Page 1 of 1

learning to trust my ears not my eyes

Posted: Thu Nov 19, 2009 6:26 pm
by mxomtk2k
Hello Paavo and Earle,

I have a question which may seem rather elementry since I have been a harbal user for so long now. While harbal is marketed (and with good reason) as a visual mastering system to aid the proccess, I find all too often I get caught up in the visual aspect of the program and often put the sound second to the shape of the curve. Using Harbal has helped me to improve the quality of my mixes to no end but I find myself striving to achieve as flat a response in Harbal as possible and I can easily convince myself that what I thought sounded good only moments before I looked at it in Harbal now doesn't so much once I see the curve is not flat enough. I have learnt the dangers of matching the curve of other albums so I am assuming it is ok for each song and/or album to have a unique curve of its own, and not nesessarily flat either? Is it just a matter of retaining the natural curve of the song but taiming the extreme peaks and dips?
Or is it ok for more obvious wide Q type eq to be used or is that best to be done on the stereo bus in my DAW before exporting and leave Harbal for the finer details?
Ok thats more than one question I guess but I hope it all makes sence, while I have been using Harbal happily for a while now I still could not get this question out of my head so thought it be best to ask.

Thanks for your time,

Nigel.

Re: learning to trust my ears not my eyes

Posted: Thu Nov 19, 2009 7:35 pm
by HarBal
Hi Nigel,

The "visual mastering" thing is marketing spin. I have some intention of toning it down in the future but I've got more important things to worry about than monkeying around with marketing spiel.

"Eyes to the ears" is a more realistic slogan that I like. It does not mean give way to your eyes if your ears tell you everything you need. More the case that when you are struggling to hear what is wrong (but you know something is wrong) it can give visual insight. We don't suggest you simply trust what you see and go by that. Its an interactive process of listening and seeing and experimenting with changes. The key word here is "experimenting".

You might see a large whole in the spectrum and wonder is that bad acoustics or phase cancellation or simply a result of the instrumentation in the track. To answer the question you dive in and make some changes and listen to the effect. Did it improve? Yes, then perhaps we have phase cancellation issues in the mix, No, then it is a natural signature of the track instrumentation. That is essentially how I use it. I never know with certainty, apriori, whether a track spectrum clearly demonstrates a particular issue. I may have strong suspicions but they are only confirmed after adjustment and listening.

IntuitQ and IntuitMatch are merely tools for the user that is unwilling to experiment and does not have monitoring that they can trust. However, if you want to make good sounding recordings you should try and get monitoring you can trust. In the case of speakers most of that is the room more than the speakers, though good speakers are important.

How can you know if you can reasonably trust your monitoring? I would say the easiest way would be to invest in some decent quality headphones and compare the tonality of what you hear in the headphones with what you hear in the room. If there is a general consistency then you can, if not, then your monitor may be biased one way or the other.

Then it's a matter of treating your room until you come up with something comparable. Now don't read too much into this because I'm not saying they should sound the same as they won't. Speakers should sound spatially richer that headphones as you have the room ambiance (hopefully) adding to the performance. The imaging in headphones is also totally different, but that does not mean you can't have comparable tonal balance and that is the consistency I'm referring too.

So in summary, yes, you need to trust what you hear and if you don't have faith in your monitoring then you should probably look and rectifying that. If nothing else, use good quality headphones to judge tonality and speakers for imaging.

Cheers,


Paavo.

Re: learning to trust my ears not my eyes

Posted: Mon Nov 23, 2009 5:14 pm
by mxomtk2k
Paavo,

Thanks for the reply, It makes sence for sure and I think I have always known that but when you already doubt your own ability it is easy to find an excuse anywhere :) There is no doubt that using Harbal has opened my ears AND my eyes in a way I never though possible and thats a skill you can take to the bank.
On monitoring, I noticed on a few older posts you recommended Dynaudio BM5a monitors. I am planning on getting new monitors in the next week or two as I only have Behringer Truth monitors which, have served their purpose, but are not detailed enough. I can get a pair for round $2000 which I think seems reasonible but I wondered if there was anything else I should consider. I can stretch to round $2500. I have a smallish room that will be treated soon too and I mix a wide range of styles from classical to death metal and everthing in between.

Thanks for your time,

Nigel Mauchline.

Re: learning to trust my ears not my eyes

Posted: Mon Nov 23, 2009 5:59 pm
by HarBal
I recommended Dynaudio BM5a monitors (it was someone elses query about them) on the strength of the drivers they make and the reviews I read about them. I haven't heard them myself personally, so I can't give my own opinion of them.

As a general rule, I'd suggest looking at the specs for the speakers your thinking of getting and check the crossover frequency and the woofer driver size (assuming typical two way monitor design). The crossover frequency should be as low as practically possible (around 2kHz) and the woofer no bigger than 8 inch (if you need extra bass extension look into a separate sub woofer).

The reason for both those suggestions has to do with dispersion. Whilst you may have two way speaker systems with higher crossover frequencies and larger woofers that have great looking on axis frequency response a problem arises because of what happens off axis. Larger drivers generally have a more beaming polar radiation pattern at higher frequencies than smaller ones, which makes large woofers not very good at wide dispersion of upper-mid-range energy. At the same frequency the tweeter has very good dispersion, so the net effect of a too high crossover and/or too large a woofer (in a two way system) is a region in the mid-range where dispersion is weak compared with other regions. Put such a system into a typical room and it will sound mid-range light as a consequence.

In my view, the dispersion characteristics of a loudspeaker are far more important than on axis frequency response, yet it is something rarely comprehensively measured or documented, I guess partly out of the difficulty in actually measuring and presenting that data.

The only other advice I'd give is don't be put off by something if it doesn't have a huge price tag. Judge it strictly by the quality of the sound you hear. Take a reference CD with you that has a collection of your favourite well recorded pieces and play them back on a variety of systems and then decide which works best for you. If you have to visit a number of different shops try and take note of the environment you are listening to them in to try, as best you can, to factor in what that might be doing to the sound you hear here as opposed to another place you visited.

Cheers,


Paavo.