<?xml version="1.0" encoding="UTF-8"?>
<feed xmlns="http://www.w3.org/2005/Atom" xml:lang="en-gb">
	<link rel="self" type="application/atom+xml" href="https://pybullet.org/Bullet/phpBB3/app.php/feed/topic/13288" />

	<title>Real-Time Physics Simulation Forum</title>
	
	<link href="https://pybullet.org/Bullet/phpBB3/index.php" />
	<updated>2021-10-23T17:22:08+00:00</updated>

	<author><name><![CDATA[Real-Time Physics Simulation Forum]]></name></author>
	<id>https://pybullet.org/Bullet/phpBB3/app.php/feed/topic/13288</id>

		<entry>
		<author><name><![CDATA[KKlouzal]]></name></author>
		<updated>2021-10-23T17:22:08+00:00</updated>

		<published>2021-10-23T17:22:08+00:00</published>
		<id>https://pybullet.org/Bullet/phpBB3/viewtopic.php?p=43658#p43658</id>
		<link href="https://pybullet.org/Bullet/phpBB3/viewtopic.php?p=43658#p43658"/>
		<title type="html"><![CDATA[Re: 30x perfomance difference between PyBullet and Bullet3]]></title>

		
		<content type="html" xml:base="https://pybullet.org/Bullet/phpBB3/viewtopic.php?p=43658#p43658"><![CDATA[
While I wont say it isn't possible to speed up computation of PyBullet, I will say it should be expected that it underperforms compared to Bullet3. Python is an abstraction layer atop C/C++ so there will be overhead.<p>Statistics: Posted by <a href="https://pybullet.org/Bullet/phpBB3/memberlist.php?mode=viewprofile&amp;u=10757">KKlouzal</a> — Sat Oct 23, 2021 5:22 pm</p><hr />
]]></content>
	</entry>
		<entry>
		<author><name><![CDATA[Mark97]]></name></author>
		<updated>2021-10-02T10:10:15+00:00</updated>

		<published>2021-10-02T10:10:15+00:00</published>
		<id>https://pybullet.org/Bullet/phpBB3/viewtopic.php?p=43631#p43631</id>
		<link href="https://pybullet.org/Bullet/phpBB3/viewtopic.php?p=43631#p43631"/>
		<title type="html"><![CDATA[30x perfomance difference between PyBullet and Bullet3]]></title>

		
		<content type="html" xml:base="https://pybullet.org/Bullet/phpBB3/viewtopic.php?p=43631#p43631"><![CDATA[
I've just ran the benchmark of 3000 cubes in Bullet3 (using Visual Studio), and found a total step time of about 7 ms. (4 threads)<br>I've copied the C++ code and translated it into Python code, and ran 3000 standard cubes in PyBullet (using Anaconda Navigator, Notebook), and found a total step time of about 200 ms. (single thread probably)<br><br>That is a performance difference of about 30x. That is a big difference. Therefore I would like to check if others have found the same performance difference, or if PyBullet can be sped up in any way.<p>Statistics: Posted by <a href="https://pybullet.org/Bullet/phpBB3/memberlist.php?mode=viewprofile&amp;u=14099">Mark97</a> — Sat Oct 02, 2021 10:10 am</p><hr />
]]></content>
	</entry>
	</feed>
