# Link-Time Optimisations resulting in much slower code with MSVC 2017

**URL:** <https://forum.juce.com/t/link-time-optimisations-resulting-in-much-slower-code-with-msvc-2017/31657>\
**Category:** Windows\
**Created:** [February 7, 2019, 5:56pm UTC](https://forum.juce.com/t/link-time-optimisations-resulting-in-much-slower-code-with-msvc-2017/31657 "2019-02-07T17:56:05Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![WilliamkWusik](https://sea2.discourse-cdn.com/flex026/user_avatar/forum.juce.com/williamkwusik/32/3821_2.png) [@WilliamkWusik](https://forum.juce.com/u/WilliamkWusik)\
**Post date:** [February 7, 2019, 5:56pm UTC](https://forum.juce.com/t/link-time-optimisations-resulting-in-much-slower-code-with-msvc-2017/31657/1 "2019-02-07T17:56:05Z")

</div>

This is weird. Today I decided to run some tests and with Link-Time Optimisations enabled I’m getting a much higher CPU usage on a test patch I created for Wusik Station (128 voices playing all at once). Not to mention that it takes forever to compile… anyone had the same problems? On OSX it also takes too long to compile, but I didn’t check CPU usage as I compile on a virtual-machine box. 😉

Cheers, WilliamK

---

<div class="post-metadata">

**Author:** ![DaveH](https://sea2.discourse-cdn.com/flex026/user_avatar/forum.juce.com/daveh/32/4924_2.png) [@DaveH](https://forum.juce.com/u/DaveH)\
**Post date:** [February 7, 2019, 7:07pm UTC](https://forum.juce.com/t/link-time-optimisations-resulting-in-much-slower-code-with-msvc-2017/31657/2 "2019-02-07T19:07:31Z")

</div>

This is running Release I presume? Much slower than what exactly?

---

<div class="post-metadata">

**Author:** ![WilliamkWusik](https://sea2.discourse-cdn.com/flex026/user_avatar/forum.juce.com/williamkwusik/32/3821_2.png) [@WilliamkWusik](https://forum.juce.com/u/WilliamkWusik)\
**Post date:** [February 8, 2019, 3:21pm UTC](https://forum.juce.com/t/link-time-optimisations-resulting-in-much-slower-code-with-msvc-2017/31657/3 "2019-02-08T15:21:56Z")

</div>

I have to run some tests so I have a more scientific result. Will also ask some users to test if they can. 🙂
