Jon White

On 6/20/2020 at 12:05 AM, LarsF said:

Different daws may set different ASIO latency, and some have special features like Cubase/Nuendo ASIO guard or whatever it was. Reaper and StudioOne add processing buffers apart from ASIO buffer setting etc. So many things may differ in what daw have as load. Larger latency give more headroom before crackles and pops.

Did you look in task manager what cpu load is and what is ASIO buffer setting?

(With and without last plugin, and depending on preset you used)

The example I mentioned it was added cpu, and when at 55-60%(task manager) or so, some crackles can appear. I run 64 samples ASIO buffers, and a 10 year old quadcore i7 (benchmarking at 5000, where modern cpu i9-9900k of today rate about 20 000) so old cpu in this regard.

If not related to cpu, something close to what you suggest - something wrong with audio engine - may be relevant.

I did not bother with 64-bit engine yet, if that matters.

This is another setting that may differ between daws what is default.

 

For me with Cakewalk the default threadscheduling worked the best in that it distribute even on 8 cores(4 physical with hyperthreading give 8 logical).

Another setting made it use 4 cores only(don't remember which one), so changed it back.

 

And my reasoning with what is the largest project that runs normally?

Thinking that some special plugin here really raise cpu. It could be running a mono plugin on stereo interleave track or other way around. Page 191 in manual mention some special settings that are in plugin manager in how mono is handled.

 

Overall thinking is that some default settings in Cakewalk does not make optimal result on your system.

What a thorough and detailed response.  Thank you!

I'm always at 48KHz/24-bit/128 samples.  I use the Lynx ASIO driver that works with my soundcard.  I use two of eight stereo outputs and inputs on that interface.  All of my drives are SSDs.  My system crackles when the Cake resouce meters show the cores bouncing around 25%, and always fails when higher.  I did find that my CD/DVD port was causing latency, so I disable that when using Cakewalk (Latency Mon showed me that), but, again, no issues in any other DAWs.

I truly think Cake does something different in some ways that ends up being problematic for the audio engine.  Just years of experience with it amidst other products.  You make great points, though, as do others.

I record upright bass to jazz quite often, and I do it by having my tone "live", as with all my guitar and analog input -- that being the tone that I present to Cakewalk is from a mixer output and already has my commitment to the final voice.  Short statement: I don't reamp.  Now, if latency didn't exist, I might record tones with live plugins in the channel.  For now, no.

SO, I have my preamps and analog boxes dialed in, and then my plugins (two ASIO devices can work at the same time if controlled by different applications) doing the rest. 

Typical guitar for me is BiasFX2 or SGear via my TonePort UX2 (ASIO, 24-bit, 48KHz) and Cakewalk using my Lynx soundcard at the same specs. The TonePort sends its output to my mixer, which sends an analogue signal to the Lynx inputs, and that is the final track sound, overall (but may take on digital/plugin tweaks later).  It takes (pun intended) deliberate commitment and tone planning, but I'm used to that.  The bottom line, so to speak, is that I'm so bad at playing that I need the tone of my guitar, bass, sax, piano to be inspiring as I play, and to be what I see as the final voice.

Anyway, that sets up EQ fashions in the hardware AND the plugins.  I would say that the preamps and stuff prior to plugins shape my EQ stance more than a plugin.  Message here is, maybe: have the analog voice be the stuff as much as possible.  That will be outboard stuff that you would commit to if gigging, right?  Your tone as the "you" being presented to recording efforts.  I learned this in the studio, as well.  

22 hours ago, Craig Anderton said:

Thanks for the props! The most appropriate tools are sometimes the least-expected ones.

Mr. Anderton, I've enjoyed your engagement and contributions to the recording world for over twenty-five years (Electronic Musician magazines that I still have).  It is my privilege to absorb your thoughts and approaches to music production.   I truly mean that.

11 hours ago, bitflipper said:

My favorite for bass is Meldaproductions' MDynamicEQ.

It's a great static EQ, but being dynamic as well, it's useful for dealing with bass resonances. The dynamic action also lets you add some high-frequency boosts only on the notes that need it. These qualities also make it well-suited for vocals.

