Why Does It Null? (with ReaEQ) Harrison 32C Channel Strip

[Music] hi and welcome back let's just have a little chat about channel strip plugins to start with because generally speaking i don't think these tend to be well thought through first of all let's have a think about the channel strip on a real analog console one of the main limiting factors is available space imagine a console which provided fully parametric 4-band eq plus full-featured dynamics on every channel as well as the usual dozen or so aux sends and routing switches you would need arms like mr tickle to operate it in case you don't get that reference i just mean you'd need really long arms of course you could make each channel twice as wide but then you'd only get half the channel count or else you double the console's footprint so decisions like limiting compressor controls to fast slow switches rather than attack and release knobs stem mostly from that basic limitation you might be of the opinion that the reduced control set is inherently beneficial and helps you to work quicker and i can't argue with you if so but that wasn't the reason for these control decisions console designers would undoubtedly have provided attack and release controls for the compressors and queue for the eq bands had they had the space available and it's notable that every hardware digital console i've ever used has indeed provided all those controls designers of digital consoles don't appear to regard these limitations as desirable so let's have a think about plug-in channel strips basically i think these fit into two different categories but the developers of these plugins don't all appear to have realized that and consequently many of the channel strip plugins i've tried seem to sit awkwardly between the two having apparently not quite decided which category they want to fit into let's start with the first category which is covered quite nicely by the waves channel strips shown here the feature set is limited to what you get on the hardware console clearly the intention is to provide the color of that specific console rather than to be your primary mixing tool i therefore think of this type of strip as a tracking console channel strip if you're using an instance on every channel chances are you're applying some preliminary eq maybe a bit of preamp saturation then largely leaving those settings alone while you finish the mix using other tools if you use an eve flavored channel strip at the start of every plug-in chain you can pretend you're mixing a project that was tracked through a need console but with the added bonus that you can reach back to the tracking stage and dial something back if you overdid it in this case the most important factor is how well the plugin emulates that console's behavior or perhaps just how good does it sound and if the plugin preserves all the same limitations as that console that's arguably a good thing for it to sound the same as the real console it needs to guide you towards the same kinds of decisions as the real console if those limitations mean it can't get you all the way to a finished mix that's okay the tracking console isn't supposed to produce the final mix that's the job of the mixing console and that brings me to the other type of plug-in channel strip ones that aim to be a mixing tool obviously these still need to sound good the processing on offer needs to be comparable to other rival plugins and preferably better than the stock daw options but in this case the main reason to use a channel strip rather than individual separate plugins is workflow having the majority of your mixing tools in a single plugin saves you time loading individual plugins and subsequently having to open multiple different interfaces i have a lot of sympathy with this philosophy mixing is an iterative process and nothing is final until you actually hit render it's therefore quite normal and remarkable to have to revisit the settings for a specific channel multiple times during the course of a mix and only having one interface to pull up to do so can save a significant amount of time for a channel strip to succeed by this criteria it needs two possibly contradictory qualities it needs to be quick and fast to operate so the ergonomics need to be well thought out and indeed this is much more important than it looking like a real console in my opinion and it also needs to be flexible and powerful enough to get you at least most of the way to a finished mix because as soon as you run into a limitation that means you have to load another plugin instead the strip has failed in a key criteria slate's virtual mixrack plugin definitely fits into my mixing channel strip category in my opinion the goal here is quite clearly to provide every possible flavor of processing within the channel strip so as far as possible each daw channel will only need a single vmr and no other plugins of course it achieves that by essentially providing plugins within a plug-in which you could argue is just moving the problem somewhere else but once you've picked your chain or loaded a preset you have everything within a single interface with all parameters available at once well at least until you've loaded too many modules and you have to start scrolling them but slate made the canny decision to emulate 500 series modules and limit the width of each module accordingly this means there's limited space for controls on each module and they therefore tend to stick to the core essential functions but there is still