Clicks at the start of a rendered file

Today I experienced a weird situation on Wavelab Pro 10.
I was mastering a 11 track record and when I got to track 6, every time I tried to render it there were some clicks at the beginning of it. At first I thought that it might be just a playback issue, but it wasn’t. The clicks stayed in the rendered file. I didn’t have any issues on the other tracks and the plugins used were the same, all of them inserted on the clip.
I tried to remove some plugins, but the only way to stop those clicks was to remove the external effects insert, where my analog chain is. Of course that wasn’t a solution.
Maximizing the waveform I could see that the clicks were audio interruptions. If I edit the rendered file and cut the 2 or 3 bits of silence, I get a perfect rendered file.
I changed several times the buffer size for my Lynx Hilo but that didn’t solve it.
Any help would be appreciated.

Is it something new? I mean, did you use the external effect before, without this problem?
Also, where is it inserted in WaveLab?

Can this be related to “the Playback Fade in” ?
check
File > Preferences > Audio Connections >
Playback tab “Options”
untick
“Playback Short Fade in When Starting Playback”

regards S-EH

Thanks PG for your answer.
Randomly this issue has occurred since Last December when I converted myself to wavelab after 28 years of being a Protools user.
Every time this happens I can figure that it has something to do with the external effects insert. I forgot to tell you that it never happens in normal playback, just when I’m rendering.
Maybe I’ll just have to change my workflow inside wavelab and use a reference track to feed a recording track with my AD/DA to get the process more stable.

Thanks S-EH for your help.
I’ve never used the playback fade in. But today I checked that box just to see what could happen and it’s still the same.
Thanks once again
All the best!

Im having this same issue - clicks at the beginning of tracks while rendering and using external effects. Im using Wavelab 11. Did you ever get it resolved?

I’m having the same. 17 track LP mastering , all tracks render fine as ‘all selected clips’. Four vinyl sides as CD groups ABCD. Three of the sides B,C & D render with a noticeable click within the first second or two. I got one of those sides to render clean by using ‘union of clips’ instead of ’ CD Track Group’, but not succeeded so far with the others. I’m on version 11, latest Monterey. I have no external chain. user since v10 and I’ve always had to set fades front and back before I insert my plugin chain as the intermittent clicks are unbearable on playback as soon as anything oversampled or high linear phase like Fabfilter goes in. Impossible to tell if it’s the raw file or the plugin chain clicking - hence fade first, process second workflow. This isn’t related to that at all as far as I can ascertain. Never had a click in a render before. Never noticed it do it before when rendering sides like this either. And I would have noticed. So wondering what’s happened differently. Either way up it’s far from ideal for the ‘best mastering software on the planet’ or whatever the claims are.

Actually I was wrong. The clips didn’t render fine. The 16 bits had no clicks - they had linear dither on the master section. One of the 24 bit files clicks after the start = as it does when it opens the side D render described above. These were rendered with the lin dither bypassed on the master section, identical otherwise. First time I’ve ever had any start click on any render as opposed to that playback annoyance . Same version, 1.1 as dozens of other mastered Eps and LPs. And same Moterey as well as far as i can fugure.

Turns out I got that wayward 24 bit master to render clean by enabling the lin dither on the master section, setting word depth to 24 and switching off dither. Never had to do that before. And never dither to 24bit. That didn’t work to get the Vinyl side off clean though, neither with CD track group nor union of clips. Clicked in first second or so every time. I had to duplicate the montage, replace that clip at the start of side D with its 24bit apple cert master, bypass the clip plugin chain , then render the union of clips. Exact same process with the C side. Both those side start tracks had been bounced in the montage as part of vinyl restoration, so were 32 bit I’m assuming, and the replaced 24bit masters solved getting the sequences rendered. My guess is this is a bug to do with high bit rate reduction to 24 bit. It’s repeatable. But I’ve no idea why reducing the word depth with lin dither worked to get the errant individual 24bit master off without the click, but not the sequence of four files following on. Turns out the A side 24 bit cut was the same. Had to redo it as above. Click at the start of the file but not at the start of the 3 following tracks on the side. Kinda potentially lines up with original click post above perhaps, not had a look at the file, but a glitch in starting processing when going for multiple (32 bit?) clips/Cd tracks but not individual clips? Weird. Made my already long day stupidly long. Last orders, need a pint.

