Pre-mastering...

[Brainstormer]

Ok, so we have an EQ chart up, which is more geared for tracking, so how about pre-mastering...

Pre-mastering is not all about smashing all the dynamics out of a mix to get it ten times louder than the next tune - leave that to the guys at the mastering houses, they have more leeway when taking it to vinyl. It is also the last stage in mastering.

When it comes to EQing the master, You can't always trust your ears when there's heaps of other factors to consider, such as frequency flaws in your monitors, or room modes that cancel or enhance certain frequencies. You've probably noticed certain sounds appearing louder or quieter when you play your tune on another system?

If you use a program such as Wavelab, you'll find a spectrum analyser tucked away in the analysis menu. this is what you need to get the final EQ as close as possible. It's quite easy to use. it has a logarithmic table showing the frequency response. The first thing you should notice is that your bass appears higher and the treble slopes down at the other end. This is normal, what you're concentrating on is the middle section.. This range spans from around 500Hz (it still might have a slight slope coming down from the bass end, but that's ok. but should remain fairly flat to about 10KHz, so you need to use a parametric mastering EQ to flatten this out if anything is looking out of place. You may need to lift some frequencies and drop others, but use a wide bandwidth on your EQ.

You may even find a few flaws where you need to go back to the tracking stage to sort the odd sound here and there. These will show up as peaks when certain sounds come in. Check what frequency it's peaking at, then go back to the mix to draw them out of the offending sound.

Last step is to make sure the overall level is not clipping, so use a maximizer such as Waves L1,2 or 3... or similar, place this as the last step of the plugin chain (not talking about the BBE sonic maximizer - that's a 2 band EQ which you won't need since you're using a proper parametric mastering EQ to sort the flaws out) Basically pull up the level enough without peaking - the final render should not have any sections squared off where it's peaked. As I said before, the guys in the mastering house can take care of boosting the level far more than you can on your PC.

I hope this helps. If you need to know any more, just ask.

[StuartJ]

Interesting read, thanks rich!

[DJ_VENOM]

Very Helpful.

[DJDeluxe]

I use a similar technique. I what is into soundforge, and spectrum analyse it. This brings up a report of where the EQ is lacking. So I do a paragraphic EQ to boost the frequencies needed.

Only recently started doing this, and getting better results.

but yeah, Rich. Good advice there mate

[Sharkfin_Records]

Thanks for the tips there Rich.

On the same topic, can I ask what everyone does about mastering before sending a track off to be pressed? I have always sent off a mastered (by myself) version of the track, and this is also what I've had from other artists on the label. By mastered I mean compressed/limited/eq'd, as you would to play it from a cd. The results generally seem to have been good.

I've heard various conflicting advice on this however - some people have said to compress the f*ck out of a track and limit it to the max before sending it off, others have said dont touch the mixdown and leave it all to the cutting engineer. I had always thought that to have the track properly mastered you'd need to pay extra for it to be done in the studio before cutting teh track, and that only minimal adjustments necessary to get the sound onto vinyl are made at the final stage. Am I completely wrong about this?! What does everyone else do?

[Electric Tribe]

We initially used to leave it to engineer to master, but that was mainly cause we were able to attend the cut and get feedback. For later releases we switched to trying to do an element of "mastering" ourselves, primarily because we we're doing 4 trackers with tracks from different artists/studio setups and wanted to harmonise tracks on each side. This was because found that it caused alot of grief if vastly different as quality of inner track ends up being sacrificed purely because the cutting engineer has all of 2-3 seconds between tracks which isn't long enough to switch settings on units.

Mastering for vinyl is a f*cking nightmare especially for bass heavy music. your trying to squash the dynamic range of the entire track into a 6dB space. As Brainstormer alluded to in the above post, the normal "balanced" frequency spectrum actually has a kind of U shape to it so bass and higher end frequencies are way way louder. This is down to the ear acting as an asymetric band pass filter around 2-4 KHz.

The tools you got for "mastering" are compressors, limiters and EQ's.

EQ's you all know about. things to watch out for are that you got more control over frequency response by cutting than boosting signals (purely due to how shape of filter acts). Also if you find that you're boosting lower frequencies to get a more pleasing presence to the sound, your probably better of using a distortion effect or trying to cut the bass and boost the mid range a touch, because it'll be the harmonics caused by the lower end distorting that your hearing...

Limiters are basically to make sure random wayward "transient" peaks are dealt with, that would otehrwise cause record to jump. If you can "hear" it making things louder or quieter when used without gain compensation, your effecively applying an element of very heavy compression, which your better of using a compressor for...

Compressors act on the whole mix, so they have tendency to alter the frequency spectrum by reducing bass and high end frequencies. This results in an apparent boost in the high to low mid range overall, because they're not being as affected. The other limitation is that the entire mix is reduced when the compressor is active (think of the "pumping" effect so beloved in house nowadays)

To get around this problem, "multi-band compression" is used so that as much of the original frequency spectrum can be maintained. It's not precise though, which is why you inevitabley get a "trade off", with the cutting room engineers skill being how he judges what elements to try and keep.

as an aside, this is probably a major part of the reason why vinyl sounds better/warmer than CD. with CD the narrow dynamic range is less of an issue and the main focus is to avoid clipping. To give you an idea of what I'm talking about, listen to songs on the radio, then listen to the CD version. you may notice a similar "warmth issue".

anyhoo getting back on topic. Simon if you can get someone understands the music your making to "master" the tracks before hand, then send of to cutting room and ask them to cut it "flat".

If not, then ideally try and attend the cut and get your "pre-master" mixdown as tight as possible. sounds really complicated but actually if your starting of with good quality sound sources and moderate compression on each bit then shouldn't be much else needed to do. (Trade off is that style of music preferentially is quite sparse as a result).

Penfold