Killer, killer software.  Great citation, Bit.

22 hours ago, Craig Anderton said:

All DAWs have different preferences. It's unfortunate that you have to be a rocket scientist to make music these days, but I'm afraid that's the case.  Think of all those preference parameters as a combination lock - get the numbers right, and things work...and then there are all the Windows variables. 

I still don't miss tape, though

Wow, yes.  What insight and wisdom you show here, Craig.  Truly.

I had a Tascam deck (4-ch), and it was, although a privilege in the early eighties, a challenge.  15-ips was the blessing, I thought!  But we innovated and got our art down!

What kind of tone do you want?  What does it sound like now that you want to change?  That makes pursuit of effects a bit more deliberate.  Mr. Anderton's suggestion is super.  I use band-based compression most often.  Waves has two good ones.

On 6/18/2020 at 5:34 PM, John said:

Ozone 5 and all Ozone versions are very CPU intensive. In many years past I used it on a mixed down stereo track in a mastering template due to my computer would have trouble otherwise. I still, more or less, do that even though I have a system that is very capable. 

One  trick to try is up your latency enough so you no longer have problems. You can put it back for recording. 

 

Edit to add. A rundown of your system would be useful. 

Thank you, John.  It's a puzzle.  Much of my thoughts on this center on the number of issues with Cake audio when compared to other systems.  I use Pro Tools, too, and it skates through this stuff.

On 6/18/2020 at 2:28 PM, Colin Nicholls said:

So, you're just here to bitch?

Hmm.  What does "bitch" mean?  If I guess correctly, that's what you're doing regarding my posting, correct?

On 6/18/2020 at 4:50 PM, David Baay said:

Any chance you enabled the non-default  'aggressive' ThreadSchedulingModel=3 in AUD.INI (Preferences > Audio > Config File) despite warnings that it's experimental? Many who have tried it had a result like you describe, and many of those promptly forgot they had made the change and blamed the update... as usual. ;^)

 

No, I remembered.  Thanks for the response.

On 6/18/2020 at 5:22 PM, GreenLight said:

Maybe it was the 'aggressive' ForumPostSchedulingModel = 3 that was enabled by mistake in AUD.INI by the OP?

I'm not saying things can't be improved, but manners could. 😉

Thank you for the tip.  I've covered that one prior, hoping for better luck long ago.  I appreciate the response.

On 6/18/2020 at 2:21 PM, LarsF said:

You had this and everybody else must have too?

Last time I had crackle in Cakewalk was when I reached 48 tracks, and sends and stuff everywhere. And this was a mistake due to that Prochannel had activated two plugins on every track, 96 instances.

Removing those I now am on 80 tracks in same project and still no crackles.

I run RME HDSP 9632.

What is your reference with any other daw and tracks and so on?

Maybe your interface is on the verge already doing what you did, and just that bit better working with other daws.

So which interface is it, or your feedback is futile?

 

Lynx audio card.  Again, though, what hits me is that other DAWs have no issue.

Created drum track in Superior 3, guitar raw input from mic'd amp, bass same, then sang a track.  Three minute song, with a TrueVerb bus, Tony Maserati vocal plugin and Ozone 5 in the master bus.

Added VocalSynth and immediate popping and clicking for three of the five types of effect.  Rebooted, tried again, same.

Setup same song in Sequoia, same add of VocSyn afterwards.  Beautiful.  Added VocSyn to the guitar track for fun.  Smooth as silk.

Went back to Cake and added VocSyn to guitar track -- immediate audio dropuout upon hitting play.  Won't even play.

Please consider that asking me for specifics on my resources and all misses the comparison point.  I'm quite sure that Reaper and the wonderful Nuendo would have smiled through this, too.   Cakewalk, you must know something about how you are handling audio in detail.  What is it?  It's got to be fixed before anyone but hobbyists, if even they, would feel comfortable even starting a project with it.  Imagine a studio!

Was Sonar stable as an audio engine?  See if Cakewalk is the same.

