I've just been testing 1.81 - i've been with PLP right from the beginning, and it's amazing the progress. This is starting to 'feel' like an expensive piece of software.
An unusual 'feature' which i've noticed: i've been testing PLP with a sample of white noise, duplicated and phase inverted. So the two tracks sum to mono. When insert a plugin on one of these tracks, I can hear any differences between them.
I'm pleased to report that with most of the filters the behaviour is exactly as expected - e.g. a low pass set to 30Khz, the signal is unharmed and phase cancels perfectly.
However - the All Pass filter is strange - it harms the signal, in other words they no longer cancel. Oddly enought, moving the frequency changes the difference - even at zero gain. And the gain control doesn't work like it used to - it used to change the gain, but now the curves stays flat.
Considering the surgical accuracy of the other filters, this basic All Pass filter would appear to have a bug or two. Unless it's not expected to be a 'straight wire with gain' ... but I would expect it to be that.
But apart from that minor detail, i'm loving it.
10-Band PLParEQ is Here!
-
- KVRian
- 1442 posts since 30 May, 2005
Hi greendoor,
this is no bug you're experiencing with the all pass filter there.
An all pass filters purose is to shift the phase of a signal while keeping the gain at unity. Therefore it doesn't make sense to have the gain parameter to do anything.
The all pass is useful for phaser fx. If you duplicate a signal and put an all pass filter on one of them you can hear some frequency cancellation due to the phase shift. If you now move the frequency (which actually moves the phase shift) of the all pass you get the phaser fx. Therefor the name: phaser!
All the best, FRitz
P.S. Could be that you now love it even more.
this is no bug you're experiencing with the all pass filter there.
An all pass filters purose is to shift the phase of a signal while keeping the gain at unity. Therefore it doesn't make sense to have the gain parameter to do anything.
The all pass is useful for phaser fx. If you duplicate a signal and put an all pass filter on one of them you can hear some frequency cancellation due to the phase shift. If you now move the frequency (which actually moves the phase shift) of the all pass you get the phaser fx. Therefor the name: phaser!
All the best, FRitz
P.S. Could be that you now love it even more.
In the end will be the word.
Check out some of my music at www.fritzmetal.de
Check out some of my music at www.fritzmetal.de
-
- KVRian
- 747 posts since 5 Sep, 2002 from Redondo Beach
I'm late to the party appearantly, but this is one fantastic EQ!!
I did a bit of a shootout earlier with some of the popular EQ plugs and PLPAR comes out ahead of the pack in my opinion.
I tested it against (in my decending order of preference)
Algorhythmics RED
UA Precision EQ
Waves Lin EQ
Voxengo Harmony EQ Demo.
I tested them in a mastering type situation on a stereo track. I used only 3 bands on all EQS with a 4db Low shelf boost at 100hz, a 2db bell curve cut at 250hz, and a 3db high shelf starting at 8K.
The PLPAR is the most transparent to be sure.
The algorhythmics is way less transparent, but does sound very pleasant. There seems to be a slight mids boost even when flat.
The UA precision EQ is transparent also and stands up well.
The waves is pretty good but adds a small amount of unpleasant harshness.
Voxengo was the least good sounding of the bunch. The low end sounded very strange.
Thanks for the great free 3 band EQ. Must test the 10 band demo now!!
Jeff
I did a bit of a shootout earlier with some of the popular EQ plugs and PLPAR comes out ahead of the pack in my opinion.
I tested it against (in my decending order of preference)
Algorhythmics RED
UA Precision EQ
Waves Lin EQ
Voxengo Harmony EQ Demo.
I tested them in a mastering type situation on a stereo track. I used only 3 bands on all EQS with a 4db Low shelf boost at 100hz, a 2db bell curve cut at 250hz, and a 3db high shelf starting at 8K.
The PLPAR is the most transparent to be sure.
The algorhythmics is way less transparent, but does sound very pleasant. There seems to be a slight mids boost even when flat.
The UA precision EQ is transparent also and stands up well.
The waves is pretty good but adds a small amount of unpleasant harshness.
Voxengo was the least good sounding of the bunch. The low end sounded very strange.
Thanks for the great free 3 band EQ. Must test the 10 band demo now!!
