KVR Audio

IIRs · Post by **IIRs** » Tue Jul 11, 2006 2:24 pm

Interesting!

I'm getting a lot of crackling when adjusting parameters on the LP one though (in Tracktion 2.1) and.. is there any reason why the Q range is reversed?

IIRs · Post by **IIRs** » Tue Jul 11, 2006 2:26 pm

And indeed, why such a limited range of Q values for the low 'Baxandall' band..?

--= Ju =-- · Post by **--= Ju =--** » Tue Jul 11, 2006 4:20 pm

Getting some wierd results here. In EnergyXT with this plugged into the output of Battery2 with everything set at default I get pops / bass rumble when it plays a breakbeat sample. This disappears if I disable the low shelf or play the sample at any pitch other than its root. Anyone else getting similar results or should I call for the men in white coats?

docdued · Post by **docdued** » Tue Jul 11, 2006 4:48 pm

IIRs wrote:Interesting!

I'm getting a lot of crackling when adjusting parameters on the LP one though (in Tracktion 2.1) and.. is there any reason why the Q range is reversed?

And indeed, why such a limited range of Q values for the low 'Baxandall' band..?

Indeed, most of the work is done when a parameter is updated, so if you have a small soundcard buffer or a slow CPU, this might lead to problems. Your remark actually made me think about the code again, and I think I have made a slight improvement. Could you download it again and see if it still crackles?

The Q is reversed because I saw most other EQs do it that way. I haven't followed that philosophy in the IIEQ though

I wanted the low pass filter to act really only in the low frequency regime. I could make the available Q values lower, though, to allow for a broader filter, if that would be helpful for you. To go to higher Qs doesn't really make sense, as then we would start influencing only frequencies below 20 Hz or so, i.e. there would be basically no audible difference.

TotcProductions · Post by **TotcProductions** » Tue Jul 11, 2006 5:10 pm

Thank you DocDued. Might be picking up LP10 along the way as a track eq.

Peace

IIRs · Post by **IIRs** » Tue Jul 11, 2006 5:18 pm

docdued wrote: Your remark actually made me think about the code again, and I think I have made a slight improvement. Could you download it again and see if it still crackles?

Repeating the same test as before (dropping it into the mix buss of a fairly cpu intensive mix) resulted in similar crackling, but freezing a few tracks to drop the cpu load cured it. I didn't try freezing with v.1 so I can't say if that's an improvement, sorry!

docdued wrote: To go to higher Qs doesn't really make sense, as then we would start influencing only frequencies below 20 Hz or so, i.e. there would be basically no audible difference.

Sometimes you need to cut that low rubbish out of a mix without affecting the audible range.. no big deal though. Add a bypass for each band and you probably have a sale

docdued · Post by **docdued** » Wed Jul 12, 2006 8:29 am

Well, I just realized that in one of the recent code opimizations I must have accidentally removed the output silence thingy... ah, what the hell, consider it your lucky day. Will be working on the band bypasses, though.

IIRs · Post by **IIRs** » Wed Jul 12, 2006 8:47 am

docdued wrote:Well, I just realized that in one of the recent code opimizations I must have accidentally removed the output silence thingy...

I wondered about that actually! Still happy to pay for bypass buttons though

Here's another suggestion that just occured to me: how about rationalising the two plugs so they both offer the same control set, ie: add a 'Baxandall' type mode for the low & high bands in the minimum phase plug, and a conventional shelf mode for the same bands in the linear phase plug.

Then: add a simple patch saving / loading function so you can set up an EQ curve with one of them, and then load it into either.. that way you could use the minimum phase EQ for tracking, then at mixdown you could quickly and easily try out a linear phase version of the same filter to see if that worked better.

I would suggest putting both type of EQ into the same plug, but I guess that would cause problems with PDC..

docdued · Post by **docdued** » Wed Jul 12, 2006 10:59 am

IIRs wrote:
I wondered about that actually! Still happy to pay for bypass buttons though

Go ahead, they're implemented...

Here's another suggestion that just occured to me: how about rationalising the two plugs so they both offer the same control set, ie: add a 'Baxandall' type mode for the low & high bands in the minimum phase plug, and a conventional shelf mode for the same bands in the linear phase plug.

Then: add a simple patch saving / loading function so you can set up an EQ curve with one of them, and then load it into either.. that way you could use the minimum phase EQ for tracking, then at mixdown you could quickly and easily try out a linear phase version of the same filter to see if that worked better.

I would suggest putting both type of EQ into the same plug, but I guess that would cause problems with PDC..

Nice idea, I'm gonna think about it. The next few days I am busy with other stuff though, so don't expect anything to soon.

IIRs · Post by **IIRs** » Wed Jul 12, 2006 12:01 pm

docdued wrote:
Go ahead, they're implemented...

Ok, I bought a licence.

The demo now inserts gaps as it should.. so hurry up with that serial number

oblagon · Post by **oblagon** » Wed Jul 12, 2006 2:37 pm

docdued wrote:Well, I just realized that in one of the recent code opimizations I must have accidentally removed the output silence thingy... ah, what the hell, consider it your lucky day. Will be working on the band bypasses, though.

Would turning bands completely off reduce latency and CPU consumption further ? Using LP 10 as a track eq introduces quite a delay.

docdued · Post by **docdued** » Wed Jul 12, 2006 4:11 pm

No, for two reasons: first, the latency has nothing to do with CPU load, but is solely caused by the way the linear phase algorithm works. No matter whether you run it on a 50 MHz or a 4 GHz PC, the latency will always be the same (that is why the LP10 should only be used as a track EQ if your host compensates for the latency automatically). You will hear crackling, though, on the 50 MHz machine because it is not fast enough to deliver the filtered samples in real time. However, this CPU load is, in the case of the LP10, not reduced when turning off single bands. The on/off knobs are just a convenient feature to quickly A/B things.

2NDMOUSE · Post by **2NDMOUSE** » Wed Jul 12, 2006 11:28 pm

Docdued --- thanks a mil !! I use EQ a lot on my material ( usually one on every mixer track ).
Ive downloaded this , but I've yet to try it out ( in FL6 )
Many thanks for the hard graft and allowing it to be a freebie !!

Hovmod · Post by **Hovmod** » Thu Jul 13, 2006 12:06 am

Thanks! I downloaded the IIEQ, and I'm giving it a go now.
I'm getting a loud rumble from the low shelf when there's no audio coming through. Disabling the lowest band hushes it up.

I'm also getting some pops and ... uh ... stepping effect while turning the gain knobs, so I wouldn't be able to automate it. Once set the sound is pristine and I like it a lot.

Debutante · Post by **Debutante** » Thu Jul 13, 2006 2:18 am

I'm getting no pops or crackles OR *latency* in FL6. Aren't LP eq's supposed to give some kind of latency?!?!? It's great, but just curious

. Automation also seems to work very well here. Thanks for this one!

Is there a way to do micro-tweaking... like using a keyboard hold or something....?

Free 10-band parametric EQ from ddmf