VPS Avenger 2.4 [RELEASED - FAQ in OP]

[m-ac]

Same question for m-ac, do you have the synth? Because if not, I don't see what is your base to say the synth has still issues (except than a guy not having the synth previously said it, it is a never ending story).

Yep, and I've posted several video clips, screenshots, and detailed tests over the past few years in both the threads on here. I'm telling the OP that some people who use the synth still have CPU spike issues in use, and they've posted in this very thread. You're telling them that there are no issues with CPU spikes which isn't true and could lead to some unhappy purchases.

Look, Im not sh!tting on the synth. I love it and I've used it since before release when I designed one of its original skins. I know of at least 8 users who have purchased it due to my recommendation, but I always warn them how it performs in an OSX + logic scenario (I don't use windows, but it seems to perform better there). The fact I'm still willing to buy expansions shows how I feel about Manuel and Renes work, and I just work around the CPU issues when they arise.

[Jac459]

Same question for m-ac, do you have the synth? Because if not, I don't see what is your base to say the synth has still issues (except than a guy not having the synth previously said it, it is a never ending story).

Yep, and I've posted several video clips, screenshots, and detailed tests over the past few years in both the threads on here. I'm telling the OP that some people who use the synth still have CPU spike issues in use, and they've posted in this very thread. You're telling them that there are no issues with CPU spikes which isn't true and could lead to some unhappy purchases.

Look, Im not sh!tting on the synth. I love it and I've used it since before release when I designed one of its original skins. I know of at least 8 users who have purchased it due to my recommendation, but I always warn them how it performs in an OSX + logic scenario (I don't use windows, but it seems to perform better there). The fact I'm still willing to buy expansions shows how I feel about Manuel and Renes work, and I just work around the CPU issues when they arise.

Ah ok.
Apologies then, I was in the wrong assumption that you only got the v1.

My bad, I was more reacting to some past discussion where some people not having the synth were pretending to know better.

Again, my apologies.

So what I can say is that on apple silicon (M2 pro) and Bitwig I have absolutely no issue. The synth is snappy, fast UI, normal CPU usage, no spike, and having a lot of instances in // is absolutely not a problem.

[kraster]

It spikes pretty bad for me on Ableton 11 on a Mac M1.

[Jac459]

It spikes pretty bad for me on Ableton 11 on a Mac M1.

Are you in v2.0.5???

[kraster]

It spikes pretty bad for me on Ableton 11 on a Mac M1.

Are you in v2.0.5???

Yeah.

[Jac459]

It spikes pretty bad for me on Ableton 11 on a Mac M1.

Are you in v2.0.5???

Yeah.

Oh that's weird. Any sounds you want to point me to? I can check tonight in my setup

[zvenx]

I too have 2.0.latestVersion
I honestly haven't used it or don't use it enough to say all spiking has gone. But for sure it has been reduced compared to the earlier versions and for sure the v1.

And yes I had v1 for many years and bought v2 upgrade on day 1.
rsp

[kraster]

It spikes pretty bad for me on Ableton 11 on a Mac M1.

Are you in v2.0.5???

Yeah.

Oh that's weird. Any sounds you want to point me to? I can check tonight in my setup

Nearly every sound on load spikes the CPU.

There's a Pad sound called "PD Heaven's Gate" that runs at 20%. It's not a particularly complex sound.

That's with a four note chord.

20% is a lot for that.

[zvenx]

I assume by spiking they don't necessarily mean consistent high cpu, just, you know, spikes of high cpu that are audible.
rsp

[kraster]

I assume by spiking they don't necessarily mean consistent high cpu, just, you know, spikes of high cpu that are audible.
rsp

I'd class a spike as something that triggers the CPU warning indicator in Ableton.

[zvenx]

Fair enough.. I didn't realise the 20% patch was indeed triggering that cpu warning indicator.
rsp

[Jac459]

It spikes pretty bad for me on Ableton 11 on a Mac M1.

Are you in v2.0.5???

Yeah.

