Python sort can be 80x FASTER!

Python itself is pretty slow as I demonstrate here with a sort of a large data set, but has a pretty neat feature that allows you to use precompiled code to get a ridiculous speed boost!

##### Related videos:

#python

(Take 15% off any premium NativeScript course by using the coupon code YT2020)

— — — — — — — — —

❤️ SUBSCRIBE TO MY YOUTUBE CHANNEL 📺

— — — — — — — — —

🏫 FREE COURSES

— — — — — — — — —

📱LET'S CONNECT ON SOCIAL MEDIA

Raffle!
Here are the rules - Give the video a thumbs up, make sure you're subscribed and leave a comment down below. I'll pick a winner at random two weeks after the video is published. I'll contact you by replying to your comment. So make sure you have some sort of notification turned on. You'll have to respond within 48 hours or else I'll have to pick someone new.

[AZisk]

Hey all! This video is based on your suggestions. Keep 'em coming and I'll make more 😀. Also, what do you think of my Python song?

[KrzysztofOsiecki]

Inspired by these tests I did similar one, testing all kinds of numpy.sort on 16GB macbook air M1 and my 8th gen i7 Dell with 16GB. 10^9 array elements made i7 with 16GB of RAM go boom - process was silently killed, but macbook air kept going. Seems like memory management is also slightly better on M1.

[gdthegreat]

wow man, your videos are really great, waiting on your view as "DEV" on new m1x chip, after monday.
Also, these sorting algos always scare me, but I am improving on it.

[frz_akbar]

JavaScript + Web assembly, that will be cool

[matts]

0:33 Is there some advantage to "import numpy as numpy" instead of simply "import numpy"?
Anyway, great video. I'd love to see how long the python only Tim sort takes on an M1!

[plasmatch]

Python has built-in sorting functions that are perfectly fine to use. [1, 2, 3].sort() and sorted([1, 2, 3]). They might be a little slower than numpy.

[caciocosta]

The test should be on same dataset to make it more accurate.

[erproerpro]

Alex, have you tried running SQL Server 2019 in Parallels Windows ARM?

[MauricioSierra]

Will you compare m1 vs MX and AMD in the next videos?

[woolfel]

LOL, there are too many people who use python and don't know how it works under the cover. Lots of people have been trying to improve python over the last 10 years. So far none of them have reached 100% compatibility, which means you really can't drop in a different version and magically get 10x speed up. Python is a nice bootstrap for calling C or C++ code, but it's a terrible language for high performance operations.

[ryzenforce]

You are referring a lot of sorting algos that I never heard about. In my programming days, Quick Sort algo was the fastest and according to 2021 sorting algos ranks, it is still the fastest so why don't you use that instead? Quick Sort also only use O(log(n)) for memory usage (that could benefit the M1) whereas a lot of the other sorting algos uses O(n) for memory usage.

[metallicafan]

Hey I just saw your GoLang livestream, was wondering if you could do a performance comparison for GoLang

macOS ARM vs macOS Intel x86_64
Windows x86_64
Linux x86_64 (and ARM if it's possible)

Also, I'm thinking of learning GoLang. According yo you is it worth the time?

[knofi]

Shutup and take Julia instead! And you'll be happy!:)