enough space to include all those essential controls like attack and release for a compressor or an eq with fully parametric mid bands you could argue that limiting the control set to these essential parameters is a good thing and i would have a lot of sympathy with your argument bear with me while i veer off on a tangent to a tangent but more controls does not equal more control let me illustrate that with a reverb plugin behind the scenes this effect is in actual fact hundreds of individual delay tabs with individual feedback and diffusion and level and panning can you imagine what the interface would look like if we had individual controls for each of those behind the scenes parameters in that case we would have more controls but way less control we'd have the freedom to create an infinite number of probably really bad sounding settings and infinite insignificantly different variations of every setting but useful settings and meaningful tweaks would become way harder to find in fact one of the most important aspects of reverb design is the way those thousands of internal parameters are mapped to just a small handful of controls the most successful designs will maximize the good sounding settings combinations available and minimise the bad sounding ones and make it as easy as possible to steer towards your goal in that respect i think virtual mixrack strikes a pretty good balance there's enough controls to cover most situations but still most of the time you'll be able to see all the settings for a channel at once so that brings me to the harrison 32c channel strip plugin i've been seeing lots of adwords for this recently so i thought i'd grab the demo and try it out the first thing you might notice is that this is still the demo version so i'm going to have to keep clicking the nag screen away it means that i'm afraid i've not been convinced to click the buy button even with the introductory discount i'll explain why in a moment but if the lag screen irritates you as a viewer my apologies but this is actually my favorite feature of the plugin so far it's vastly preferable to an audible hiss or periodic muting of the signal then it means i can actually use it in real projects and evaluate the sound properly so first question which category of channel strip should i put this into is it a tracking strip that i'll just use to apply some initial colour before finishing the mix as normal or is it intended as your main mixing toolkit i'm not sure to be honest and i'm not sure the developers are either on the one hand the eq and filters are apparently modeled on the original analog harrison 32c console according to the marketing the 32c channel plugin provides a complex emulation of the original harrison 32c eq every resistor capacitor and transistor is included in the model this strongly suggests there's some special colour in the eq that might make you want to use it as a tracking channel strip to emulate the sound you get from tracking through an original 32c console however there's no saturation at all anywhere the eq is clean as a whistle just like pro q3 or your stock eq probably there's no preamp emulation no harmonics added anywhere except presumably the usual ones you get from dynamics processing this is an interesting choice now i made the mistake of stumbling into a thread on the subject and gear sorry gear snobs i mean gear space this was mostly just multiple car pilots really but there were some interesting responses from a harrison representative that's true we don't add any distortion or noise in our channel eqs in case you doubted me on that point as i'm aware i haven't provided any evidence for it yet uad makes a great oversampled emulation of the console eq which adds some color if that's what you're after as the original developer of the console we aren't that focused on the artifacts of an old console which largely weren't there when the consoles were new but rather the painstaking work that went into voicing the eq itself [Music] so this strongly suggests that the plugin isn't intended as a coloring box indeed do i even detect a hint of disdain but the suggestion a brand new harrison analog console would be any less clean and pristine sounding than the daw i don't have one of those in my studio you won't be surprised to learn so i'll have to take their word for it that under normal operation a new harrison board can be considered totally clean but of course there's no analog console anywhere with infinite headroom if you keep cranking the game sooner or later you'll get distortion and extra harmonics whether that's good or bad it's an option you'd have with the analog hardware that you don't get in the plugin so considered as a coloring tracking channel strip the only possible source of any magic source can be either the particular behavior of the compressor section or the special sexy curves on offer from the eq so let's try the eq first after all every resistor capacitor and transistor is included in the model so this should be pretty special right here's a low shelving boost i'll give it a realistic plus 6 db at 150 hertz [Music] it sounds much as i would expect a low shelf boost to sound to be honest it sounds good but i'm not sure i'm hearing any special magic yet [Music] so we need a reference i guess let's pull up my stock eq which as i'm a reaper guy means re-eq and try a similar boost matching the frequency and gain is easy i can just type them in but i'll have to use my judgment when setting the bandwidth in re-eq as