Algorithmix compressor & Flux DSP overs

[Graemme]

Potentially better news:

As of a few hours ago, there is now a '2.0.27' version up on the Flux user download site. The fixes indicated are for an 'over-installation' problem and the 'de-normalling' issue that's been causing the DSP 'overs' we're seeing.

Best,

Graemme

[Jonathon Stevens]

Downloaded and installed the new versions this morning and I can confirm Epure2 and Limiter2 no longer show VST overs when playback is stopped. Very lovely!

[DubStudio]

I'm a new owner of the Algorithmix LP split comp (great compressor btw). I've noticed there's a significant increase in the DSP core usage when I open the GUI for the plug-in. The Load also seems to become more erratic, jumping around and sometimes causing dsp core overloads. This may be more of a general question, but is this normal behavior from a VST plugin for the GUI to consume additional resources like this?

I am having the exact same problems with the LP split comp. With the GUI open, I get a popping sound occasionally, and when I close it - problem goes away. I have tried changing the graphics card to an ATI Radeon HD 4550 and no luck. Jonathon, did you get this sorted in the end? Is the LPSC working now? Any recommendations?

[Jonathon Stevens]

It's been a while since i've used the LP split comp, but I think there was an improvement when i went to the 4350 video card. Sorry no other real advice. I'll give it a go again soon and let you know.

[Graemme]

The Flux Alchemist GUI also seems to use a lot of system resources and sometimes the audio is affected when the plug-in GUI is open. I also changed to the same video card that Jonathon is using and I have not noticed this issue since then.

Speaking of Flux, in the latest build (2.40.4000) the 'de-normai' problem is back, leading to DSP overs when there is no signal feeding the plug-in (noticed in Pure Limiter II). I worked around this last night by feeding a pink noise generator onto the same bus, with a level of -143.6dBfs. Of course, that completely changed the sound of the project, so i had to turn it off on mixdown...

Graemme

[Jonathon Stevens]

I've noticed the return of the denormaling problem too but haven't filed a bug report on their website yet, I switched to using PSP Xenon cause I was starting to get irritated with it.

[Graemme]

I've noticed the return of the denormaling problem too but haven't filed a bug report on their website yet, I switched to using PSP Xenon cause I was starting to get irritated with it.

Really?

Please tell me how you've been able to make the Xenon plug as (or nearly) as colourless as the PureLimiter?!

gb

[tas]

a tad off topic, but I thought try out the freeware algorithmix TT dynamic range meter plugin, only to find out that it produces crackle in the audio, which doesn't seem to be drop-outs but some other problem. Perhaps someone else has tried it? Not very important as you get the same functions from pmx vs3 plugs

[fl]

a tad off topic, but I thought try out the freeware algorithmix TT dynamic range meter plugin, only to find out that it produces crackle in the audio, which doesn't seem to be drop-outs but some other problem. Perhaps someone else has tried it? Not very important as you get the same functions from pmx vs3 plugs

On my Native system, I found that the Algorithmix TT plug to be a real CPU hog, loading the system so much it caused the kind of crackles you mentioned. It's a marginally useful plugin, for when you want to get some idea of the crest factor of your mix, but once testing is done, I take it out of the mixer before I do any mixdown or CD imaging.

[Jonathon Stevens]

@ Graemme: I'm not making any comparisons of sonic quality, but it doesn't give me those scary looking VS3 overs when the clients leaning over my shoulder saying "what's that flashing red thing, that doesn't look good".

[DubStudio]

I tried just about everything I could to solve this problem but it turns out the only way to really get rid of the clicks and pops was to replace the Lynx AES16e with an RME AES-32. The two sound cards are really pretty similar, except that the RME uses two spaces (but only one pci or pcie slot) and uses standard breakouts (D-Sub, TASCAM pinouts), whereas the Lynx card uses their own breakouts (presumably so they can fit the word clock i/o on the same card). The real difference is in the drivers though, RME have a maximum buffer size of 4096 samples, which makes it ideal for using CPU-intensive VST plugins in Pyramix like the algorithmix LPSC and PEQ. The Lynx AES16e only goes up to 1024, so I would not recommend using it with this combination of software. I was told by Lynx that they might be increasing the buffer size at some point, I waited a year and nothing happened so I gave up waiting and went for an RME card - problem solved.

[reenfoo]

replacing the Lynx AES16e with an RME AES-32 works !!

[DubStudio]

Yes and the RME is generally more stable than the Lynx. The Lynx would cause PMX to display an "ASIO configurations" box every time I launched it which became very tiresome.