bitflipper

I don't generally use multi-band compressors (prefer dynamic equalizers) but I've had good results in the past with the Sonitus plugin. A pretty UI isn't everything. I have FabFilter Pro-MB with its colorful visuals, but the Sonitus multiband is quicker to dial in. Sure, I get it when people refer to the Sonitus suite's look as "dated", but if it works it works, and a good design is still a good design 20 years later. Nobody turns their nose up at a '59 Les Paul because it looks dated.

I'll be curious to hear about peoples' experience running Win11 in a VM. As a software developer, I'll have no choice but to start experimenting soon, but there's no way it's going near my production machines unless safely tucked away within a VM.

$5k for a new computer is a small price to pay for a center-justified taskbar. Ever notice Microsoft's surprisingly consistent pattern of alternating good and bad versions? Windows 2 was useless, Windows 3 changed the world. Win 95 made half your applications stop working, Win 98 addressed those problems. Windows ME, 'nuff said. XP was great, Vista sucked. Win 7 was troublesome but ultimately recognized as forward progress, but Win 8 was widely ridiculed for trying to make desktop displays look and act like tablets. Then Win 10 redeemed the brand again. Let's hope Win 12 swoops in to save the day.

Yup, CW thinks the project ends at 5:27, even though the last MIDI event (pedal up) occurs at 2:54:17. The longest clip contains a single MIDI note at 2:32. Clearly, the clip length as defined in the cwp file does not reflect the actual distribution of events within the clip. This is not an uncommon occurrence. This is because within the project, clips are their own objects with their own start and end times not necessarily correlated to the data within them. You can see this via the following experiment. Create a new MIDI track and open it in the PRV. Draw in a note, and watch a visual representation of the clip appear in the track view. Add a second note and see that the displayed representation of the clip automatically stretches to encompass both notes. Now delete the last note. The apparent length of the clip as visualized in the track view does not change. This is a routine performance optimization - you wouldn't want the program to have to scan the track to determine its effective endpoint every time you make an edit. Plus sometimes you'll intentionally want the clip to extend beyond the last note, e.g. so a long release doesn't get truncated. Fortunately, there is an easy solution: the slip edit. That's what the feature is for, to let the user determine where the clip starts and ends irrespective of the data behind it. I did that with your project, no problem. Note that it can be difficult to get the slip-edit cursor(s) to show when the track is minimized. Try vertically expanding those too-long tracks before hovering your mouse over the trailing edge of the displayed clip to see the cursor. Drag the end of the clip back to where you want it to end. For the most part, however, trimming clips isn't necessary before exporting the entire song. Just use start and end selection points as I described earlier in this thread. If a long clip bothers you, use a slip edit and bounce.

May sound like a dumb question, but have you verified that there is actually data in the source files? I've been sent many blank files over the years.

Question, resolution and self-deprecation, all in a single post. Well done! All it needed was one irrelevant and unhelpful reply to complete the perfect thread.

We need to resist the normalization of normalization. Leave that to the statisticians, relational database designers and anybody still using a slide rule. Tangents are good. The most valuable lessons I've learned about electronics, recording, mixing and digital audio were acquired by accident while stumbling down some tangential path.

I looked at the project. Before I fix the problem, tell me where the observed project end time is and what time you'd like it to be. I show the track labeled "Split Note F#3" to be the longest, ending at 5:26. Is that what you see?

Well, you've piqued my curiosity so I had to go over to Bandcamp and give a listen. There's still plenty of bass in there, particularly the kick drum. It actually sounds pretty good, though. Maybe the answer lies in compression rather than more filtering.

On the rare occasions when I use synthetic drums, the only vanilla processing I apply is EQ, often to reduce that "unnaturally hyped high end" you describe. More often, the fx will be things that make them sound even more unnatural, such as delays, reverb, distortion and modulators. Acoustic instruments are far more tonally complex and dynamic, which lends themselves to treatments that either highlight or hide the many overtones that are in there, and/or their dynamic characteristics. Electronic percussion just doesn't have that depth. So acoustic fx tend to be subtractive in nature, while electronic drum fx tend to be additive. Also consider combining electronic and acoustic drums. You can end up with an instrument that sounds like it might exist in the real world, but with an interesting twist. The classic example is mixing an 808-style gated sine "kick" under a real kick drum sample, for a deeper low-frequency component. But you can do the same thing with snares and toms. There's no rule that says electronic music must be 100% electronic.

treesha's lovely jam is indeed a testament to the precept that free instruments can be creative catalysts. Reminded me that I recently dug out an old freebie that I'd never used, the Janggu. It's a traditional Korean bongo-like instrument, part of a collection offered by the Seoul National University. I threw it in as an experiment, just because I wanted some non-standard percussion on a song. It surprisingly changed the direction of the composition. I had originally discovered that instrument thanks to a thread similar to this one on the old forum.