Great ideas, all.  It creates "NTOSKernel" (or close, my spelling and caps may be wrong) in the top 3 list of resource burdens in LatMon.  The other two are, of course, NVIDIA resources.  Wish I had another brand of vid card, frankly.  

Just now, Jon White said:

Just one thing I've found in my little corner of the world:  disabling my CD/DVD device on two of my music computers has dropped latency and dumb stuff dramaticaly.  

Although I will say that it was only Cake/Sonar that manifested the problems.  Latency Monitor helped show the situation.  That situation did not seem to make Sequoia, Nuendo or Reaper fail.  Just our baby.

Just one thing I've found in my little corner of the world:  disabling my CD/DVD device on two of my music computers has dropped latency and dumb stuff dramaticaly.  

On 6/4/2020 at 6:11 AM, Starship Krupa said:

You must be speaking of performances that didn't have PA systems with a FOH mix, unless you interviewed the sound mixer after each of these shows to make sure they weren't using a compressor or limiter on the main bus or individual instruments. It's common enough that many consoles designed for mixing live shows have compressors built in to each channel.

No, the point was that these performances are mainly heard directly by the audience.

We have to come to grips with the instability of Cakewalk.  Only then can it move forward.  Finding a million things that might fix audio glitches while other programs work fine is contrary to wise thinking.  All of them work better in terms of audio stabiity.

On 5/31/2020 at 5:47 PM, mdiemer said:

Been following this thread with interest. It's great to get such expert opinions on this rather mysterious subject. I'm currently trying to improve my orchestral mixes, which is really tricky. For a compressor, I've been putting the Sonitus on the mix or master bus. I find that the Vintage Neve or Complete Mix presets seem to work best for me, but in trying to make my own settings, I have not been able to do anything that helps (although I have had great success making it much worse). What did help was this: I took the Vintage Neve preset, changed "type" from normal to vintage, and put the TCR button on. Amazingly, these small changes resulted in noticeable improvement. I have no idea why, but it did help. Just thought I'd share that.

I grew up listening to every genre you could find at Oberlin College's Finney Chapel.  From Dizzy to folks no one here would believe (classical artist prodigies and intelligent rock gods -- although the night a certain complex artist named Patti rocked a deliberate urination out on stage, that was tested).  I was privileged with fun at studios once in college and since have loved recording.  My first Cakewalk was in DOS, I always joke.

I have a conviction.  It is that not one performance I've ever seen and relished and absorbed with human being had compression on the stage image.  Sure, various artists in rock and fusion may have had pedals and such compressing their stuff, and various house mixes of more modern music may have had something going on like compression, too.  But the stage, at least in classical, delivered the instruments without compression and the inclinations behind it.  It was open air and truth in conveyance.  I'd minimize compression in classical as an imperative, and I challenge it as the future in any music once we humans morph a bit more into nature.

I did a resource monitor after noting some heavy hard disk (non-ssd) loading and droputs in Cake, and saw that it was reading every muted audio track during playback.  Each take lane was being read by the drive.  Why is that?  Isn't that a bit of a waste?

Every convention we have now was once formed on a "new day" in the past.  Let your ear bring you new days, and put whatever you like wherever you like, based on the audible result and your personal flair.  Heck, compressors across the board can all have different sounds due to varied phasing and frequency responses, so go for what you think sounds great.  It could be a new day.

(For demonstration of what various compressor plugins do to parameters other than compression, try Plugin Doctor.  Load 'em up and see their frequence and phase character, and other things.  It's vegetable soup -- potpourri.

C'mon, man, check all of your settings.  

What software is playing it as a wav file?  

People come here asking the most unresearched questions.

Try Reaper or something.  If no issues, there you have the state of things.

11 hours ago, mark skinner said:

Sorry , I did not mean to sound sour. I actually still have Reaper installed on another pc. If I ever decide to make it my daw of choice , I will quietly leave and not complain about constantly improved  "Free" software...  mark

Very cool, Mark.  I'm always impressed with what blossoms to be the intelligence here.  I'll erase my comment, knowing that you just tilted a bit to the pedantic!

Try another DAW.