Hello,

I'm sure variations of this question have been asked a hundred times, but I was hoping for some specific answers to my questions so thank you for indulging me...

I am trying to set up my little recording environment in Reaper, and I'm quite happy with it, for the most part. I am trying to make music that has a classic sound. That means the finished product could be a record that came out last year, or 35 years ago. I am using almost exclusively freeware VSTis for vintage synths, electric keyboards and old drum machines. I have classic phasers, flanges, amp sims and stomp boxes, and everything I play sounds pretty much the way I want it to sound.

Now I need to learn mixing and mastering, and this is where my questions come in. I am feeling overwhelmed by the different options, even just for vintage style mastering freeware. I've heard a lot of good things about Bootsy's VSTs, and now I found these Sir Elliot ones that rival Bootsy's, and I see people on this board recommend the Molot compressor, and other suggestions. I was hoping I would only need one good compressor and one EQ, but I don't know if this is the case.

Do I need at least two EQs? I see a lot of 3 and 5 band EQs, and then these elaborate, 21 band EQs, and I wonder if I should have a small one to put on individual channels, and then a master EQ for the master channel. I have heard the Neve 1073 EQ is a classic, so I wonder if I should use a VST that emulates this.

Is there one compressor (or a couple) that can serve as my workhorse? I have read that the Urei 1176 is such a compressor, so maybe I should try an emulation of this.

Is it good practice to put an EQ on every channel? Is there a better alternative?

Is it useful to have a preamplifier (like Sir Elliot's), and/or a tape saturation simulator like Bootsy's Ferric?

Is there already a thread I missed that will answer these questions for me?

One more question: from what I understand, mixing and mastering a record involves spacing and spreading out your channels, using volume control, equalization, dynamics (compression, expansion) and manipulating the stereo field, using pan, stereo width and pseudo stereo effects. Another dimension would be to have a reverb send and control the wet/dry mix of every channel to place different instruments closer to the listener, or further back in the room. Would it be safe to say that I have a handle on how this process works, based on the description above?

One final question: what is the best stereo separation method, and what (free) VSTs do you recommend to accomplish this?

Sorry this post is so wordy, and thank you so much for your help.[/list][/list]

It doesn't have to be that complex:

http://tarekith.com/assets/mixdowns.html

http://tarekith.com/assets/mastering.html

Everything you need is already included in Reaper so just learn on that.

Buy yourself a decent set of monitoring speakers and get busy for a couple of years.

Thanks, I'll give these a read and come back.

Your post has too many questions for me to find time to answer, but I will say a bit on this one.

The main things to do to get stereo separation are:
a) use panning wisely. Genearlly have bass, kick drum, vocal and or a lead instrument panned center. Pan other instruments around the stereo field to give them their own space in the mix.
b) use reverbs and delays to help make some instruments sound furtherback in the mix. Panning the delays and reverbs to different parts of the spectrum can also work well
c) Use volume contols, automation to put things at the right levels for different parts of the song
d) Use Eq wisely to make sure that each instrument is not fighting with others for the same frequency space.

There are no magic VST's to do this. It takes tria;l and error and experience to get good at mixing, as well as a good ear.
There are many useful VST's that help in doing this, but some are costly. Here are some brands to google..all have great plugins, and there are many others too:
Waves, Sound Toys, PSP, Voxengo, URS, TC Native, iZotope

Hmmm...not sure that's entirely feasible. 35 years ago would be 1977. Some exceptionally well recorded records of that era stand up well today, but the general sound of the 70s and certainly into the 80s was quite utterly different to the mixing aims of today. Bass wasn't as heavy then (at least it wasn't so sub-heavy), reverbs tended to be more obvious and denser - often quite metallic. Mixes weren't as dense - it tended to be bass, guitar, synth, vocals. Or maybe some synth layers, but not nearly as many as some tracks today. The way they f**k up vocals today is nothing like what engineers aimed for with vocals in the late 70s. 35 yrs ago the way you mixed was determined by the fact that everything you used was analogue. Digital mixing today can mean entirely different possibilities and aims.