Yeh, there's that; no point in sharing an articulation map for a customized instrument. Or, for that matter, an instrument that isn't widely used. I've begun work on an articulation map for an older library, Kirk Hunter Concert Strings 2. I used to love this library but haven't used it in a while, mostly because I rarely need its level of detail. I'm more likely to reach for Amadeus Symphonic Orchestra, which isn't nearly as deep but sounds just as good. My thinking is that if I had articulation maps for CS2, I might start using it again.

It's a logical detour in any discussion of freeware. You can't talk about freeware without acknowledging the reason commercial developers offer it in the first place: to encourage interest in their paid products. Granted, there are some great freebies out there that were created by dedicated hobbyists and altruistically shared (e.g.Thomas Mundt's Loudmax limiter). But you have to sift through a lot of klunkers to find them, which is why this kind of knowledge crowd-sourcing remains such a longstanding staple of recording forums. And you cannot confidently pronounce a freebie as useful without comparing it to its commercial alternatives. The basic premise of the whole thread, as stated by Starship Krupa, is the belief that "a person can put together an excellent system entirely with freeware". He can make such a proclamation only because he has extensive experience with both free and non-free software. So yeh, talking about commercial software in a freeware context is legit.

It's true that articulation maps are most helpful in orchestration. So useful that they actually make composing and arranging more fun and less tedious, and thus inspire greater experimentation. But that's not the only use-case. Any virtual instrument based on strings is a candidate for AM joy, especially faux guitars. Another application would be the more sophisticated voice and choir libraries that offer articulations beyond basic oohs and aahs. Speaking of being scared off by excessive complication, sjoen's mention of pedal steel reminded me of the time I decided against buying a pedal steel VI for that very reason. The demos sounded great, very expressive. But making that happen required some deep articulation switching that didn't look fun at all. I might have to revisit that decision.

Don't normalize. That should be printed onto a t-shirt or wall chart for easy reference. The exception is when, like in Aidan's case, you get a bunch of files that are so quiet that they need extra gain just to get into the ballpark of the rest of the mix. Technically no different from turning up the gain knob, just faster and more permanent.

Cakewalk doesn't actually have to "recognize" the RealiWhistle library, just Kontakt itself. Any library you can load into Kontakt will be fine with Cakewalk. The only time you'll have problems is if the library requires a newer version of Kontakt than what you're running, or, rarely, a library that needs more memory than your computer has.

Some of the Melda plugins haven't turned out to be as useful as I'd anticipated. Spectral Delay is one of them, along with MMultibandChorus and that multiband autopanner, forget what it's called. And speaking of compressors, MModernCompressor was frustratingly unintuitive, and even after figuring out the UI was ultimately just another compressor. However, the good ones more than make up for the few disappointments. MDynamicEQ and MSpectralDynamics are surgical life savers.

Just curious. Articulation maps are the coolest addition to Cakewalk since, um, maybe drum maps twenty years ago? Possible reasons: They're so simple to use that no discussion is needed They look too complicated to get started with Building art maps is so time-consuming that they're treated as trade secrets by composers Users have spent years memorizing keyswitches and are too proud of that accomplishment to switch Everybody's into death metal or EDM and don't need no frickin' pansy-***** articulations "What's an articulation?"

