Hi,

I'm working on my first sample-based instrument in HISE. I’ve sampled an electronic piano and I’m now at the stage where I need to edit and trim the samples properly.

The thing is some of the samples have very long tails, in the spectrogram, I can still see content going down to around -115 dB. I'm unsure how far I should go when trimming those tails.

My initial take was to cut around -90 dB and apply a short fade-out to keep things smooth, but I’m not sure if that’s the best practice.

I'm really curious to hear how more experienced users approach this, especially when it comes to how low you usually go in dB when trimming tails. Do you have a typical threshold you stick to ?

I feel like this kind of information could also be useful for other beginners.

Thanks in advance

@d-healey Thank you !

Hi,

I'm using the spectrum analyser from the Parametric EQ effect, but I noticed that it freezes, it continues to display peaks or shapes even when there is no audio signal.

Reducing the decay time helps, but it only hides the problem, I’d like to keep a long decay for a smoother visual response, but still have the analyser clear completely when there's no input signal.

I've searched through some older forum topics, but I couldn't find a clear solution to this issue.

@Orvillain
Oh ok, my bad then. Maybe it’s your sample that’s below 0dB, so it goes up and then drops back down. Anyway, glad it’s working for you now!

@Orvillain

From the doc :
"this value will be used as start ramp value whenever the node is being resetted (eg. when the network is unbypassed)."

I suppose that setting a smoothing value along with a resetValue of -100 dB will create an attack-like effect, gradually ramping the signal up to the target gain after a bypass or a initialize start

Hi everyone,

I’ve been going through the Stereo Widener example in the HISE documentation, and I noticed something that seems strange for me: the routing.ms_decode node appears before the routing.ms_encode node in the signal chain.

However, according to the documentation for routing.ms_encode, it should be placed at the beginning of the chain to encode a stereo signal into Mid-Side, which can then be processed and later decoded with routing.ms_decode.

From Doc about ms_encode : "To use the routing.ms_encode node, you would typically place it in a processing chain with the stereo input you want to encode, and then use a container.multi node to split the signal into its mid and side components. You can then process the components separately and recombine them using a routing.ms_decode node placed later in the chain."

This seems like a contradiction

Image from Hise Documentation :

@modularsamples

Thank you! That was very helpful.

I just have one last question:
If I want to modulate the gain using the LFO from GlobalModulator Container (the one in your snippet), how can I make the last gain node from my screenshot effectively control the gain of the AudioLoopPlayer1, which is located higher in the module tree?

Hi,

I'm wondering if it's possible to make the Global Modulation LFO start or retrigger only when an audio signal is detected, like when sound enters.

@Lindon
Mhh, I'm not sure I fully understand what "compiling the project" means exactly.
I get the part about compiling HISE itself, but I thought compiling a project was just the same as exporting it as a VST or instrument.

Anyway, I simply created a new project from scratch and now 5sec delay with jdelay in scriptnode works!
Thanks a lot!

@Lindon
Sorry, I'm having an issue again, I must be doing something wrong.
I tried compiling with the extra definition in JUCE, and I can see that the extra delay time is recognized in the Help > Toolbar info.

However, as soon as I use jDelay with a delay time above 3 seconds, HISE crashes.
.

Do I just need to write the line in the HISE settings?

But then, how do I compile it so that it actually takes effect?

@Lindon

Thanks!

@Lindon
Thanks!

But sorry , I still can't figure out how to access or modify HISE_MAX_DELAY_TIME_SAMPLES.

For example, in scriptnode, when using djDelay, how exactly do I access the extra definition you're referring to?

Hi,

I'm experimenting with different delay effects in scriptnode, (jDelay, core delay and i tried build-in FX Delay). However, I've noticed that no matter what I do, the delay time doesn't go beyond 1 second.

For example, even if I set the built-in delay to 20,000 ms, the actual delay still sounds like it's capped at around 1 second. The same thing happens with jDelay and core delay.

Is there a known limitation on delay time in these effects? Or is there something I'm missing in the configuration?

Thanks in advance,

@Mighty23 said in fix_block - I can't understand it:

Can I have some concrete examples?

For example, if you have a very fast LFO, using a frame_block is probably more interesting than a fix8_block to achieve more precise and smoother modulation.

As I understand it, the fix_block node allows you to force a specific block size for processing (e.g. 2, 4, 8, etc.). This can be useful for certain modulations, where a more regular update frequency improves rendering accuracy or stability.

For example, a fix8_block will divide the signal into blocks of 8 samples and execute the child modules at each block. In comparison, a frame_block performs the processing for each sample, i.e. 8 times more often in this example, which improves accuracy but greatly increases the CPU load.

I'm not sure if this is very clear but I hope it helps you understand.