there's no equivalent parameter in the 32c plugin [Music] what do you think do they sound the same [Music] well i personally don't think they do it seems to me like the 32c plug-in sounds a little warmer richer fuller-bodied the kick sounds a bit deeper with the channel strip plug-in to my ears it's quite subtle but i'm confident i can hear it or am i am i just imagining it let's try a null test i know we're nowhere near and null regardless of my bandwidth setting but this is a surprisingly big difference but i'm not inclined to leave it there because it doesn't quite smell right there shouldn't be this much difference between two eqs that are working correctly in my opinion so i loaded them up in plug-in doctor and sure enough there are more differences beyond just the bandwidth setting i had to set a different frequency in re-eq to match the curve from 32c fair enough you could say we're arguably picking an arbitrary point on the shelf slope to define as the turnover frequency so it's understandable that two different eqs choose to define this differently at least it's nothing like the crazy discrepancies i found in the amic console eq when the glow and sheen options were enabled however i also needed to increase the gain in re-eq this one isn't really open to debate the 32c shelf is applying more than the 12 db of boost i've dialed in that's just a fact armed with this information let's go back to the null test and it's now not too difficult to find a perfect null which drops right off the bottom of my analyzer which is interesting what happened to all those resistors capacitors and transistors let's try a mid bell at 1k and i'm going to check this in the doctor first the gain is off again though not by very much this time but the frequency is also wrong and unlike a shelf you can say definitively where the center of a bell is especially if you switch to show phase shift that zero phase crossing indicates the center of the bell when it's not at one kilohertz [Applause] okay so perfect null again let's try a 10 kilohertz bell same drill perfect null and a 10k shelf perfect null if the developers really did model every resistor capacitor and transistor in the original eq they could have saved themselves some time by just grabbing the robert bristow johnson algorithms that i used to use when building in reactor which also null perfectly with re-eq and as a bonus they'd be showing the correct frequency and gain settings too there's a cynical part of me that wonders if the incorrect frequency and gain settings is actually just an attempt to foil a superficial null test they don't want to make it too easy to prove that this is just a standard digital eq then there's a much less cynical part of me that thinks this is probably a genuine attempt to match the original analog console eq and that it's motivated by a genuine belief that there's something special about this behavior maybe limiting your eq options to four bands with only these curves available will actually improve your mixes but i confess i'm not convinced re-eq can recreate any of those curves perfectly but can also do very different shapes when needed and provides as many bands as you want you're going to struggle to convince me that's not better and yes by the way that means this eq cramps near nyquist i've made a video explaining what that means so i won't repeat myself here hopefully you're seeing a card for that one now i've also made a video that compares re-eq with pro q3 and attempts to assess how much of a problem the cramping really is in practice again card hopefully but in case you don't want to sit through that the short version is that assuming you adjust the bandwidth to compensate for the cramping the differences are actually pretty subtle but of course you can't adjust the bandwidth in 32c because reasons i'm going to come back to that but it means the high bell gets narrower as you approach the upper limits and there's nothing you can do about it now if there was some colorful saturation on offer from the eq or a preamp section that could be used to add harmonics i would regard this as disappointing and surprising but not a deal breaker in itself because a cramping eq is still fine for 90 or more of your eq duties but when the eq is the main color on offer supposedly at least this is kind of a problem particularly if recreating the behavior of the original board was a priority i guarantee the analog 32c eqs don't cramp near nyquist shall we try the high and low pass filters i'm struggling to get excited about them to be honest knowing them to be clean and free of any potentially interesting non-linearities but let's try nulling them against re-eq's filters i have to do both at once sure enough there's a perfect null again it also bothers me slightly but i can't have a high pass filter without also enabling the cramping resonance of the low-pass filter i don't really understand why these aren't separately switchable so as far as color is concerned so far we have no more than is provided by reaper's stock eq plugin but perhaps there's something special about those particular resonance settings they've chosen i'll give the filters a fair chance and try a trick i often like to use with colorful filters i'll high pass filter the low end of the drum bus with just a touch of a resonant bump the normal bump is enough i don't think i want the extra in this case it's too much and i'll also bracket the high end with the low pass filter around 10k usually does the trick but that's