Neither naïve nor irrational, just prudent. Most of the time, the kick and snare tracks being in stereo is an unintentional mistake, and most of the time the left/right channels are actually identical. So there'll be no tonal difference after conversion. Yes, you do get a volume boost, but that's why I do the mono conversion first, before starting to balance the kit pieces. In the rare instances when a track sounds thin after conversion, that's an indication that it really was recorded in stereo with two or more microphones, and that the engineer did not take phasing into account. Leaving such tracks in stereo because collapsing to mono makes them sound worse just pushes the problem back, as the overall mix will likely have mono compatibility issues. In this scenario, rather than using the "convert to mono" shortcut, split the stereo into two separate mono tracks and either delete one of them or phase-align it. Though such a scenario is fairly rare on drums, it is common with synthesizers that insist on stereo output whether the patch calls for it or not. Like you, I fear degradation and will often leave the track stereo for that reason. However, before making that decision it's helpful to check if the track is really stereo to begin with. Quite often, such a check reveals that the patch isn't stereo at all but just has a widening effect (chorus, reverb or delay) added, in order to make it sound stereophonic. They design patches that way so they'll sound good in isolation, without regard as to how they'll fit into a mix. If, however, the track is truly stereo (e.g. a Leslie, auto-panned pad or acoustic piano) and needs to stay that way, then a true stereo panner (such as Boz's Pan Knob) can save the day.

An added plus: all the free fx are feature-limited versions of their paid counterparts, so it's a great way to experiment and see which ones you might be interested in upgrading to the full version. I've bought a few Melda products as a direct result of using their free versions and liking them. (OK, I admit I've purchased more than just a few Melda products, ~40 of them at last count. It's quality stuff and very nicely priced when they go on sale, with a random selection of 4 of them half-price every week.)

I've been lobbying for years for start-of-song and end-of-song markers. These would work like regular markers, but have an additional flag and checkbox in the marker dialog, and maybe be shown in a different color. Beyond that, they would behave like any other markers: show in the Marker List, be moveable, name-able and delete-able. In the absence of such a feature, what I do is create my own start-of-song and end-of-song markers and label them "Start" and "End". Then, when I export, I place the cursor at the Start marker and press F9 to make that the start of the timeline selection, then move to the End marker and press F10. That way, I can precisely control where the export begins and ends and it'll be consistent every time. If I listen back to the exported file and notice it's truncated (e.g. reverb tails or ambience being cut off abruptly), I can easily nudge the End marker a little to the right.

Here's my drum-track organization practice. Bear in mind I've only been doing this for 50 years, and only 20 of those on a computer, so the process is still being refined. I like to place drum tracks in the order that I initially mix them: kick, snare, toms, overheads, hats, room. When I get tracks from someone, they are often all in stereo, so the very first step is to convert kick and snare to mono. These all go into a "Drums" folder, along with any other percussion instruments, and all are routed to a Drums bus. (If I'm using MIDI drums, all the MIDI tracks go into their own "Drums MIDI" folder so I can easily hide them when I'm done fiddling with the MIDI.) Depending on the genre, there will also be one or two additional busses for distortion and reverb. I like to send a little kick, snare and toms to the distortion bus. Unless I'm using reverb as a special effect, it usually works better when you apply reverb to the entire kit via a separate bus. Each drum track will have its own EQ, sometimes a limiter and/or compressor on just the kick, sometimes a compressor on the snare, and usually compression on room mics. Dynamics here are generally just for shaping hits and trimming excessive peaks. All this is done very early in the mix process. Yes, it's common wisdom that you don't put fx on individual tracks before mixing, but for me drums are the exception to that rule - get them sounding good on their own first, then tailor the rest of the mix around them. I prefer to do the main compression and limiting on the entire kit at the bus. Any volume automation will also be done at the bus. This helps to maintain cohesion between the individual drums. Once they've been balanced between themselves I generally don't touch them, preferring to treat them from then on as a single instrument. It also makes it easier to balance the drums with the rest of the mix later on, or to export them as a stem.

^^^ Well said, Bob. Input/Output buffering isn't the only source of latency. Some plugins necessarily introduce large latencies due to the way they work, and should be applied only after tracking is complete. If you need to add another track late in the mix process, that's the main reason the Global Bypass feature is there. Btw, the reference to "oven" comes from having interface clock circuits literally placed into a warm box to prevent the oscillator from varying due to heat changes. This makes the clock frequency very stable, and is the reason internal clocks outperform even very sophisticated external clocks.

I might just give it a go. Once I get over my current phase, which is 70's style synth anthems. I blame it on Wookie. Every time he posts a song, I get all nostalgic for vintage Larry Fast and Vangelis. That'll pass, though. I was thinking about 60's surf next, although that's challenging for a piano player.

Just did my first gig in a year-and-a-half. It wasn't my fingers that hurt afterward (they've been working out), but my back (which has not).