You can get by with just one. But some Eqs have a certain character and sometimes you want different character. For example some bog standard Eqs really don't sound that nice if you boost by much. If mostly what you're doing is cutting, then TBH I don't think you need more than one. If you are boosting - high frequencies can give away the character. Some are more musical (i.e. less precise) and some just sound harsh when you boost at, say, 2.5 kHz and 6kHz and above.



Channel Eq can be both important or unimportant depending on the impact you want a specific channel to have. For example if you're making some dubstep type car-destroyer with ear bleeding bass, you might want to spend on a decent Eq that handles bass well. But generally IMO, channel Eq doesn't usually need to be as flash as a master Eq.
TBH I've never used a real Manley, but some impulses I've heard of them just sound sweet with some bass boost. Sweet in a way that your average cheapo Eq plugin can't be. And I've certainly had desks in the past where the Eq was musical (i.e. you could use a heavy boost and it didn't grate - and I haven't heard too many digital ones that can do that)



For sure. Most instances of compression is just bread and butter stuff that any decent comp can do. Always IMO better to go for a better one first than keep on buying slightly less cheaper ones again and again. The market's loaded with comps. The Glue, almost anything on the UAD catalogue, I use Kjaerhus GCO1 (don't think he's in business any more, sadly) mostly amongst some others - but I'd be happy with that one if I only had to have one. That Fabfilter one looks pretty useful - certainly it has all the features you could want. Again, some comps emulate the colour and character of old gear - it's choice you have to decide - what charactyer you want, or none (sometimes you simply want to tame dynamics transparently without any colour.) It's useful to have a decent comp that has different modes - VCA, optical etc.



Personally I rarely use an Eq on every channel. But no absolute reason why not. If you're going to use that many Eqs, all I'll say is that it's both easier to use and generally easier on the ear if you use Eqs to mostly cut rather than boost.



Yes. Doesn't mean you have to use it all the time though. Not all recordings were saturated. And there are plenty of superb digital mixes made without any analogue saturation at all (or emulation of it).



Thousands of them. Take a trawl through several pages of productiuon techniques, and probably good idea to use the search function with some key words.





Rocksure mentioned a few things. Panning is highly important. It's possibly one of the first and main things you should exhaust before getting fancy with esoteric mixing techniques. Decent use of panning can make an immense difference to a mix. Pan your channels, often pan your FX.



Yes. Don't go mad with reverbs though. I read posts where some people use what I consider way too many reverbs. Keep things simple and IMO there's usually no need to use reverbs as inserts. 2 reverbs is generally enough, maybe three. Using them as sends should suffice if you mix decently. Not all channels need reverb. Reverb has better effect if it stands on its own sometimes - i.e use a standout reverb on only one or maybe two channels. If everything is swamped with a huge Lexicon hall, it'll all sound mushy. And the Lexicon will no longer sound like a standout. A completely sweeping generalisation is to keep drum reverbs relatively short. Pads often don't need reverb, and sometimes simply having longer release amp envelopes on synth lines negates the need to swamp them in reverb. It's another huge topic though.



Best method is good panning, and keep stereo FX swamping down to a minimum. Mostly I don't think stereo separation plugins are a good idea. Do your panning properly and mostly you won't need them. Having said that I personally like using an enhancer over the mix, or at least a big group bus. Most enhancers will somehow or other help clear up stereo mush, or at least give the impression of it. Some like enhancers some don't, but having fancy stereo plugins on channels to me indicates that basic channel controls haven't been used properly yet.

TBH these are all very short answers - it's all a massive topic.

Tarekith: I read both your essays and I found them both to be very informative. For those of you who haven't read them, this is the main point I took away from them: If you are producing your own records, you don't need to 'master' anything.

Mastering is about teasing a better sound out of a lackluster recording, or a recording that doesn't sound the way the client wants it. Examples where mastering is useful: Taking an old recording and making it sound newer, cleaner; taking an amateurish record the band recording and making it sound more professional; 'flipping' samples from other records to fit into your track. Basically, mastering is about operating on a finished recording.

If you're writing the songs, recording the instruments and mixing the levels, you have so much control that you can improve the recording by mixing it down properly. If a synth lead sounds too dull in the mix, I can go into the synth itself and sharpen the sound of the patch. This is going to have better results than trying to tweak a 21-band equalizer.

Two mastering techniques were recommended: running a master limiter on the final recording to make it seem a little louder, and then dithering before converting from 24bit to 16bit audio. That is apparently all you need in terms of mastering.

