High Frequncy Roll-off Test: Pultec Pro v.s. Neve 1073

[Arys Chien]

I just did a simple test with the spectral analyzer, comparing the rolling off above 16khz when the Pultec Pro and the Neve 1073 were inserted with original setting (no knobs touched).

Pultec Pro: -5dB above 16khz
Neve 1073: -1~2dB on and above 16khz

And it did sound this way.

(Please mind that the \"initial volumn change\" of the two plug-ins are different, and have been taken into consideration. The Pultec Pro, when inserted, boosts the RMS level 0.4dB up, while the Neve 1073 cuts 0.8dB.)

 

[pogo]

I never noticed the cut by the Pultec (it obviously sounds different but, in a flattering way) until I tried the 1073 and switched to the Pultec on the same track. I've always loved that the Pultec made my voice almost tolerable to listen to. I was blinded by it's beauty... I really need to spend more time training my ears

 

[electro77]

I just did a simple test with the spectral analyzer, comparing the rolling off above 16khz when the Pultec Pro and the Neve 1073 were inserted with original setting (no knobs touched).

Pultec Pro: -5dB above 16khz
Neve 1073: -1~2dB on and above 16khz

And it did sound this way.

(Please mind that the "initial volumn change" of the two plug-ins are different, and have been taken into consideration. The Pultec Pro, when inserted, boosts the RMS level 0.4dB up, while the Neve 1073 cuts 0.8dB.)

What are you using to test this? Pink noise, white noise, sine wave?

 

[Arys Chien]

Pink and White Noise both.

 

[svs95]

I am sure UA are aware of this, and Arys is entirely correct. It is there, and it is audible - but only with 44.1kHz sources (and only on certain material with plenty of HF content in that spectral range).

It's one of the many reasons I try to work at higher sampling rates as much as possible. It also makes the workaround for the Pultec obvious, if you find it necessary. Just upsample, use the Pultec, then downsample with a higher quality converter than the one in the Pultec.

The Pultec already uses up/downsampling, of course, but the anti-alias filter is a bit too shallow by current standards. Still, it's better than what was built into most DAW's until their most recent versions.

I'm pretty sure UA are aware of the problem, and working on it. The resampling technology developed for the 1073 could find its way to other plug-ins in the future.

Having