Thanks. It’s rambling cos I was trying to get hard deadline clients’ LP out the door late at night. With a major glitch - and altruistically show my workaround in progress… I’ve thirty years experience as a commercial mix and mastering engineer with billboard #1s, as both. I have a mathematical physics degree and I did PhD in AI. I’ve lectured music tech at UK universities as adjunct lecturer since 1995. It’s not the dither plug. The lin dither word reduction solved the problem with the one 24 bit apple cert master render that clicked as a rendered single file. All the other clicks were vinyl comp renders. I’ve NEVER has this clicking on render before. It’s almost certainly something to do with the initial sub rendering i did in the montage with the vinyl restoration. It happens on each and every 24 bit render of each vinyl side of 4 - 5 tracks. A single click and not the same sound on each. The sub rendering to de-crackle is the only thing I did differently in this session to any other of the last dozens of LPs an EPs I’ve done on Wavelab 11.1 - to my knowledge so far.
I’m frankly deeply unimpressed with Steinberg Wavelab. Especially now.
I already have to get the sound right in Tools or LPX as Wavelab won’t phase lock solo swap in sync between reference track, proposed master chain and raw file with just a limiter on. It’s ok with trivial plugin chains but glitches hugely as soon as any plugs go linear phase or oversampled in the master chain. Wavelab is not plugin delay compensating like a tracking DAW does. Unless there’s an option somewhere to selct for this sort of tracking behaviour that I’ve missed?
I use the awesome BC patchwork to get the plugin chain in a DAW that will allow me to solo swap instantaneously between as many refs and proposed master variations as I like, so I can hear the absolute nuances, Then go render full res in a Wavelab session.
Patchwork also obviates the really odd max plugin chain limitations of Wavelab. I regularly have 16 + plugins on in various swap out combos till I’m happy with the chain.

and btw , none of the 16bit renders had a click as individual files - just the one 24 bit render. So the lin dither did it’s job properly and perhaps solved/obviated the issue with that one 24 bit master that had the click? But ALL the 24bit sequenced sides had a click. whether the lin dither was reducing to 24bit or not as far as I can see. Just reporting the facts publicly so others can see if there’s a bug or spot my operator error.
The guys above reported a similar render click only at the start of files on an external chain path, which comes out the master section right?. My intuition says this is the same problem. It’s 100% repeatable and I’m happy to upload the project to Steinberg if anyone there would like to investigate.

If your project does not contain a third-party plugin, and the problem is reproducible, yes please send it to me. If you need an upload place, I can provide you with one in a PM.

Hi. So after a few more months I’ve not had the same problem until today. Dozens of vinyl comps and nada. It seems it may be a combination of Acon Digital Restoration suite plugs - specifically the Denoise2 and rendering CD track groups, where that is in the clip chain on one of them - possibly the leading track of the group. The click apparently disappears if I use ‘union of selected clips’ to render the vinyl side, rather than ‘all CD tracks in group’. Beats me why there’s any difference, but for the record, that’s what I’m observing. And yes I now firmly suspect Acon Denoise, despite it being a great plugin, something happens when it’s rendered in a CD group for some reason, but not union of clips. THankfully the union of clips seems to preserve the gaps/SMPTE offset relationship.

Remember, the individual, muti-output, clip renders always seem to be ok, it’s when I do the vinyl side CD group renders that I got a big click in the first few seconds of groups as above.

To be absolutely honest Steinberg’s investigation and support stipulation of ‘no third party plugins’ attitude to this ‘World Class Mastering Software’ leaves me baffled. Plenty top class mastering engineers use Wavelab for some of their workflow, I’m sure. But I doubt you can name one that uses solely stock Steinberg plugs.

My plugs are Izotope, Fabfilter, Plugin Alliance, Waves. PSP… you know , the best of the best processing money can buy. They never glitch Tools or Logic etc. Why are Steinberg so averse to investigating and coming back and saying it’s a plugin issue? -What if it’s a combination issue as looks likely in this case?

