Graphic EQs are named so because of the fact that every frequency band is represented by a linear potentiometer and they are aligned next to each other.
Looking at all sliders at once you see how the frequency attenuation and amplification looks like over the whole band.
On the other side you may just dont want to use a graphic EQ because you have special requirements like these:
- manipulating the corner frequency of a band
- adjusting Q (the quality)
where usually the corner frequency (i.e. 700Hz) can be set in a range (like 400Hz and 1400Hz) and Q (the width of the band - meaning frequency range effected) between 1/12 octave to 4 octaves.
A Q of 20 will mean 1/12 octave in frequency range. You could boost or attenuate single notes with this. So, this is only for very special tasks.
A Q of 0.6 is 3 octaves and therefore a quite wide range being manipulated.
Useful Q ranges from 2 to 5 in standard environments.
There are two philosophies with param EQs. The first one - lets call it "set and forget" uses internal trimpots to define the corner frequency and Q. Only the level pots are accessible. This is usually used in a fixed setup.
The other is "full control" meaning all parameters can be dialed in via pots from the front side. This is very comfortable in a flexible environment but there is a chance that you will never be able to dial in a specific setting you did like again after working on something else 🙂
Parametric EQs usually are made from gyrator circuits. There is a version with transistors and one with OpAmps.
A calculator for the corner frequencies and Q can be found here: http://www.muzique.com/lab/gyrator.htm