what matters. It doesn't have to point out what you're doing right (tho that's usually appreciated). It's doesn't even even have to point out what you're doing wrong. Ultimately, good feedback tells you what and/or how to improve.
Another way to define good feedback is to define bad feedback. Incorrect, incoherent or irrelevant feedback is bad feedback. Imprecise isn't good, but it's better to hear that "your mixing is bad" than to not hear anything about the mixing at all. That's assuming the person offering the feedback doesn't get his terms mixed up and meant your writing instead. Inexistent feedback also counts, as you won't even learn whether random ppl like your mix or not if they don't say anything. Praise (even sincere, if excessive) is also bad feedback - hearing that your mix is the greatest thing ever does not help you grow as an artist. I mean, what do you learn from hearing that you know everything?
You should always question the feedback you're getting, if all you're getting is "it sucks" or "it rocks", ask why, ask what specifically sucks or rock, ask what you can do to improve it. Don't take anything at face value, question everything and stick to what makes sense. Try to understand all feedback you're getting, even if you don't agree with it or know what the big words mean. And nothing's stopping you from reading feedback on other ppl's works and trying to hear the things they hear.
Repeat: try to understand _all_ feedback.
The checklist mentioned above.
I think good feedback comes from good listening. This means understanding what you're listening to, understanding the artist's intention but also a typical listener's reaction and negotiating some kind of useful response out of that. Knowledge, whether music theory, audio engineering, performance, sound design, mixing, music history or anything else is also useful, so learning any of that will help.
Be aware of the artist's intention. On this site you might come across releases, works-in-progress, experiments and all kinds of things, and some of it is made with to suit ocr's standards and some of it isn't. And elsewhere on the internet ocr's standards aren't relevant (the vgm interpretation stuff anyway).
Then comes the psychology of how to actually deliver the feedback in a way that's constructive. I've screwed up on this a few times (apparently the word mediocre means bad), and you will too, probably. Don't tell the artist what to do. Offer your perspective. Suggestions are fine, but be more descriptive than prescriptive. There's a saying about how usually when people say something's wrong they're right, and when they say what's wrong they're not. I try to offer multiple solutions when I identify a problem, as in "you might be able to solve this by EQ-carving some space in the other instruments, or side-chain compressing them out of the way". That gives the artist options to consider rather than directives to obey.
If you make music (I haven't seen you around), think about the feedback you'd want on your mixes, and how you'd want it delivered. And then write it a little softer, a little nicer than that, because tone is difficult to convey in text.
I'm not exactly in my best head space right now, so this might not be entirely coherent. I hope it's still useful.
-
26 minutes ago, audio fidelity said:
looking like the next sd3 release
It's not that bad.
Not yet.
-
What the... again?
-
Dude, you have every reason to be proud of this one. It's great. It always puts a smile on my face when I hear it.
-
Sounds like a blast.
One of the ways to tell that you've improved is being able to tell that your old work was worse. There's always more to learn. But it's easy to forget how far you've come already. Enjoy your music making, dude.
-
Long-time necro post! I have no memory of this thread, but now I want to hear your old song. Do you still have it saved somewhere? It'd be fun to compare.
-
Same as Jorito on point 1, although I won't just take the lead melody, but other elements in isolation. A cool bassline can be a lead melody. Or vice versa.
I'll then iterate on whatever elements I want to use, over original backing, or in isolation. It helps to be able to play an instrument here, even if you're not very good at it, as it's much faster to iterate on the parts you're using that way. Just improvise, jam, mess around, and you'll come up with a lot of cool ideas that you'll want to incorporate into an arrangement. This is why my arrangements end up a bit loooong.
Then it's just a matter of tying them together in a way that makes sense. Sometimes I do that well. Sometimes... not.
Making "originals with stolen melodies" (great phrase) is a good way to break away from the structure of the sources. You can start from a rhythm not found in the original and add elements from the original to it, and adapt them as needed, or write new material to support them. A new rhythm goes a long way to preventing a remix form being too much like the original, even if chords and melodies are the same.
- Jorito and Rapidkirby3k
-
2
An OverClocked Christmas v.XVII
in General Discussion
Posted
IIRC, there was a Zelda fansite that had a time-based splash image and color scheme, switching between day and night styles based on the user's clock. If you want a fun but not too complicated js coding project, you can do that. Though maybe not a day-night cycle for AOCC. Making the colors change with the seasons and turn wintry as the next release gets near would make sense, but would take a while for any changes to show. Still, it could be good practice.