# Filter Selection

On this page, you select the type of filter/equalizer that you want to train on.

{% hint style="info" %}
‘Filter’ generally refers to a cut-only device, whereas an equalizer can either boost or cut.
{% endhint %}

![The Filter Selection Page](https://3465913129-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F35ssi0AM5G2ho2oSgGf3%2Fuploads%2FSY9AEW0EUBiSMURRrWqA%2FScreen%20Shot%202022-05-29%20at%2012.00.02%20pm.png?alt=media\&token=b0e2c48c-6c24-46cd-932f-035e5e8bc05a)

{% hint style="info" %}
**FR**<mark style="color:red;">**EQ**</mark>**UIA only allows you to train on one filter or equalizer at any given time**. The rationale for this is that very few people can correctly identify +/- 1dB changes made to audio signals, and even fewer can consistently achieve this in [Absolute Identification](https://frequia.gitbook.io/frequia/advanced-training/training-mode) mode.
{% endhint %}

### High/Low-pass Filters

A High-pass Filter (HPF) passes (doesn't boost or cut) frequencies ABOVE the selected frequency, whereas a Low-pass Filter (LPF) passes frequencies BELOW the selected frequency. A high Q value may boost certain frequencies, even though there is no gain applied.

The parameters of High/Low-pass Filters that you can train to identify are Frequency and Q (these filters do not have a Gain parameter).

![](https://3465913129-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F35ssi0AM5G2ho2oSgGf3%2Fuploads%2Fjx673F1YCCoUief55NX9%2FScreen%20Shot%202022-05-29%20at%2012.04.03%20pm.png?alt=media\&token=38be061b-12ef-4ebf-87fc-7e4a068fdb7a)

![](https://3465913129-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F35ssi0AM5G2ho2oSgGf3%2Fuploads%2FsNsHnJWPkm5CpBdFKftL%2FScreen%20Shot%202022-05-29%20at%2012.04.31%20pm.png?alt=media\&token=d6d5a497-38ee-4772-8e1d-a2571ee386cb)

If you choose to train on this type of filter, FR<mark style="color:red;">EQ</mark>UIA will randomly choose between a High-pass or a Low-pass filter during training.

### Shelving Equalizers

Shelving equalizers boost or cut all frequencies above or below the selected frequency. There are two types of Shelving Equalizers: Low Shelf and High Shelf. A Low Shelf boosts or cuts all frequencies BELOW the selected frequency.  A High Shelf boosts or cuts all frequencies ABOVE the selected frequency.

The parameters of Shelving Equalizers that you can train to identify are Frequency, Gain, and Q.

![](https://3465913129-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F35ssi0AM5G2ho2oSgGf3%2Fuploads%2F35PzMjxmSCvBZ8Q7njQ6%2FScreen%20Shot%202022-05-29%20at%2012.05.28%20pm.png?alt=media\&token=9d9bf8fc-1086-4e6a-bf20-0a8843acd2c9)

![](https://3465913129-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F35ssi0AM5G2ho2oSgGf3%2Fuploads%2FcGoatb51rm6TJGdtq6e7%2FScreen%20Shot%202022-05-29%20at%2012.05.35%20pm.png?alt=media\&token=7c86444d-7a3a-4dbc-aba5-7177206afc7f)

If you choose to train on this type of equalizer, FR<mark style="color:red;">EQ</mark>UIA will randomly choose between a High-shelf or a Low-shelf during training.

### Parametric Equalizers

Parametric Equalizers boost or cut frequencies symmetrically around a selected center frequency. Gain controls the amount of boost or cut, and Q controls the bandwidth of the boost or cut.

The parameters of a Parametric Equalizer that you can train to identify are Frequency, Gain, and Q.

![](https://3465913129-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F35ssi0AM5G2ho2oSgGf3%2Fuploads%2FLowSwU73cW6YrOyEUfh8%2FScreen%20Shot%202022-05-29%20at%2012.06.45%20pm.png?alt=media\&token=1425fade-0859-4492-9b87-2335ff2fa287)

![](https://3465913129-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F35ssi0AM5G2ho2oSgGf3%2Fuploads%2FuA6zXqmP3Jtno7Fvvdqd%2FScreen%20Shot%202022-05-29%20at%2012.06.58%20pm.png?alt=media\&token=409787a2-a271-4fdc-ab6d-2231e6d93f1d)

### Bandpass Filters

Bandpass Filters pass all frequencies within a certain range and cut all others. In reality, these filters cut all frequencies except those at the very center.

The parameters of a Bandpass Filter that you can train to identify are Frequency and Q (these filters do not have a Gain parameter).

![](https://3465913129-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F35ssi0AM5G2ho2oSgGf3%2Fuploads%2F6FzhYoxS9u2MknM4Qr0e%2FScreen%20Shot%202022-05-29%20at%2012.07.56%20pm.png?alt=media\&token=ee24a0f6-c959-4793-9f79-5a8517c2f4af)

### Bandstop Filters

Bandstop Filters cut all frequencies within a certain range and pass all others.

The parameters of a Bandstop Filter that you can train to identify are Frequency and Q (these filters do not have a Gain parameter).

![](https://3465913129-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F35ssi0AM5G2ho2oSgGf3%2Fuploads%2FqgDUJrldOVaf2jKXxNax%2FScreen%20Shot%202022-05-29%20at%2012.08.58%20pm.png?alt=media\&token=004a7a02-fc01-476e-a512-73c21a9d16d4)