So that's what I took away. I don't need a 30-band master EQ, or a master compressor. I do probably need a good, all-purpose EQ that I can reuse on different channels, a good compressor that I may or may not even need, a good master limiter, and a good spectrum analyzer to visualize the bass, mid and treble levels of the recording.




kritikon: Thank you for your lengthy response. I like those 70s sounds, the simpler arrangements, the mellower bass, the plate and spring reverbs. A lot of times the verbs don't even sound like they're in stereo. It's like someone's doing hand claps at the other end of a sewer pipe. I like that though

I don't use impulses for anything because they're so CPU heavy and they give me latency. The only thing I can't find (for free) that I would consider using impulses for is spring reverb, and even that doesn't have that realistic of a sound to me. I am using reverbs as sends, though, like you recommended.

I always assumed that if I'm using reverb, that everything should have at least a little verb on it, so it sounds like all the instruments are in the same room. I'm starting to realize, from listening to other records, however, that this is not the case. Keep in mind that I'm making more old-fasioned music, more disco than techno, but even here it seems like you can put a heavy reverb channel with a stereo separated channel with no reverb, even for different drum parts in the same rhythm, and it sounds good. In fact, when I put verb on everything it seems to sound amateurish.

You mentioned a Kjaerhus compressor. I have Kjaerhus Classic Compressor. Have you tried that one? Do you recommend it?

Thank for your help!

Great post, kritikon.

If they're from the 70s they almost certainly won't be. Right through into the 90s, most reverb units except for the high end ones were some form of dual mono. Mono input, stereo output - although it really wasn't true stereo at all. It was simply the same output split into two. There were plenty of pretty expensive desks where there was no such thing as stereo sends. That's only become standard fairly recently with DAWs. And IMO often not needed - sometimes true stereo can be nice for impact on certain channels, but I'm still more than happy with mono sends in this day and age. Nice but not necessary.



Absolutely. It's so easy nowadays to ignore the finer points of mixing and slam limiters at ridiculous levels over mixes in an attempt to correct what should have been done earlier in the mix. If you've got to boost any more than 2 or 3 dB of any frequency range in a master, the mix should be fixed. If you ever have the good fortune to hear someone gifted at mixing in action, you will be astounded at what good mixing can do - I've known people who can mix the same stuff I mixed and make it sound leagues above what I achieve. I'm not necessarily that bad, but I've seen that mixing is a talent. I used to work with someone who intuitively used Eqs etc in the mix, and got results that I couldn't even get with loads of fancy FX over the master channel. And he did it with no master FX.



Yeah - I still have that. It's quite useable, but it's very limited in it's flexibility. The better ones give you far more fine control over the envelope settings, knee, and different modes of compression (not essential, but occasionally very nice to have). Having said that, I'd happily use the Classic comp if that's all I had. For general duties, it's perfectly OK.

Very helpful. I didn't know that about mono reverb. This is really useful information to me.

When I started shooting video with my digital camera, I discovered that videos looked better at the lower frame rates than the higher settings on the camera. I remembered that this was true when I used to make flash videos, and I learned that because film is generally shot at 24-30 frames per seconds, including motion pictures, that this rate resonates with us, and that a faster frame rate will actually look more artificial, and less professional. I thought this was very interesting.

I've heard audiophiles say that music started sounding bad in the 80s because all the gear switched to solid state technology. I don't know if this is true but I agree that a lot of music from the 80s, compared to the previous decade, sounds too cold, too crisp. Maybe people had just come to associate the sounds produced by music's technical limitations with the art form itself.

I've heard a trend in the last few years (especially around the Chillwave sound) of digital fake lo-fi. Records sound like they're mastered in a way that gives them an artificially obscure sound. I think that this production style can lack authenticity---but I do like the warmth and familiarity of actual old records. It does help to know how they recorded reverb in the 60s and 70s, how they faked stereo and so forth.

I found a plugin that simulates ADT multitracking, which was an Abbey road trick of re-recording a track onto tape and then playing both recordings at the same time. The subtle wow and flutter of the tape would make it seem like two distinct voices. I feel like I need to know these types of tricks to get the sound I'm after.

