bitflipper

The reason this doesn't affect RMX, Trilian and Keyscape is that the root cause of the problem was a change made only to Omnisphere, in which a 9th output was added. I have no idea why they did that, since you can have at most 8 individual voices in a Multi. It means that when you created your project prior to the Omnisphere update, CW saw that it had 8 outputs and calculated all the other instruments' input source offsets accordingly (think of them as being numbered 1...N internally; inserting a new one throws the whole list off). Then, when you subsequently re-open that same project all the offsets are thrown off by 1 (or the number of Omnisphere instances). Kudos to Noel for figuring out a fix quickly. Know that this was not an easy fix to implement. As a fellow coder, I can testify that our Noel is a frickin' bit-ninja. [EDIT] I was wrong; Trilian was also affected. See below.

There are less-expensive mastering tools out there, Marcello. They won't be do-all suites like Ozone, so you'll have to run down individual plugins for EQ, compression, limiting, phase analysis, saturation, and all the other cool stuff Ozone does. For example, there is Limiter No. 6. It's free and works well, with its only downside being it's complicated and not very user-friendly. Which just means its learning curve will be steeper than Ozone. But think of it as "sweat equity", trading your time to save money.

Thanks for taking a closer look, which I was too lazy - er, too busy - to do. The pictures certainly reinforce what I heard: a scooped spectral imbalance. First thing I'd do before commencing any mastering or master-bus processing at all, would be to do something about that 2K bump. The OP is probably being held back by his monitoring environment (headphones?) while mixing. That's where spectral images come in handy, showing you what you maybe can't hear well. I'd also note that this type of composition can't be evaluated from a whole-song analysis, because it's sparse in places and full in others. OP picked a tough song to first try mastering on. I'd concentrate on the loudest part of the song when setting levels and dynamic reduction, as the quieter parts may well fall into place naturally.

Your DIY master is definitely better. The automated one sounds dreadful, with obvious compression artifacts and distortion.

No, if you're doing it right (i.e. not smashing the bejeezus out of it with compression) the two peak values will often be slightly different. As long as they're within 3-6 dB of one another, it's unlikely to indicate a problem. The tiny 0.9 dB difference shown in your screenshot will not be audible. That's for peak values, though. RMS values should be fairly close. If there's more than, say, a 3dB difference, the mix may sound unbalanced to the ear. In that case, you'll want to review each track's pan position to make sure left and right channels are equally represented, such that they sound balanced to the ear. Headphones are good for that. In your screenshot you show a mere 0.4 dB difference, and nobody's going to notice that.

I ask because the Kontakt universe offers an amazing number and depth of virtual instruments, ranging in price from free to several hundred dollars. But the free and cheap ones require the full Kontakt product, not the free player version. Kontakt itself comes bundled with a lot of sampled instruments, including an acceptable French horn.

^^^ Excellent point, Cactus. Excessive sub-bass can make mastering difficult. I wouldn't expect an automated mastering tool to be able to handle it.

Or, if it's no big deal to simply detect a change, just a warning messagebox would have saved me a week of head scratching. Like "WARNING: This version of _______ may be incompatible with this project."

Sorry, my bad. Guess I figured everybody here has Boz on speed dial in their Favorites bar. Das Boot Le Snappet El Clapo $15 each, $29 for the lot.

Yeh, Larry, this has already been posted before. But the sale price is back on again until the 29th and I couldn't find the previous post to update it. I bought the bundle the last time it was on sale. It was merely an impulse buy, like grabbing gum at the grocery store checkout line. But I have gotten a LOT of use out of these simple one-trick ponies. Here's a tune that uses only these for percussion, plus another favorite cheapie, Skaka by Klevgrand (sadly not on sale atm, but keep an eye out). https://soundclick.com/r/s8hb75

And "how do I plug it into my iPhone?"

Which in turn is a stripped down version of anything we'd have called a sampler post-20th century.

Yes, it is entirely possible to create a mix that's too dynamic. We just don't talk about it as much because it's not nearly as big a problem as over-compressed music is. But you do have to think about where the listener will listen to your music. In the car barreling down the freeway, on an airplane or a train, running on a treadmill - all way more likely than someone sitting in a quiet room listening on a high-end hi-fi.

