# Same code much faster on a Ryzen than on a Xeon?

**URL:** <https://discourse.julialang.org/t/same-code-much-faster-on-a-ryzen-than-on-a-xeon/60534>\
**Category:** Performance\
**Tags:** question, first-steps, performance\
**Created:** [May 4, 2021, 5:09pm UTC](https://discourse.julialang.org/t/same-code-much-faster-on-a-ryzen-than-on-a-xeon/60534 "2021-05-04T17:09:45Z")\
**Posts on this page:** 1\
**Showing post:** 11

<div class="post-metadata">

**Author:** ![ederag](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/ederag/32/4106_2.png) [@ederag](https://discourse.julialang.org/u/ederag)\
**Post date:** [May 16, 2021, 10:41pm UTC](https://discourse.julialang.org/t/same-code-much-faster-on-a-ryzen-than-on-a-xeon/60534/11 "2021-05-16T22:41:04Z")

</div>

Timings (lower better) plotted below, with the same scales.

Machine A ( `AMD Ryzen 9 3900X`):

 ![amber](https://global.discourse-cdn.com/julialang/original/3X/5/7/57909f649c24f448f00b6450c2c2703f127010d7.png)

Machine B (`bi-Xeon Gold 6146`):

 ![basalt](https://global.discourse-cdn.com/julialang/original/3X/f/0/f0797415aa5048789ff82dba932451de6fb89164.png)

Bottom axis is the number of points, top axis is the memory used by the 6 arrays of N points.  
The blue vertical lines indicate the L1, L2, L3 cache sizes.

Bottom line: the Ryzen has much faster L3 caches than the xeon.

---

_[View the full topic](https://discourse.julialang.org/t/same-code-much-faster-on-a-ryzen-than-on-a-xeon/60534)._