I have a related question about stereo: is guitar distortion ever in stereo? I ask because most, if not all, of the plugins I've used with my guitar produce monophonic fuzz, even if it's a huge metal sound. Then, when I thought about it, stereo overdrive on records usually sounds pretty symmetrical, even though noise is pretty random, so that makes me think that it's usually a mono channel that's split to make it sound bigger. A lot of Shoegaze records sound more 3-dimensional to me than that, though...

Sounds dubious to me. I suspect it's more of a nostalgia thing - audiophiles tend towards being Luddites to anything new. In the 80s there was still plenty of analogue gear around that was similar to 60/70s - tape was still being used, although it was simply better quality machines it was recorded on. Certainly the reverbs and FX were changing quite drastically - I personally find much of the 80s reverbs metallic, but when you compare to today and to the 60s they were metallic. Lots of new and budget digital reverbs were coming onto the market but they were in their infancy. In the 60s it would have mostly been plates - still metallic, but quite different. And in the 90s, companies like Lexicon et al were already making quite sophisticated digital reverbs. But in the early 80s, the analogue synths were still much the same.
Musical styles were radically changing then, so...



Yes, I think so too.



I must admit that guitar's really not my thing. I'd imagine from the odd stomp box I've used over the years though that it's the same dual mono thing. I doubt most of them are stereo. And there are well known mixing tricks to stereo-ise sounds by things like adding panned delays. There's a fancy name for it (can't remember of the top of my head) but you can add simple L and R delays where the one side is delayed by a few mSec more than the other side. Both delays are very short to start with. And you can ever so slightly increase the impression of stereo by lengthening one side's delay. I've an old Digitech multiFX unit that has what it calls "stereo widening" and it is a simple L/R delay.

And don't forget that with most guitar FX, they tend to be multiFX - so it's likely a fuzz has some chorus or flanging or some modFX either before or afterwards, so mostly you don't even need to bother trying to stereoize the fuzz.

you know what I'm sorry about this and I'm sure you mean well and everything but the 70's ended a long time ago and it's about time people realized this, vinyl is pretty shit in reproducing the sound, Neve's designs were pretty garbage until he got actual engineers in to redesign everything, The Beatles and Pink Floyd are okay, no better and no worse and as for valves well everyone in the 70s were busy replacing that crap with transistors because they were sick of them burning out, I'm pretty sure they didn't give a shit about how the "warmth" was gone because in the 70's they were trying to make their sound better not trying to revert back to the dam 40's.

if you want a record from the 70's just go and break out your "stuck in the past" collection.

however if you want to sound all vintage here is how you do it.. cut the highs from 12K onwards boost 310hz for "warmth" and then stick it up your ass for that vintage timbre.

and if you think my response to this is "trolling" or "uncalled for" I'm sorry but I am sick of people holding the 70's up to some unobtainable level that you can only get by spending too much on inferior equipment. try smoking a shitload of pot in your studio dropping beer and stuff down the mixing desk - THAT is "70's"

Hahaha! I wouldn't call sharing your opinion here "trolling". I never liked either Pink Floyd or Beatles recordings as recordings or example of the art of the studio (the tunes stand on their own, of course). The great 70s recordings were in the schlock and schmalz (TV shows) and underground (early Funkadelic, Sun Ra home recordings), IMO. Yeah and I always found Moog synths to be thick and classy, yet bland and not synaesthetically stimulating.

To the original poster, though, something Brian Eno said many years ago: a synthesizer has a voice but no body. So he would not record direct, but through various amps, and mic it. If you try this I think you might find that you don't need plugins to simulate all the qualities you are probably looking for.

precisely! I always found moog to be out of tune! I just don't know where this clinical obsession with the 70's comes from, some of the recordings such as "Meddle" by pink floyd is insufferable.

I have a TX81z synth which is the nicest synth I own because it doesn't overload on the "warmth" frequencies which always do my head in and I end up eq-ing out.
my favorite trick when recording other people is to record their guitar going into a (hidden) buffered splitter so they can think it's going into their 70's correct valve amp which is all fizzy and "warm" warm like fresh turd and the other output goes into the V-Amp Pro which is the only thing being recorded and so far everyone has loved the sound of their guitar on the recordings I've made
analog is only king when they don't realize it's actually digital.