I recently reported a CD report issue that prints pauses in the wrong order in my client ‘PQ code’ pdf sheets for the cutting engineer. You verified it’s a bug from my project and were totally non committal at a fix at any point for sure. No thanks, no apparent interest at the lack of proper functionality as it impacts audio pros. Maybe you only pay real attention to your select beta testers, rather than experienced users actually at the sharp end of your software?

FWIW. I don’t use ANY Steinberg plugins, and I only use 3rd party plugins. That said, I use ones that I have put through their paces and trust to work in WaveLab. I even have a “Render Test” montage that I use if I want to consider introducing a new plugin to my daily workflow.

FabFilter, DMG, Sonnox, Goodhertz, Plugin Alliance, UADx/UAD, iZotope, Leapwing, Eiosis AirEQ, Toneprojects, Oeksound, Tokyo Dawn to name a few. In my experience those work well in WaveLab.

That said, I’ve contacted most of those companies at least once in the last decade about an issue with their plugin(s) in WaveLab and they were all able to resolve the issue, without blaming WaveLab.

WaveLab is a host for plugins and can’t force every plugin manufacturer to test their plugins in WaveLab, or follow VST3 spec close enough that there will be no issues on playback or rendering.

Aside from Pro Tools and Logic using totally different plugin formats (AAX and AU), the bigger DAWs like Pro Tools, Logic, Cubase, and to some degree REAPER get the most testing attention from most plugin developers and often more niche mastering DAWs like WaveLab, Sequoia, Samplitude, SoundBlade (RIP), SADIE, and Pyramix to name a few are more likely to experience a 3rd party plugin issue. Comparing VST3 plugin performance in one DAW to AU performance in another DAW is apples and oranges.

I used to have your attitude about it in that if it works in Pro Tools it should work in every DAW but after over a decade of using WaveLab and as a byproduct, becoming a beta tester for many 3rd party plugin companies to help them understand what isn’t working in WaveLab and encourage them to fix it, I see why and how these issues must be fixed by the plugin developer and can’t really be fixed by WaveLab.

I can’t think of one issue in over 10 years that was a WaveLab specific issue when it came to plugin rendering or playback. WaveLab is simply a host for plugins and Steinberg or PG have no control over what plugin developers do when coding their plugins.

1 Like

I’m not sure of the origin of this information, but I can assert that Steinberg does not exclusively advocate using its plugins. However, when you encounter processing issues, it might be interesting to experiment with substituting the problematic plugin with one from Steinberg. Doing this can help in identifying the root of the problem. As you might expect, Steinberg plugins meticulously adhere to the VST-3 standard.

1 Like

@Karl_Junga FYI I attempted to reproduce what you are hearing using DeNoise2 and so far can’t get a repro of the click. DeNoise2 seems perfectly clean here… I tried rendering with Union of selected clips, All tracks in CD group and CD track Group. What is the format / resolution of the source files and what resolution are you rendering to and, most importantly, what other plugins are in the chain, either in the Inspector Clip Effects, Track Effects or Output Effects, or in the Master Section? What settings are you using in the DeNoise2 plugin (screenshot please)?

This is almost certainly a plugin issue. If you put together a precise and repeatable step-by-step procedure that produces the click I’ll gladly test it here (but of course only if I have the same plugins in my collection).

1 Like

Hi ,Thanks for follow up.
So I’ve never had this group render issue without an Acon plugin in the clip chain, and it only seems to happen on ‘all cd tracks in group’ not ‘union of clips’. It’s been the same type of click each time it’s occurred.

Acon is no shirk as a manufacturer and I was mighty surprised. It’s probably only arisen because i’m currently preferring Acon to Izotope RX or Waves noise reduction algorithms despite price, and because I’ve had 2 remastering double pack vinyls in a few months.

I use lots of thoroughly tested high-end plugins as you indicate also. Non have ever given this issue, or any other, in Wavelab. And I don’t recall ever having a render click issue with multi output individual clip renders with or without Acon.

You mention VST3 but I use mostly AU. Without any issue ever , save this. I don’t use anything on the master save Steinberg dither. Anything else on the master bus is metering… and again, never presents an issue.