sounding a little dull so i'll try a touch higher [Music] unfortunately i don't have the option to increase the resonance for this one and let's toggle bypass [Music] i'm not convinced to be honest i think i prefer it without for reference here are some genuinely colorful nonlinear filters doing the same kind of thing but with the gain carefully tuned so they're saturating nicely [Music] the compressor section is going to have to be good to save this one let's give it a spin there are three different modes on offer here which are apparently derived from the most commonly used settings in harrison's hardware consoles and that's quite an interesting idea the first setting is described in the blurb as a compressor mode with an adjustable ratio moderate attack and programmed dependent auto release for general purpose guitar keyboard and overall mix compression they say adjustable ratio as if that's some kind of unique selling point for a compressor but in fact it's just code for there are no attack or release controls and i mean none not even a measly fast or slow switch [Music] next up the leveller mode with an adjustable attack a fast release and a very gentle ratio for always on vocal compression and this really does have a very gentle ratio which is interesting or there's a limiter with adjustable release timing instant attack aggressive ratio and a multi-stage hold plus release envelope for drums and other percussive sources [Music] and yes this is definitely a limiter so okay for a limiter i guess i'm okay with only having control over release time the slowest release time on offer isn't particularly slow but the fastest promises to be really fast one millisecond suggests to me that we'll be exploring the gray area between limiting and distortion meant adding growl or crunchiness to the sound [Music] but unfortunately it just sounds bad [Music] i can hear the program dependency struggling to prevent any distortion when there's nothing about that sound i like honestly i'd rather just have the distortion in a mixing context a crunchy limiter with character would be much more useful than a mediocre mastering style limiter that attempts to do brick wall limiting without the benefits of look ahead in my opinion for reference here's a good character limiter doing a similar amount of squash [Music] let's talk about the leveller mode and goodness me it's gentle this is suggested as a vocal setting and sure enough this type of setting can be useful on vocals but it's unlikely to be the only compressor you need on a vocal unless it was tracked through compressors that were spanking it hard on the way in there are two possible ways i could imagine the leveler mode becoming way more useful first of all it could have a very gentle knee so that you get into higher ratios if you really push it hard this would make it much more useful and versatile in my opinion while still allowing it to perform its intended function but second and more importantly why is this choice forced upon me this isn't hardware there are no physical limitations why can't i run the compressor and the leveler at the same time with the limiter also active at the end of the chain to catch rogue peaks these great big buttons here could be used to just switch which controls were visible right there's no reason all three dynamics models couldn't be active simultaneously so let's have a listen to the first compressor mode [Music] and actually this sounds good [Music] really good in fact [Music] for the first time i'm hearing some proper character from this plugin when i really like it [Music] there's a nicely aggressive snap added to transients and the auto release has reacted in useful and interesting ways to everything i've thrown at it so far i was hoping i'd be able to find a redeeming feature in this plugin because i don't like being too negative and this seems to be it almost because actually i'm finding this quite deeply frustrating my natural reaction to hearing delicious compression like this is to grab the attack knob and listen to how it responds when i go really fast and smash off the transients completely or when i slow down the attack what kind of punchy thumpy characteristics emerge can i adjust the release to make it pump and breathe in interesting ways or groove better with the song i'm dying to explore this compressor further and find out all that it's capable of but i can't because all i've got is a ratio control so if we're judging this channel strip by how close it can get me to a final mix before i have to load something else well i don't think i'm going to get very far before i want to control attack or release times and need to load a different compressor and i have to ask why aren't there attack and release controls it's not a hardware channel strip with physical space limitations you can just make the interface bigger if you run out of space but in this case that wouldn't even be necessary i mean do these buttons really need to be that large all three could easily fit in the space that one currently occupies leaving plenty of room for two more knobs failing that i would quite happily swap the leveler in limiter modes for proper control over one really good compressor attack and release controls are not superfluous extra features that just get in the way they're crucial to getting a mix just right removing these options isn't the same as stripping things down to the core essential controls as slate manages quite successfully with the virtual mix rack it goes a step