I have to imagine CW isn't the only DAW that works this way. Could be that Spectrasonics just tests Logic and Pro Tools and calls it a day. But QA aside, I am surprised that the coders didn't anticipate this, as changing a DLL's interface almost guarantees broken backward compatibility. That's Programming 101. Then again, the whole reason for VST's existence is to implement standardized interfaces, so the programmers might be forgiven for forgetting that rule. OTOH, that's why QA exists - to test if what the programmer thinks will happen really does.

Checked the release notes, and sure enough there it was, the smoking gun: I asked Noel if changing the number of outputs in a VI update could corrupt a project, and he said yup. Bottom line: do not reopen any Omnisphere projects created before June 2021 and you'll be fine. If you must revisit an old project that has Omnisphere in it, be prepared to perform extensive repairs. And if one of the affected tracks is frozen, leave it frozen to avoid subsequent crashes.

Interesting observation re Spotify. I'm not a Spotify user, so I can't confirm, but I've heard that they don't treat all artists the same. Could be something to that, I don't know. Spotify, however, is not a big moneymaker for most artists. I can tell you that major artists don't break the rules if they want their music heard on radio, TV or motion pictures. Well, first let's be clear that artists themselves rarely have any say in the matter. There will often be separate masters for different applications. For example, the movie soundtrack will typically have a wider dynamic range than what's broadcast on radio/TV. A point that should be made here is that it's not just about average levels (integrated LUFS), but also dynamic range. You could, in theory, have a master at -6dB loudness but still be sufficiently dynamic that even if playback is automatically adjusted downward, the mix will still sound good. Oh yeh, and those adjustments can go both ways; if your master is quiet the streaming services will turn it up - and it'll sound better than one that was turned down to match. Yeh, it's complicated. There's a reason folks pay specialists (MEs) to handle this stuff.

"Too loud" is really a subjective thing, and somewhat dependent on the genre. Many popular releases are far louder than -8.3 LUFS. But streaming services and YouTube will automatically turn them down, so for them yes, it's too loud. For a CD, no problem. For me personally, that would be too loud. I usually shoot for -14 for my own stuff. I'll go as high as -12 (integrated, -10 short-term) for aggressive rock that I'm mastering for someone else. If -14 sounds too quiet, turn up your monitors. Seriously, that's how it's done. Google the K-system and speaker calibration, it'll be a game-changer for you.

If it's intermittent, does it depend on where you stop playback? If you happen to stop playback after a pedal-down event but before the corresponding pedal-up event, the piano will keep playing the note until it decays.

Do you have full Kontakt?

My go-to solo violin these days is the SWAM violin from Audio Modeling. It's a modeled instrument, not a sample library, and I like that for its small memory footprint and quick load times. One of the things I like best about a modeled violin is the ability to do slides of any length or speed, something that doesn't work well in most affordable sampled strings. Here it is in action. It's a project I made while writing up the SWAM instruments for SoundBytes Magazine. It features several of the SWAM instruments: violin, clarinet, flute, cello and double bass. The string ensemble you hear in there is another nice but very affordable ($35) string library, Arctic Strings from FrozenPlain.

Update: I have now seen this problem in at least six recent projects, with mis-routed audio inputs on many dozens of tracks. All of them had been last opened before I updated Omnisphere. I wouldn't have thought that updating a VI could screw up a project's routing, but that does appear to be the case. In at least one instance, the routing change was to a frozen track, which could not be un-frozen without crashing the synth (Zebra2) it had been connected to when frozen. I was able to salvage the project by removing the synth and inserting a new instance.

Deprecated AND depreciated. Seriously, though, we all have favorite plugins that were officially retired long ago. Nothing inherently wrong with older 32-bit plugins -- as long as they still work. Sounds like this one's got some issues, and Steinberg is never going to address them. I'd starting poking around for a replacement. Sorry, I have no suggestions, as I've never used this specific one. I'm guessing it's a bass instrument with some canned MIDI or loops/phrases? If you're just using it as a soft synth, try SI-Bass that came with Cakewalk.

Much ado about nothing. I wonder how many irony-deprived people were scared enough to relay this information via Facebook.

Sorry, not a pun but I saw this on Reddit and it was just too funny not to share. It's amazing how music can take you to another place. I was in a café, they put on BTS. Now I am in a different café.

Yes. For the most part, you can freely mix 'n match VST2 and VST3 plugins. Even if it's the same plugin, e.g. Omnisphere.