I actually use BlueCat Patchwork for my mastering chains and AUs within that. This is because the limit of total number and order of plugs in Wavelab is far too restrictive when one wants to swap out known similar units quickly for comparison; and because Wavelab refuses to swap its audio engine seamlessly from track to track for audio comparison. Not with any reasonable latency like a Fabfiler high lin phase EQ in the chain.

So I get the master chain sounding right against multiple iterative trial master versions and references in another DAW, then go to Wavelab, check fades and starts first ( because Wavelab engine clicks randomly and intermittently at file end and start with any reasonable latency plugs) then import the Patchwork chains into a single clip instantiation; then final checks and comps in Wavelab for batch renders.

Like I said, incredibly robust and super flexible so far. Standard 44.1kHz sessions usually. And I know from decades of pro audio practice, associate uni lecturing, and several years of coding AI at PhD that repeatability is paramount in problem solving.

It repeats on our trash can 2017 Mac Pro with latest mavericks, but that may be as far as we can get it for now, bearing in mind that they’re ALL 3rd party plugs.

I’m leaving a public, searchable breadcrumb trail here because this issue may or may not be a complex Wavelab interaction.

Any time I’ve had an offline render click close to the start of a file on any other DAW, historically, it’s been easy to figure it’s a plugin issue and fix it, then render. This is different as it only seems to happen with a specific command to group tracks together.

I’ve also mentioned here several other issue with Wavelab’s audio engine that I’ve no idea how others work around - the clicking at file ends and starts with any decent latency . I’ve tried buffers etc but I just had to find a work around. The clicks just happen with a single simple plugin - anything with decent latency. Maybe it’s only my Mac Pro, but it’s really tough when you’re listening for tiny file clicks, not audio engine clicks.

The fact that I never experience clicks with the same plugins and chains in Tools, Logic, Ableton, Triumph etc etc etc leaves me logically leaning towards Wavelab’s audio engine as a candidate. Strictly, it’s a hunch. But with the evidence I have and the experience I have, it’s worth leaving here for others to make of what they will - in good faith.

Wavelab IS the best rendering mastering software available imo, or I’d simply not bother trying to raise constructive issues which may or may not fall between the beta testing cracks. Maybe I’m missing something obvious. 100%. But I’d be more than happy to be corrected in a ‘doh!’ moment… so I can get on with my dayjob - mastering and mixing.

@Karl_Junga I fully appreciate your dilemmas and all the ins and out you describe here, and just how difficult and subtle it can be to find solutions for these click issues. Unfortunately, without a specific step-by-step procedure to replicate the issue you are experiencing, I doubt that anyone here would be able to help. I use a PC and I notice you are on MacOS with AU plugins, so you can probably count me out for providing further meaningful analysis.

All I can repeat is that coming up with a specific precise and repeatable step-by-step procedure for reproducing the click is likely the only way you’d find a solution (and I really do mean a VERY precise numbered step-by-step procedure including details of all plugins used and their settings), but I notice that you don’t seem to want to do this… alternatively, could you upload a simple montage which suffers from the clicks? and I’d highly recommend @Justin_Perkins 's practice of checking all plugins you regularly use by performing a rendering test on each, listening out for any clicks.

Unfortunately, suggesting that you don’t have clicks in Tools, Logic and so on is not really a valid argument in all this. Like it or not, this is Wavelab, which asks more of plugin developers where conforming to the official plugin specification is concerned.

Don’t get me wrong, all this is not to avoid the issue, but there are some unavoidable details required (that step-by-step procedure and/or a simple montage, for example) for anyone to be able to test and search for the issue.

I personally wouldn’t recommend using Patchwork in a complex mastering chain. At least, not without thoroughly testing in multiple scenarios.

Right. So, if you get the time, it would probably help if you could provide a precise step-by-step procedure, as suggested above.

For example, the post from an hour ago is the first I’ve seen mention of using AU versions via BlueCat Patchwork in WaveLab which is a HUGE detail. Maybe it was mentioned elsewhere but if it was, I missed it.

1 Like

Yep indeed, that is a pretty huge detail.