Loopy Pro: Create music, your way.

What is Loopy Pro? — Loopy Pro is a powerful, flexible, and intuitive live looper, sampler, clip launcher and DAW for iPhone and iPad. At its core, it allows you to record and layer sounds in real-time to create complex musical arrangements. But it doesn’t stop there—Loopy Pro offers advanced tools to customize your workflow, build dynamic performance setups, and create a seamless connection between instruments, effects, and external gear.

Use it for live looping, sequencing, arranging, mixing, and much more. Whether you're a live performer, a producer, or just experimenting with sound, Loopy Pro helps you take control of your creative process.

Download on the App Store

Loopy Pro is your all-in-one musical toolkit. Try it for free today.

Most CPU friendly IR loader AUv3?

jacoujacou
in General App Discussion

I’m looking for the most cpu friendly IR loader AUv3.
It would be great if it had a pre-delay function and if it also ran on iPhone.

I want to use it with a good spring reverb stereo IR wav. All spring reverb apps either don’t have that fun splosh/drip sound or are too expensive.

So far I tried out Nembrinis IR Loader and GMs Impulsation. Nembrinis IR Loader is significantly more CPU friendly compared to Impulsation but doesn’t have a pre-delay knob.

Anyone here tried out the other IR loaders? I saw there’s also Thafknar, JAX Convoluter PE, etc
Any tip or suggestion is welcome.
Thank you!

Comments

  • jacoujacou

    I just thought maybe it’s possible to just manipulate the IR wav file and put some silence in the beginning. That should have the same effect like a pre delay knob in the IR loader I guess.
    Then I can use the most CPU friendly IR loader with a minimal feature set. Yay!

    Anyone here know a very CPU efficient IR loader AUv3?

  • pbelgiumpbelgium

    I have Nembrini IR Loader, Impulsation and AltiSpace 2. My testing showed Impulsation (~13%) to be a little more CPU friendly than IR Loader (~16%), and AltiSpace 2 (~22%) to be the least CPU friendly. Sorry, not much help. Isn't Thafknar free?

  • jacoujacou
    edited November 2025

    Oh thank you for testing! :)
    That’s quite surprising! What audio host app are you using? I’m using AUM on an iPad M1. They might handle AUv3s differently.

    IR Loader was quite a bit more efficient than Impulsation on my system. Nearly by 10%. My guess is that the host apps distribute the work load differently between the cpu cores.

    I will try out Thafknar.

    Edit: It costs 3€ IAP to use IR files.

    Anyone here that can talk about Thafknars CPU efficiency compared to other IR loader AUv3s?

  • uncledaveuncledave

    @jacou said:
    Oh thank you for testing! :)
    That’s quite surprising! What audio host app are you using? I’m using AUM on an iPad M1. They might handle AUv3s differently.

    IR Loader was quite a bit more efficient than Impulsation on my system. Nearly by 10%. My guess is that the host apps distribute the work load differently between the cpu cores.

    I will try out Thafknar.

    Hi. AUM, like most AUv3 hosts, is definitively single-threaded. So there is no distributing of the workload between CPU cores. The system moves processing to different cores to equalize thermal stress. And remember that your M1 has two types of core, economy (low power) and performance. In the AUM DSP display, the first 4 cores are economy, the last 4 are performance. You're likely only using the economy cores, so there's a lot of performance in reserve.

    The computational load of the IR convolution process should be pretty much the same for all apps, unless they cut off the tail of the IR. Any extra processing is likely due to additional operations like filter or delay, which must be applied separately.

  • SquishySquishy

    @jacou said:
    Oh thank you for testing! :)
    That’s quite surprising! What audio host app are you using? I’m using AUM on an iPad M1. They might handle AUv3s differently.

    IR Loader was quite a bit more efficient than Impulsation on my system. Nearly by 10%. My guess is that the host apps distribute the work load differently between the cpu cores.

    I will try out Thafknar.

    Edit: It costs 3€ IAP to use IR files.

    Anyone here that can talk about Thafknars CPU efficiency compared to other IR loader AUv3s?

    There’s also Jamp IR (from the dev that made bleep machine). Also free. I struggled with that one, couldn’t make heads or tails of the sliders, but might be worth a shot. I mainly use NAM XT now, but it conSUMES cpu haha

  • jacoujacou

    @uncledave said:

    @jacou said:
    Oh thank you for testing! :)
    That’s quite surprising! What audio host app are you using? I’m using AUM on an iPad M1. They might handle AUv3s differently.

    IR Loader was quite a bit more efficient than Impulsation on my system. Nearly by 10%. My guess is that the host apps distribute the work load differently between the cpu cores.

    I will try out Thafknar.

    Hi. AUM, like most AUv3 hosts, is definitively single-threaded. So there is no distributing of the workload between CPU cores. The system moves processing to different cores to equalize thermal stress. And remember that your M1 has two types of core, economy (low power) and performance. In the AUM DSP display, the first 4 cores are economy, the last 4 are performance. You're likely only using the economy cores, so there's a lot of performance in reserve.

    The computational load of the IR convolution process should be pretty much the same for all apps, unless they cut off the tail of the IR. Any extra processing is likely due to additional operations like filter or delay, which must be applied separately.

    Thank you for clarifying!
    So there is no audio host app that can distribute the CPU load better between the cores? Is it an iPadOS restriction?
    That really sucks. So I can get the latest expensive iPad but the difference in audio performance will always be minimal?

    I hope whoever is in charge of that will change that asap!