too far in my opinion when constitutes dumbing down while we're on the subject why don't we have q controls for the eq bands it's not like there isn't room there's plenty but if there wasn't you could make room now i know the magic of the harrison eq is all in the game queue interaction it must be all there because i can't find any anywhere else for a couple of points having a cue control doesn't preclude also having gain cue interaction you can keep it just as it is but still allow us to make the bell wider or narrower as needed but also having a cue control makes the game queue interaction rather less important and i would go further and say having a cue control that you can adjust with the mouse wheel while simultaneously setting the frequency and gain as i routinely do in pro q3 makes gamecube interaction totally irrelevant the fabfilter eq does have an option to link gain and queue by the way but i never use it not because i think it's bad i genuinely have no opinion because i just have no use for it with pro q3 the queue is where i want it with just a quick flick of the wheel and that's all that matters so again the lack of queue controls in the eq just makes it that much more likely that i'll need to load a different eq plugin that's without even mentioning the likelihood of running out of bands with only four available yes i know analog consoles have four band eqs and when i mixed on them i regularly run out of bands and while i'm at it why won't it let me adjust the low mid band lower than 180 hertz or the high mid band higher than 8 kilohertz these are totally arbitrary limits i don't care if the original harrison eq also had those parameter ranges it would annoy me on the real console too but it's even more frustrating when i know for a fact there's no good reason for it and again it increases the likelihood that the 32c eq won't cut the mustard for some sources and i'll be forced to load another eq now with all these compromises in mind consider the fact that they nevertheless gave us both input and output gain controls for a totally linear plug-in with no saturation this annoys me more than it probably should but it seems to me they're included just because that's what a channel strip is supposed to have rather than with any thought as to how they would be used of course you only need one gain control for a saturation free channel strip but anyway in my opinion a much more useful implementation would be to just provide the eq section with its own gain control which would be bypassed along with the eq when you bypass that as the dynamics section already has a gain compensation knob the eq is the only other reason you'll need to make a gain correction if that gain control was grouped with the eq it would make it easier to assess the settings when bypassing just the eq section and it would also make it faster and easier to explore the routing options at the bottom if the eq had gain compensation that moved with it you'd be less likely to have to adjust the threshold when changing the order of the compressor and eq sections so here's my summary if i'm to consider this as a tracking console strip intended to color all my channels with some harrison flavor before i go on to mix them as normal then i consider the lack of saturation a major floor that's what color mostly is after all right and while ben from harrison can claim that actually a real harrison console is totally clean as well he's also happy to endorse the uad version which apparently does have saturation goodness included so there is indeed a flavor of console saturation that harrison are prepared to put their name to now i feel like we should have some of that goodness in this channel strip before i'll consider using it as a harrison flavored tracking console on the other hand if i'm to consider this as a mixing channel strip it falls short in a variety of ways i'm going to continually run out of eq bands i'm going to need a different eq for my airy high frequency boosts i'm going to need a different compressor every time i want control over attack or release times and don't even think about anything more complex like de-essing there's no side-chain filtering for the compressor so it's not going to happen for a mixing channel strip to be successful it needs to be as powerful and flexible as possible while still being intuitive and easy to use with everything on display simultaneously in a single interface that's not an easy brief as those demands pull against one another but nevertheless that's the challenge and i feel like many developers of channel strips haven't really thought through what they're trying to achieve with their plugins is it the character of the console warts and all including all its limitations or is it a new and better way to mix in the daw pick one because you can't do both so that's why i'm going to pass on this one i'm afraid but i'll finish off with some totally unsolicited advice for harrison if you build a plugin around this compressor but with full control over attack and release and maybe also other stuff like program dependency squash these eq and filter knobs closer together and patch them to the compressor side chain instead of putting them in the main signal path and sell it as a cool sounding full-featured compressor i'd probably buy that that's all thanks for watching [Music] you

Channel: Dan Worrall

Views: 63,778

Keywords: production, mixing, DAW, plug-in

Id: U9eySYxwDXI

Length: 32min 56sec (1976 seconds)

Published: Mon Sep 13 2021