Oh that's weird. Any sounds you want to point me to? I can check tonight in my setup

Nearly every sound on load spikes the CPU.

There's a Pad sound called "PD Heaven's Gate" that runs at 20%. It's not a particularly complex sound.

That's with a four note chord.

20% is a lot for that.

Well.

This is interesting you chose this preset because this is granular preset so it is easy to compare to another synth.

So what I did is I created a sustained 1 loop bar with a 3 notes chord.
For good measure I created 10 instances of Avenger 2 with the same setup.

This is the first run:

Screenshot 2024-03-04 at 9.56.53 PM.jpg

I am slightly below 40%.

Then I created the same patch in Pigments 5 (the version with improved CPU usage) with just the granular synthesis, the same number of grains (106/seconds) and the same grain duration (0.55 seconds).

Then the same track.
And this is the run:

Screenshot 2024-03-04 at 10.01.25 PM.jpg

Then I discovered it was unfair to Avenger because the Pigments patch was without any effects, no second oscillators, and filters disabled.
Avengers patch has 2 oscillators, with the second being a VA with wave shaping with additive synthesis with 20 bands modulations. 4 reverbs, multiple filters and EQ and more.
So I disabled everything and did a second run for avenger:

Screenshot 2024-03-04 at 10.06.16 PM.png

[Jac459]

The conclusion is simple.

On Apple Silicon, Avenger is in fact more efficient than Pigments.

2 remarks:
1 - I don't deny at all your experience, Avenger may have a problem with Ableton, I have no idea. What I am telling you is that on a normal setup without bugs, Avenger is flawless.
2 - I don't really agree with you when you say that PD heavens gate is a simple preset. It consists of:
- Granular synth with 50 (50 !!!!) samples running at the same time for one note)
- A simple VA but with 20 bands spectral modulation.
- A vintage phaser
- A Reverb
- A Transgate
- 2 filters,
- 1 Equalizer
- 1 Limiter.

PS: I just noticed that the granular part has a unison so that's 100 samples running at the same time (vs 50 for Pigments).
I also noticed than the VA has 3 unison and 2 voicing, meaning 6 voices per note.

So knowing that I have 30 notes, that's 3000 samples played in // and 180 notes of a VA played in //. With about 35% usage, I am tempted to say that it's a freaking awesome result...

[zvenx]

Is a "normal setup" synonymous to "your setup" for you?

I must admit your generalisations bother me.

All you really can say again is that in your setup (and of course if you know people with other setups that have no issues, surely you can say their setups too) and the way you work, it is fine....

Also no software is flawless...(well maybe the intro hello world C intro program). Maybe what you mean is in the way you use it, on your machine it is flawless.


rsp

[Jac459]

Is a "normal setup" synonymous to "your setup" for you?

I must admit your generalisations bother me.

All you really can say again is that in your setup (and of course if you know people with other setups that have no issues, surely you can say their setups too) and the way you work, it is fine....

Also no software is flawless...(well maybe the intro hello world C intro program). Maybe what you mean is in the way you use it, on your machine it is flawless.


rsp

.....

You really take the things the wrong way my friend...

I am not saying "normal setup" to tell you my setup is better than yours or I am the center of the universe... That's a weird accusation.

What I am trying to tell you and others is that the "normal" or "standard" (I don't know how to find a word that doesn't upset you) behaviour of Avenger 2 is to be quite efficient in terms of CPU.
Let's say you have a car. And you manage to go with it only at 60mph.Wouldn't you be interested to know that the "normal" or "standard" speed this car can go is 120mph ? That people are actually able to go with it at 120mph ?
For me it would be interesting in the way I troubleshoot my setup.

What I wanted to make clear. Is that it is not a question of appreciation, like there is a population more demanding, saying there is spikes, and there is a population, less demanding saying that there is not.

In fact what happens is that Avenger 2 is working perfectly fine and "normally" for some users and isn't for some others.
Now what is the ratio of population for which Avenger 2 is working well and the ratio of population for which there is spikes, I have no idea and this isn't my debate or my point at least in these few last posts.