  • pbelgiumpbelgium

    @jacou said:
    Oh thank you for testing! :)
    That’s quite surprising! What audio host app are you using? I’m using AUM on an iPad M1. They might handle AUv3s differently.

    I used AUM on an iPhone 11 Pro - if you tap on the DSP% (top right) then 'Node Statistics' you can see the individual DSP percentages being used and CPU core distribution.

  • Philandering_BastardPhilandering_Bastard

    I tested the four IR loaders I have with an intentionally absurd wave file (75 seconds long). Results in descending order of CPU load:
    Thafknar: failed. Very efficient with normal IRs though.
    WoodVerberator: +20%
    Impulsator (within IFX Rack): +5%
    AltiSpace: +1%

  • uncledaveuncledave

    @jacou said:

    @uncledave said:

    @jacou said:
    Oh thank you for testing! :)
    That’s quite surprising! What audio host app are you using? I’m using AUM on an iPad M1. They might handle AUv3s differently.

    IR Loader was quite a bit more efficient than Impulsation on my system. Nearly by 10%. My guess is that the host apps distribute the work load differently between the cpu cores.

    I will try out Thafknar.

    Hi. AUM, like most AUv3 hosts, is definitively single-threaded. So there is no distributing of the workload between CPU cores. The system moves processing to different cores to equalize thermal stress. And remember that your M1 has two types of core, economy (low power) and performance. In the AUM DSP display, the first 4 cores are economy, the last 4 are performance. You're likely only using the economy cores, so there's a lot of performance in reserve.

    The computational load of the IR convolution process should be pretty much the same for all apps, unless they cut off the tail of the IR. Any extra processing is likely due to additional operations like filter or delay, which must be applied separately.

    Thank you for clarifying!
    So there is no audio host app that can distribute the CPU load better between the cores? Is it an iPadOS restriction?
    That really sucks. So I can get the latest expensive iPad but the difference in audio performance will always be minimal?

    I hope whoever is in charge of that will change that asap!

    I believe it's related to the specific audio processing thread, which gets special treatment for its real-time nature. I think apeMatrix might be able to use multiple cores. Cubasis uses multiple cores when freezing tracks. And there may be others. I wouldn't worry about it until you are maxing out the performance core.

  • espiegel123espiegel123

    @Philandering_Bastard said:
    I tested the four IR loaders I have with an intentionally absurd wave file (75 seconds long). Results in descending order of CPU load:
    Thafknar: failed. Very efficient with normal IRs though.
    WoodVerberator: +20%
    Impulsator (within IFX Rack): +5%
    AltiSpace: +1%

    If I recall correctly Altiverb and Impulsator truncate long IRs on import.

    What buffer size did you use? THAFKNAR doesn’t truncate IRs for such a long one, it might need a larger buffer.

  • Philandering_BastardPhilandering_Bastard

    @espiegel123 said:

    @Philandering_Bastard said:
    I tested the four IR loaders I have with an intentionally absurd wave file (75 seconds long). Results in descending order of CPU load:
    Thafknar: failed. Very efficient with normal IRs though.
    WoodVerberator: +20%
    Impulsator (within IFX Rack): +5%
    AltiSpace: +1%

    If I recall correctly Altiverb and Impulsator truncate long IRs on import.

    What buffer size did you use? THAFKNAR doesn’t truncate IRs for such a long one, it might need a larger buffer.

    Thafknar buffer was set at 4096 (max). AltiSpace cuts off at 15 seconds, WoodVerberator and Impulsation at 10 seconds.

  • espiegel123espiegel123

    @Philandering_Bastard said:

    @espiegel123 said:

    @Philandering_Bastard said:
    I tested the four IR loaders I have with an intentionally absurd wave file (75 seconds long). Results in descending order of CPU load:
    Thafknar: failed. Very efficient with normal IRs though.
    WoodVerberator: +20%
    Impulsator (within IFX Rack): +5%
    AltiSpace: +1%

    If I recall correctly Altiverb and Impulsator truncate long IRs on import.

    What buffer size did you use? THAFKNAR doesn’t truncate IRs for such a long one, it might need a larger buffer.

    Thafknar buffer was set at 4096 (max). AltiSpace cuts off at 15 seconds, WoodVerberator and Impulsation at 10 seconds.

    Based on that info, you should use a 10 second IR so that you are comparing computational efficiency fairly. The IR length is a major factor and needs to be constant between the plugins you are comparing.

  • Philandering_BastardPhilandering_Bastard

    @espiegel123 said:

    @Philandering_Bastard said:

    @espiegel123 said:

    @Philandering_Bastard said:
    I tested the four IR loaders I have with an intentionally absurd wave file (75 seconds long). Results in descending order of CPU load:
    Thafknar: failed. Very efficient with normal IRs though.
    WoodVerberator: +20%
    Impulsator (within IFX Rack): +5%
    AltiSpace: +1%

    If I recall correctly Altiverb and Impulsator truncate long IRs on import.

    What buffer size did you use? THAFKNAR doesn’t truncate IRs for such a long one, it might need a larger buffer.

    Thafknar buffer was set at 4096 (max). AltiSpace cuts off at 15 seconds, WoodVerberator and Impulsation at 10 seconds.

    Based on that info, you should use a 10 second IR so that you are comparing computational efficiency fairly. The IR length is a major factor and needs to be constant between the plugins you are comparing.

    Noted.

  • ecouecou

    I always show up to mention ChowDSP BYOD

Sign In or Register to comment.