Glenn Stanton

On 4/26/2021 at 11:07 PM, clovis said:

They did it, they are the winners of the real time remote collaboration within any DAW, both audio and MIDI, w/chat capability.  BandLab should create a plugin for a DAW to make use of its mix editor app!

actually i'd call "near real time" since there is latency etc which is solved by keeping copies of the work-in-progress in each client connected to the session rather than a true live streaming (where the latency across tracks would be readily apparent). so kudos for making this work!  once i saw this, i thought the same thing - BandLab mix editor as a VST... just add your favorite video/chat product and you'd be good to go (assuming the BL does a trick similar with the near-real time caching across clients)

fair enough ? i'm drinking my morning caffeine from a 30 year Tupperware mug which is still going strong ? i have a bunch of the Glaadware so i'll have to try it. one plus is that they're easily replaced and much less costly than a set of new drum heads ? 

coolio - i like the idea of the tupperware to introduce velocity ?

5 hours ago, Marcello said:

... I don't understand when people say the bass and kick should be mono, but isn't stereo panned center the freaking same?

yes and no. the center panned stereo phantom image is "essentially the same" as a mono track except that any stereo pan laws in use , phase differences, etc can impact the levels whereas the mono buss/track will enforce a consistent phase and energy level on your mix decisions. a mono pan law in a stereo master field can also be impactful, but you've already addressed any potential phase and level issues by forcing the mono.

you can easily test this yourself. setup your bass and kick through mono source tracks and a single mono buss. switch the tracks to stereo and back - what did you hear/measure? then set tracks to mono and switch buss between mono and stereo - same hearing and measures (phase scope)? 

note: most times, instruments are mono sources like vocals, so unless you're stereo miking the instrument (e.g. acoustic guitar, a three mic drum kit setup, M/S etc) just using mono tracks and audio clips will be best.

thanks! a lot of good information there! one quick note on sources - i find that when i need a quick and dirty "human performance" on drums, i record a few tracks with me just tapping on my laptop to perform bass drum, snare, hi-hat, percussion beats (one per track as the "notes" on the laptop case are pretty monotone) then use Melodyne (or could be any audio to midi convert) to generate a single note MIDI sequence, copy + paste + drag MIDI to appropriate drum(s) on my AD2 midi track, or just use the drum replacer.

fast and very human way to get drums going without using anything extra like my bongos or djembe.

yeah, for something like this i would use Ozone and Insight to get the levels correctly set. and RX for noise reduction etc.

well there is a normalization process but its peak level not RMS. there is also the silence process which can readily trim down below a threshold and split clips (or not) depending on settings. 

same here - prefer insert via parameters versus dragging. and not just at the beginning but anywhere i'm inserting across the board.

maybe the answer is jpeg ?  ? 

hmmmm - 64-bit plugins? did you try a different driver (e.g. like WASAPI instead of ASIO)?

maybe a re-install of the software could fix some missing bits or runtimes.

if you're using them during the mix phase - bump up your audio IO latency to a high number - this will significantly reduce the CPU load. once you have a track sounding pretty much as you want it, freeze it to save even more processing power. this should help.

one thought - use automation to "write" the setting as a very short envelope, then turn off write. the gain will bounce to the level you set. adjust the gain, then once happy, enable the automation write, turn the gain know a bit to lock it in, turn off write.

maybe just use two mono tracks. i just did a recording this weekend up at Woodstock and just two mics and two mono tracks per performance, sounds great and i can also adjust some of the levels between the tracks to change the closest instruments to a given mic. 

hmmm. are the audio latency set as high as possible? this will help ease CPU impact from the synths. if you're trying to run soft synths, versus freezing them, those will be very high resource consumers. presumably you're stacking all the same soft synths in Ableton with the same latency settings?

6 hours ago, Cristiano Sadun said:

The hardest bit would be to find a GUI design that allows the user to pick up the points and the curve he thinks fit better in a smooth and effective way.

agreed, never let the computer decide ? 

pretty sure if the simplification simply remove redundant nodes within a given % of the adjacent ones, that would seem straight forward like the video. if you're asking for it to create curves based on the node pattern as showing in the first post, i'd suggest that is way less intuitive for the programming to make - my interpretation of that would have been a bowl not a descending arc to a point... also, i wouldn't want my gaps (the dotted sections) to be automatically filled since i usually mean them to keep the last position steady until next change without needing to consume any automation processing.

yeah, the value-added tax schemes are probably the most corrupt/evil by any measure. basically the government applies the VAT to every step, so it's often 35-40% by the time a consumer gets it - with most "built-in" by wholesalers and distributors. and because many of the VAT layers are not readily seen by the buying public, the governments adjust them all the time (you can guess which direction). this not only cause a competitive hit on the markets but effectively devalues the currencies involved. (as a disclaimer had the professional experience of building systems and as a result too much info on VAT schemes...)

so for anyone thinking in the US that VAT is the way to go to "simplify" taxes, just say NO.

it's not so much complicated as it is preparation work so your projects are in the best condition possible when sharing them, otherwise the recipient has to do all the work, and they could get it wrong resulting in a lot of wasted time. plus, you're likely to do a bunch of these steps when it comes time to archive a project. knowing if you rehydrate it, you'll mostly have things organized in a way that helps you get back to mixing instead of patching up a mess...

