# FLAC3D v7.141 speed - CPU wave behavior

**URL:** <https://forum.itascainternational.com/t/flac3d-v7-141-speed-cpu-wave-behavior/102>\
**Category:** FLAC3D\
**Tags:** faq\
**Created:** [September 20, 2021, 3:08pm UTC](https://forum.itascainternational.com/t/flac3d-v7-141-speed-cpu-wave-behavior/102 "2021-09-20T15:08:46Z")\
**Posts on this page:** 9\
**Page:** 1

<div class="post-metadata">

**Author:** ![ccancino](https://avatars.discourse-cdn.com/v4/letter/c/bcef8e/32.png) [@ccancino](https://forum.itascainternational.com/u/ccancino)\
**Post date:** [September 20, 2021, 3:08pm UTC](https://forum.itascainternational.com/t/flac3d-v7-141-speed-cpu-wave-behavior/102/1 "2021-09-20T15:08:46Z")

</div>

![image](https://canada1.discourse-cdn.com/flex031/uploads/itasca/original/1X/a8e7018e7a93a7619c10e1a17646904139e4d951.png)

Is anyone familiar with this kind of “wave” behavior? FLAC3D is only using about 40% of CPU resources due to this “wave” behavior. I have done few tests, concluding that it’s not related to fish-calls, and I have also closed the plots items, but I got the same answer.

PD: I am using AMD Ryzen Threadripper 3970X 32 Core Processor

Thanks in advance.

---

<div class="post-metadata">

**Author:** ![cheng](https://avatars.discourse-cdn.com/v4/letter/c/ecccb3/32.png) [@cheng](https://forum.itascainternational.com/u/cheng)\
**Post date:** [September 24, 2021, 2:14pm UTC](https://forum.itascainternational.com/t/flac3d-v7-141-speed-cpu-wave-behavior/102/2 "2021-09-24T14:14:22Z")

</div>

We just observed 100% CPU usage in i9 machines, separately tested by Jim and myself.  
More investigation are in progress.  
We suggest you run the speed test on this specific machine. See this website: [http://benchmark.itascacloud.com/](http://benchmark.itascacloud.com/).

---

<div class="post-metadata">

**Author:** ![ccancino](https://avatars.discourse-cdn.com/v4/letter/c/bcef8e/32.png) [@ccancino](https://forum.itascainternational.com/u/ccancino)\
**Post date:** [September 24, 2021, 2:54pm UTC](https://forum.itascainternational.com/t/flac3d-v7-141-speed-cpu-wave-behavior/102/3 "2021-09-24T14:54:52Z")

</div>

Joe,

Thanks for answering. First of all, this is not happening in one specific machine only, the wave behavior is happening in 6 different crunchers. Howie tested one of these crunchers, but it is working fine, with 100% CPU over 30 mins or so, but then when I tried running FLAC3D 7, the wave behavior started.

Christian

---

<div class="post-metadata">

**Author:** ![cheng](https://avatars.discourse-cdn.com/v4/letter/c/ecccb3/32.png) [@cheng](https://forum.itascainternational.com/u/cheng)\
**Post date:** [September 27, 2021, 8:32pm UTC](https://forum.itascainternational.com/t/flac3d-v7-141-speed-cpu-wave-behavior/102/4 "2021-09-27T20:32:20Z")

</div>

I tested remotely a i9 machine, run more than 30 minutes (actually hours), I still observed 100% CPU usage. Did your tested machines include i9, or are they all AMD machines?

---

<div class="post-metadata">

**Author:** ![ccancino](https://avatars.discourse-cdn.com/v4/letter/c/bcef8e/32.png) [@ccancino](https://forum.itascainternational.com/u/ccancino)\
**Post date:** [September 27, 2021, 8:51pm UTC](https://forum.itascainternational.com/t/flac3d-v7-141-speed-cpu-wave-behavior/102/5 "2021-09-27T20:51:51Z")

</div>

![image](https://canada1.discourse-cdn.com/flex031/uploads/itasca/original/1X/88d16bb2dca2d1e2f173b9f269f896e266749f9f.png)

I tested this model in two i9 machines, same behavior.

---

<div class="post-metadata">

**Author:** ![dblanksma](https://avatars.discourse-cdn.com/v4/letter/d/858c86/32.png) [@dblanksma](https://forum.itascainternational.com/u/dblanksma)\
**Post date:** [September 30, 2021, 9:50pm UTC](https://forum.itascainternational.com/t/flac3d-v7-141-speed-cpu-wave-behavior/102/6 "2021-09-30T21:50:44Z")

</div>

Hey @ccancino I have a i9 machine an AMD Threadripper (older CPU). Could you send me the data file you are using and I’ll take a look.

---

<div class="post-metadata">

**Author:** ![ccancino](https://avatars.discourse-cdn.com/v4/letter/c/bcef8e/32.png) [@ccancino](https://forum.itascainternational.com/u/ccancino)\
**Post date:** [September 30, 2021, 10:11pm UTC](https://forum.itascainternational.com/t/flac3d-v7-141-speed-cpu-wave-behavior/102/7 "2021-09-30T22:11:18Z")

</div>

I sent you an email. Thanks, Derrick.

---

<div class="post-metadata">

**Author:** ![cheng](https://avatars.discourse-cdn.com/v4/letter/c/ecccb3/32.png) [@cheng](https://forum.itascainternational.com/u/cheng)\
**Post date:** [January 31, 2022, 8:12pm UTC](https://forum.itascainternational.com/t/flac3d-v7-141-speed-cpu-wave-behavior/102/8 "2022-01-31T20:12:14Z")

</div>

This turned out to be the unoptimized interface initialization every 10 steps triggered by the zone relax excavation command during cycling. If the model does not have a lot of interfaces, the issue will not be there. It has been greatly improved/fixed from v700.146.

---

<div class="post-metadata">

**Author:** ![ccancino](https://avatars.discourse-cdn.com/v4/letter/c/bcef8e/32.png) [@ccancino](https://forum.itascainternational.com/u/ccancino)\
**Post date:** [January 31, 2022, 8:27pm UTC](https://forum.itascainternational.com/t/flac3d-v7-141-speed-cpu-wave-behavior/102/9 "2022-01-31T20:27:22Z")

</div>

@cheng @dblanksma thanks for persisting with this issue. Much appreciated!
