three-bands EQ and delay

[nicop]

Hi, in the process of conversion of my drums library, I realized that I also need a three-band EQ and to delay a sample.
I think the opcodes involved should be <delay> and <eq1_freq> <eq1_bw> <eq1_gain>; <eq2_freq> <eq2_bw> <eq2_gain>; <eq3_freq> <eq3_bw> <eq3_gain>.
Could someone implement these opcodes?
Now I can only donate 50 euros, but I'm planning to do another donation for the development of the "internal effect" feature.

thanks

Nicola

[varpa]

This is a completely naive comment, but I'm assuming the the EQ op codes get the sampler engine to provide equalization. Wouldn't it be more efficient to process your samples with whatever effects you want and just use those samples? It seems like it would save the sampling engine CPU time and hence work better. If the effects are easily parametrized I would think you could automate the processing of sample using, say, (in linux) a script and snd, ecasound, or lvl2file, for example. But maybe there is a good reason to use opcodes and dynamic processing of samples, and I certainly don't want to discourage you from contributing to linuxsampler development.

[nicop]

Hi Varpa, I understood your point.
I asked eq because I use LS with a lot of channels (more than 100), and potentially everyone of them could have a different eq. In this situation it would be very complex to manage, for example, 100 instances of an external eq effect.
Making a "pre-production" equalization with ecasound or something similar could solve the problem, but it will make the fine tuning a difficult task, because every time I want to modify a parameter I'll have to take the original sample and process it.
Maybe a solution would be to use the LS internal effects that Grigor started to develop. http://bb.linuxsampler.org/viewtopic.php?f=6&t=611
In that case I would still be happy to contribute to the internal effect's development too

[grishata]

Hi Nicola,
I've just added experimental support for per voice equalization - sfz opcodes eq1_freq, eq2_freq, eq3_freq, eq1_freqccN, eq2_freqccN, eq3_freqccN, eq1_bw, eq2_bw, eq3_bw, eq1_bwccN, eq2_bwccN, eq3_bwccN, eq1_gain, eq2_gain, eq3_gain, eq1_gainccN, eq2_gainccN, eq3_gainccN.
Note that the current implementation does not provide internal EQ but is looking for specific EQ effects to use - currently it only looks for the triplePara (Triple band parametric with shelves) LADSPA plugin, which is part of the swh-plugins package (http://plugin.org.uk/). I guess at some later point I'll add a LSCP support for setting the desired EQ plugin to be used for per voice equalization.
Let me know if the current implementation works OK for you.

BTW, Christian was the one who implemented the internal effects in LinuxSampler, I added support for configuring them through Fantasia.

[grishata]

Note that in Ubuntu you can install the swh-plugins package with the command:

Code: Select all

sudo apt-get install swh-plugins

[grishata]

I also implemented the opcodes delay, delay_onccN, delay_random, delay_samples, delay_samples_onccN.

[nicop]

Thank you soooo much
All works flawlessly.
I'll give a try to eq*_*ccN, it would be very interesting to change the parameters in realtime!
I've donated only 25 euros because I ran out of money on my paypal I'll donate the rest when I'll recharge it.

thanks again

[grishata]

I've donated only 25 euros because I ran out of money on my paypal I'll donate the rest when I'll recharge it.

No problem. When you can.
Thanks.