hmmmm. i use BT speakers and sometimes are headset with CbB many times over the past couple of years. if you have them installed as Windows devices such as audio outputs, recording sources, etc then they (generally speaking) can be seen in CbB using WASAPI or ASIO4ALL (in this latter one i find it annoying that it sometimes reverts to the Bose headset when i had set it to the USB audio, but its easily changed in the admin console).

congrats on the new specialized high intensity training (S.H.I.T.) job! ? 

i tried this about 8-9 years ago with a fairly large server-size system (HP DL w/ 160 cores on 8 CPU , 2TB RAM, etc) and oddly enough it had a built-in audio card. running Windows Server 2012. so since the $50K machine was sidelined from a project, i loaded up a VM using SONAR PE (X2) and while i could not "hear" via audio card, since it was present in the VM, i could hear and mix via remote desktop (via 1gb hardwired network) from my laptop.

BUT (always a big but somewhere) it was flakey and had the crackles etc you'd expect on a low end PC, and even with a bunch of tweaking i could not get it to be happy. maybe the Windows 8 on a VM was not happy. whereas my Dell 545 (still running today - 3 years ago updated w/ quad core cpu , ram, and OS) was perfectly fine with the same software, OS, and general settings. 

maybe with more modern OS and Hyper-V updates it will work.

if you're planning on sharing - some preparations are needed to do this smoothly:

make sure your audio clips are bounced to clip and extended (not dragged!) to 1:0:0 mark so the person receiving them can align them to the zero mark. yes, you can export them individually as broadcast wav which have the time markers embedded, but what if the recipient cannot use that? so zero mark is best.

if you know the person on the receiving end will not have the plugins needed, print a separate track with the effects and then remove them from the project once you have documented their settings.  make sure any plugins which are yours exclusively and not likely to be available (easily) for the other person (and which you determined are critical to the mix) are documented. Word is a good choice with screenshots.

make sure all the clips and associated file names are understandable. it makes it much harder for someone to sort through clip#99 clip#A mix_clip#5 etc

create a copy of the project with it's own audio file directory. this (check the copy audio files box) makes copies of the audio clips and only the ones you're actually planning to share and not a variety of clips which will be confusing to someone needing to patch them together. this is where you can export OMF file as well - open the newly saved-as project and export.

ensure the person you're sharing with can read your files at the sample rate and bit depth of your project. it won't be helpful if the person getting your files can only support 44.1K @ 24 bit to send files which are 96K @ 32 bit... if not, convert the audio files to the required sample rate and bit depth.

export a "draft" mix (or demo/reference etc mix) so the person receiving the package have a rough idea of what you're doing at the time you sent it...

then zip the new project directory. this captures the project file, OMF, Word doc, rough mix, and audio clips into a single blob which is universally accessible. the recipient of the package can use the project file or just import the audio clips into the DAW of their choice.

now share the blob via USB drive, or via cloud services.

of course, if the person you're exchanging with use the same version of CbB, then you can skip the OMF and leave any shared plugins in place.

feel free to have people sending you their projects take the same efforts to package up their work before sending it. ? 

sorry for your loss. it's actually only 5 tracks - drums, bass, guitar, keyboard, and vocal. so no doubled parts, maybe it's the constantly shifting tempo and some flanger (guitar) and chorus (bass) causing the pitchy bits (not discounting my singing of course!)

yeah, it is a bit sloppy in parts. mostly because i started out testing the tempo mapping by dragging the guitar (played without click or other parts) onto the tempo bar, then adding a midi drum pattern and playing the other bits.

so why would you have to drop CbB if you buy into PT? i'd go out on a limb and suggest many people on this forum (and many more not on the forum) have at least 2 DAW they use, and probably many have 3 or 4 DAW products (and likely all kinds of audio processing tools as well).

for myself (having 6 DAW products), if i get audio files and no restrictions, it's CbB. if it's a mostly mixed PT project, then i open it in PT, save in PT, and re-share it back as a PT project. same for Cubase, FL, REAPER, Audacity, Studio One, etc etc... it would be silly to use a framing hammer to assemble picture frames when an air powered brad gun would be perfect... the right tool for the job is the right tool. smae for plugins - i need both VST and AAX so investing in products which support both eliminates redundancy in the plugins i use and the breadth of knowledge needed to use them across DAW...

if you're making a living off this stuff, then investing in the right tools and acquiring the skills to use them is part of the gig...

i'm guessing but the Park Control app you're using is made by the same people who make Lasso ? Lasso is so much more. however, it's mainly for technically inclined folks who want to get deep under the covers of CPU affinity and many other bits. yes, if the CPU is kept full on, full power, it will be more heat than allowing the OS to manage the CPU utilization.

12 hours ago, Blogospherianman said:

+1 for Music Rebalance.  I like to render to 4 separate stems via rebalance (when multitrack stems are unavailable).  Allows to have Vox, Bass, Percussion and Other separate for different treatment and/or balance.

yep same here. plus with RX7 (or later, perhaps earlier as well) - de-reverb, de-noise, etc all help.

on the simpler "synthetic" stems - capture a MIDI of each from Melodyne, use it for soft-synth replacements, and just keep the vocal or lead solo ?  (sort of like using new paint on a Sistine Chapel ceiling when needed ? )