Jeff
-
- KVRist
- Topic Starter
- 236 posts since 5 Oct, 2005 from Tucson, AZ, USA
For those of your experiencing problems with PLParEQ3, if you are using an AMD-based machine, please bear with me...
I have a deadline on some human detection code for a TMS342 DSP coming up. As soon as that has passed I will jump into an AMD-specific port of the DSP core to see if that speeds things up a bit. This weekend will see that...
I have a deadline on some human detection code for a TMS342 DSP coming up. As soon as that has passed I will jump into an AMD-specific port of the DSP core to see if that speeds things up a bit. This weekend will see that...
-
- KVRAF
- 3846 posts since 15 Mar, 2002 from Underworld
Yeeeepiee! 
It is no measure of health to be well adjusted to a profoundly sick society. - Jiddu Krishnamurti
-
- KVRist
- 421 posts since 12 Jun, 2004
That would be nice; have heard from other AMD-Users, they also couldn't run the GUI-versions. So i am not the only one 
-
- KVRist
- Topic Starter
- 236 posts since 5 Oct, 2005 from Tucson, AZ, USA
The Muse Receptor is also an AMD-based machine. While I can run these GUI based system on that computer just fine, they really hog the CPU.
If you have a laptop or an older AMD CPU then this hogging could well interact with the rest of your system and bring it to its knees. E.g., I can't really run quality level 5 on the Receptor. Nothing bad happens, it just sends the CPU over 100% and I get lots of dropouts. But who knows what it would do to any other host?
If you have a laptop or an older AMD CPU then this hogging could well interact with the rest of your system and bring it to its knees. E.g., I can't really run quality level 5 on the Receptor. Nothing bad happens, it just sends the CPU over 100% and I get lots of dropouts. But who knows what it would do to any other host?
-
- KVRian
- 943 posts since 15 Mar, 2005
harmonieq is too different really for comparison imo.Guitarjeff wrote:
Voxengo was the least good sounding of the bunch. The low end sounded very strange.
Thanks for the great free 3 band EQ. Must test the 10 band demo now!!
Jeff
-
- KVRian
- 1121 posts since 8 Oct, 2005
no that anyone has been paying attention to my posts 
(poor me) but the 1.81 update has made a quantum leap performance wise for me (sonar 4). i don't have the stop/start problems anymore and a boost/cut of just 1db is absolutely phenomenal.
if i cud spend $1k i would buy the 10 band today.
edit: oh yeah, up to 54% cpu...this soundz so good i don't give a hoot. i know that's a terrible thing to say for future improvement, but i trust david to take care of that.
(poor me) but the 1.81 update has made a quantum leap performance wise for me (sonar 4). i don't have the stop/start problems anymore and a boost/cut of just 1db is absolutely phenomenal. if i cud spend $1k i would buy the 10 band today.
edit: oh yeah, up to 54% cpu...this soundz so good i don't give a hoot. i know that's a terrible thing to say for future improvement, but i trust david to take care of that.
-
- KVRist
- Topic Starter
- 236 posts since 5 Oct, 2005 from Tucson, AZ, USA
Hi politcat, I was hoping that sooner or later these speed improvements would make the difference for you. Glad to hear it!
-
- KVRist
- Topic Starter
- 236 posts since 5 Oct, 2005 from Tucson, AZ, USA
I'm doing some tests here...
(shame on me.. i have a deadline approaching!!)
Running the 10 band at QL5 with now doubled buffer sizes for the 96 kHz quality at the low freqs (16384 samps). Pushing through 10 phase linear filters with at least some doing M/S filtering and using 44.1 SR -- this is the toughest possible case for timing...
Doing all 10 bands loads my CPU down by about 23%.
(shame on me.. i have a deadline approaching!!)
Running the 10 band at QL5 with now doubled buffer sizes for the 96 kHz quality at the low freqs (16384 samps). Pushing through 10 phase linear filters with at least some doing M/S filtering and using 44.1 SR -- this is the toughest possible case for timing...
Doing all 10 bands loads my CPU down by about 23%.
-
- KVRian
- 747 posts since 5 Sep, 2002 from Redondo Beach
AMD Running good here too. The 3 band takes about 10% of my CPU at 24/96 in Nuendo 3.1.
I have an Athlon 4000+ with the Asus A8V mobo. 2 gigs RAM.
I have an Athlon 4000+ with the Asus A8V mobo. 2 gigs RAM.