DIY Album

Recording & mixing guitar : My own recipe

Even if there might be a perfect way to record electro-acoustic guitar and electric guitar, I assume mine is not perfect. Below I explain the way I recorded and mixed the acoustic and electric guitar on the album, with extremely cheap gear and stuff.

What I used for the recording itself :

  • USB soundcard (E-MU tracker pre, probably the cheapest one ever) with two Jack/XLR input and a stereo output for direct monitoring (i.e. without latency).
  • Headphones
  • Korg ToneWorks Pandora Box (a very cheap multi-effect device that I got from my guitar teacher Yann-Guirec Le Bars almost for free, some years ago)
The USB soundcard, the headphones and the Pandora Box
  • Of course a Laptop (Mine is 2GHz AMD Athlon Dual Core, 4GB RAM, Windows 7-32bits)
  • Microphone : Behringer B-1 (Also used for percussions and vocal, actually for all the instruments of the album !)
  • One microphone stand.
The microphone listed above.
  • 2 Jack cables, one XLR cable
  • And the music recording software, the wonderful Cubase SX3 !!

Of course the microphone has to be put in a suitable way next to where the right hand will touch the strings… Listening carefully with isolated headphones can help to find the best setting.

Microphone setting for acoustic recording

Okay, now let’s talk about the recording itself, for acoustic guitar first. I record two mono tracks, one with the microphone and one from the Pandora Box.

Two guitar tracks

The upper track is the one from the microphone, the lower is that from the Pandora Box. See for each track the routing I used (Microphone on right-in, Jack on left-in).

Then, each track will be doubled. Jack track 1 will be sent to the right output, and its copy to the left output. Then Microphone track 1 will be sent to the left and its copy to the right. The levels of the two copies have to be reduced in order to separate them in the stereo output. I also add a time delay of say 15-20ms to those two tracks to enhance the stereo effect.

Then both tracks get doubled, a time delay of 15ms is added to the two copies.
The mixer shows the routing and levels used for each track…

And of course, basic effects such as compression should be added to each track…

“Multiband compressor” effect.

Reverb, delay, flanger etc… are also possible, only your ears can tell you what is missing, in case something is actually missing !

And what about electric guitar ? Let me show mine :

My electric guitar ??

Yes, I know, this is not really what we can call an electric guitar, I have a nice Kort electric guitar but I left it in Brittany, we have to reduce the amount of luggage when travelling from Brittany to Istanbul ! But here is the Pandora Box, EXTREMELY helpful ! So, Pandora has a stereo output, I use a special cable to split it in two mono outputs that will be connected to the soundcard as follows :

The way Pandora is connected to the soundcard, see the two thin cables on the left.

Then, I only use two tracks, one for each mono output. From Cubase, one is sent to the left output, the other to the right. Some examples of what it sounds like here or here !

I used exactly the same kind of procedures for electric baglama and electric ‘ud. For acoustic ‘ud I only used the microphone and two tracks.

Well, I’m quite satisfied of the result so far, even if this surely has to be improved ! Would you have any useful advice, please let me know, all suggestions are welcome 🙂 !
§ Yann

Buy Astrakan Project debut album

DIY Album

16-bit or not 16-bit ?

What professional sound engineers say :

It is commonly known that professional studio recordings are made using 24-bit depth and sometimes 192kHz sampling frequency. But records that are on a CD are only 16-bit and 44.1kHz which enables a sound quality that is already far above what human ear can feel.

But then our question was :

Why the hell are they using 24-bit/192kHz although 16-bit/44.1kHz is what we use for a CD ? The answer is rather simple : when processing tools are used (such as compression or reverb for example during the mix and also during the final mastering process), low bit depth and low sample frequencies result in lower sound quality after the processing, for a simple reason : numerical errors that occur during the computations (because numerical data use a finite number of digits) get more amplified as the processing chain gets longer. Therefore it is strongly advised to record and process using the highest bit depth and sampling frequency as possible, and to convert the final result to 16/44.1 before burning the master.

But this means several requirements :

  • Much larger amount of memory
  • More powerful computers and stuff to handle this
  • High quality and expensive monitors (they can be very expensive !)

As a conclusion : much more money !

And what we do have :

In our case, we have a very cheap USB soundcard (E-MU tracker pre, about 100€), Behringer MS40 monitors (among the cheapest monitors ever, around 200€), and a laptop (a PC, not even a Mac which means no firewire) with 2GHz AMD Athlon Dual Core and 4GB RAM.

The voice and the stringed instruments were recorded with a Behringer B-1 microphone that we bought for 70€ in Portugal. We also used a Korg Pandora ToneWorks for the stringed instruments effects (Compression, chorus, flanger, distorsion etc…) that we got almost for free from my previous guitar teacher some years ago. The percussions were recorded both with this B-1 and a basic SM58 microphone.

This is what we could call a cheap home don’t you think ?

The home studio

So our conclusion was :

  • First of all, we think our monitors cannot truly render the difference between 16/44.1 and 24/192.
  • Our sound recording skills, experience and listening accuracy may not be sufficient to be able to feel the difference (it is commonly said that a sound engineer’s ear has to be trained for at least 10 years before being fully accurate).
  • Most of the samples we used when setting up the pieces (i.e. snare or bass drum samples for instance) were mostly free samples collected from the internet, or even extracted from CDs, and therefore 16/44.1 samples.
  • The recording process was already quite advanced, and we were feeling doubts growing up in our mind.
  • We read somewhere that Alanis Morissette’s first CD was recorded using 16-bit…
  • We also realised that most people listen to mp3 music on their computer or smartphone directly from youtube or soundcloud with average loudspeakers or headphones !

Then we said, let’s keep on going with 16/44.1 and let’s really try to do our best !

Can you really feel the difference ?

[ yann ]