10-Band PLParEQ is Here!

[cjp]

Hi.

The colour switching thing happens in FL Studio whenever a parameter is changing. If you select the dot in the graph and hold the mouse button down the flickering happens. Could just be a graphics card thing at my end though. Also the lines in the background of the graph flicker as well (between solid and transparent). Either way it's probably not something worth woring about, crashes etc. are way more important.

Doesn't stop it from sounding fantastic (which is does!)

I can't get any other lowpass to go as low as PLParEq, it's ruined all my other EQ/filters for me

Thanks again.

(and yeah, I couldn't give a shit about the GUI either, but some potential clients might)

[dbmcclain]

Yep, I though about doing the separate L/R filtering... that's exactly what M/S is doing internally -- L is assigned to M, and R is assigned to R.

But I thought, "Surely nobody wants to treat L and R separately...." think again...

Well it is an option that would be easy to provide. But the trouble you run into is that if you mix M/S with L/R processing in the same batch of filters, you will have to go through two separate passes of the filtering operation. It only means more CPU usage...

( and then we have to come up with better colors.... I resisted Blue and Red for the M/S processing because I thought it would be too misleading...)

[dbmcclain]

I'll take another look at the flicker problem. I solved it for when sliding the mouse in the text boxes. But the graph is another story...

Anyone else having problems with Cubase SX? We have a persistent problem appearing with one user, and I'd like to get to the bottom of it.

Basically he can run the GUI-less version just fine, but any of the variants with GUI's cause a horrific crash on his system. Anyone?

[pitchdrifter]

David,

you could offer a sort of master switch for input type selection. E.g. input: Stereo/L/R or Stereo/M/S;
And all it would do is disable the internal M/S encoding and decoding when chosing input:stereo/M/S, everything else could stay the same gui-wise.

best

Markus
P.S. having dropout issues with the plpareq (SX3) here, however only when cpu load is getting to exactly 50% (it's a dual opteron setup) or 100%. In between it works fine.

[dbmcclain]

Yes, the master switch would be a good way to go without overwhelming the computer with gobs of computations. Seems a shame to limit it like this, but we are bumping the limits of available CPU cycles....

Version 1.55 now available -- fixes the graph flicker problem when dragging the curve knob...

[djsubject]

http://refinedaudiometrics.com/pages/16/index.htm

thats the link for late-comers of this thread


PS; this is without a doubet the best free eq in the world

makes me wish i could afford the big one

Subz

[OMU]

Waves LinEq is priced around 300 E... And Waves are well known for not being cheap.

And you can't say the mastering engineers didn't buy it.

The whole Waves Mastering Bundle costs 900 E.

Although it may be great sounding, your eq could be less expensive than 1000. I don't agree that everybody should take the Algoritmix pricing route.

[pitchdrifter]

Yes, the master switch would be a good way to go without overwhelming the computer with gobs of computations. Seems a shame to limit it like this, but we are bumping the limits of available CPU cycles....

Version 1.55 now available -- fixes the graph flicker problem when dragging the curve knob...

I wouldn't consider the ability to turn off the m/s encoder+decoder limiting to be honest. Everything else stays the same doesn't it?

[pitchdrifter]

addition to the dropout issue:

when getting above 50% cpu load on this dual opteron system occasional dropouts happen, the eq itself takes up between 10 and 20% cpu, the overall project is at 65-70% cpu during playback. Enabling PL for a single band will cause plenty of dropouts to appear, even though the cpu load of the entire project is still at 65-70%, no overs. I can go up to 99% cpu load without dropouts usually (that is, unless there's an overload it runs without dropouts, so the 99% are theoretical/peak, as an average the figure is closer to 90-95%).

[dbmcclain]

Try increasing your ASIOI buffer size. It can make a huge difference.

[Stupid American Pig]

... why it costs what it does... how about trying to recoup more than 6 years of intense research costs? How about because it is better than anything out there, including the nearest competitor that charges $1,500 USD for theirs? How about, because I'm a nice guy, and my business partners think we should be selling for closer to $6,000 USD to compete more in line with Massenburg and Weiss?

how about...

Its yer world man, I wasnt trying to come off as an ass. I was genuinely curious why you thought it was worth that price is all. Best of luck in yer new endevor.

[pitchdrifter]

David,

it's not the asio buffer sizes fault that these dropouts occur. As I mentioned, at the current buffer size I can go up to averaged cpu load of 90-95%, with stable dropout-free performance. And with the plpareq the dropouts start happening as soon as cpu reaches 50%.

[dbmcclain]

Hmmm.. .what kind of CPU? Intel or AMD? I have limitations here in the Receptor that don't appear in the Pentium.

[pitchdrifter]

dual opterons as mentioned above, so AMD

best

Markus

[Aleksey Vaneev]

So, we have another round of competition here?

Probably, it's time for me to start thinking about alike linear-phase filter (which is analog linear-phase - not zero-phase like CurveEQ or other digital filters). Of course, it can't be 100% linear phase, but who cares? I'm just wondering why it takes so much CPU...

Time to create IIR filter optimization procedure and find some new peaking filter shapes.

So, everyone forgot about Hydratone for a moment? I remember there was as much discussion as on this one. But at $1000 price tag that won't go too far, I'm afraid.