IKMultimedia T-Racks3 EQ                         Back to fader8's audio blog Index

IKMultimedia T-Racks3 EQ Study. Version 3 AU

Measurements at 96kHz, 24 bit, 0.7Hz resolution, Blackman-Harris windowing.

I received this plug-in free of charge as part of a IK promotion, so I thought I'd check it out. This is the T-Racks3, but it only 

came with the EQ component. 

T-Racks3 Plug-in settings for these measurements:

This setting gives the plug-in a whopping 23,165 samples of latency at 96kHz.

The plug-in does not report its latency to the host's plug-in delay compensation, so for the purposes of these measurements I had to manually compensate when measuring the transfer function.

1. Null Performance

The Level Adjust has a +18 to -18dB range. The Bypass toggle switch disables the EQ bands, but leaves the

plug-in active and imposing its null characteristics, (not that there are any to speak of).. 

Magnitude, phase transfer function and harmonics (700Hz at -9dBfs) with null settings:

2. Low Cut

With a setting of 16 Hz, the filter has no effect. It doesn't engage until you bump it up to 17 or 18Hz. 

The low cut has a fixed rolloff.

Taking a new approach here. In addition to the transfer function showing frequency and phase response on the left, the right-hand plot shows the response plotted with a slow swept sine (20 seconds) which gives me a chance to see if any distortion occurs at any given frequency. Overlaid is a sine wave in the passband and facilitates showing the increased noise floor created by the filter. 

Cutoff at 200Hz:

While that looks sort of OK, it's decidely NOT linear phase. 

(Edit: Since this writing, I learned that this EQ isn't intended to be "linear phase" and that another EQ that's part of the whole T-Racks package is.)

The following is the same low cut setting, but with the preferences

changed to the oversampling without the linear phase option.

The latency now bumps up to 41,565 samples at 96kHz. (Nearly half a second.) The expectation 

would be that the linear phase option would create higher latency, but that's not the case here. Additionally, the measured response 

changes dramatically. I'm not clear on how useful this would be as a low cut filter:

Odd, eh? As time permits, I'll do some further testing on this one.

  Back to fader8's audio blog Index