IEEE 754: Introduction to Floating Point Format

This is the first of my tutorial mini series on the IEEE 754 floating point standard. That's the way a modern computer represents real numbers. There's a lot of tutorials on this topic already but I've not yet been able to find any that describe the steps involved in converting a real number into its IEEE 754 bit pattern and back again, they all seem to gloss over the operation as if it does not need any further explanation. I'm making this tutorial series to describe exactly what the standard means for programmers and how to convert to and from these bit patterns.

[@hellbash]

Hey Chris! I'm Martin, I just wanted to take this oportunity to thank you for your generosity. To upload this video and making it free. I'm trying to become a programmer, a little bit late, but it is also said that is never too late I'm 29 years old. Anyways I just wanted to thank you again for this tutorial and all the effort. I'll try my best to keep up with your tutorials. Cheers, Martin.

[@-MahmoudElshenawy]

SO did you become a programmer?

[@vanta6lack]

I'll be 32 in a month, and I decided to become a programmer. I'm also wondering if you really did become a programmer. My ex-boyfriend had a low-paid job two years ago, spend one year learning JAVA, and then got a job in QA on his very first try (seriously, he only went to a single job interview!!!)

[@sigmawolf228]

This is the best video I've ever seen about floating-point number on youtube! I wanna thank you man!

[@pixelbogpixxelbog2090]

Wow all the comments are like 7/8 years old. I wonder where all these people are in life now and what they have done in the last years. And btw great video, still better quality than most videos these days :)

[@ddoodm]

This is incredibly helpful for my HSC Software Design exam in a few weeks! Thanks a billion again!!

[@danielsaavedragonzalez5677]

Thank you very very much for this video! This is exaclty what I am learning in my 'Numerical Methods for Engineers'!

[@nanocalp]

So well explained, thank you.

[@Dream-xr7ls]

your video are really helpful !!! thank you for your willness of helping others learn !

[@BobTheBob647]

Thank you!!! You da real MVP.

[@sandreusz]

That really helped me out! Thanks alot!

[@habaguy]

Nicely explained Sir.

[@michelarjuna]

Thanks a lot! Your video made the concept clear to me :) I Have my like.

[@leonardmohr9450]

What a legend. Thanks a bunch :)

[@SonLyAu]

Thank you!

[@aarb17]

Thank you

[@andyhitchings8826]

ty!

[@WhatsACreel]

It stores integers exactly.

[@FannyVanderbildt]

thanks :)))

[@WhatsACreel]

Ha, ha, I do things like that all the time, this is confusing stuff, thanks for watching a commenting.

[@soleuso755]

i need help, how do i Convert (263.3)10 to IEEE754 floating point format in hexadecimal format

[@cavercody]

Good video! At 12:30 you say 80-bit extended precision floats are not in C++. This is not true, they are generally exposed as the "long double" type.

[@WhatsACreel]

Yes, I was wrong :) I think it's mostly just the Microsoft C++ that doesn't support them. Pity, really, they are very cool!

[@ArkDEngal]

doubles do not have 64bit mantissa, they are 64bits so mantissa must be 64 - (sign=1) - (exp=11) = 52. I realize this video is old, but it is still around!

[@WhatsACreel]

Did I say that? Hahaha, how silly! Just a slip of the tongue. Well spied :)

[@matthewe3813]

@@WhatsACreel what a legend, responding to a comment half a decade after the video was made!

[@eboubaker3722]

what website ?

[@steelddd]

Don't see the pdfs on your website...

[@clankill3r]

steelddd He went to the beach and never came back :(

[@theexplosionist2019]

It's so stupid that games still use float instead of double.

[@wm4201]

Why is that stupid? The vast majority of the time a float is fine. A double has twice the memory footprint making it less cache friendly

[@ahmadalastal5303]

I worked on a 3d software called SimLab Composer, it uses double in its core and I learned that this is the biggest mistake that was made since the very early version of the software, for small scenes for example the size of the saved file is huge, while using floats the size will be 50% less and nothing is corrupted, only use doubles when you have to otherwise use float

[@nathanflett6427]

Who cares about C++???

[@pauliusdamanskas71]

????????

[@nathanflett6427]

@@pauliusdamanskas71 C or some other language. C++ is a waste of time.

[@pauliusdamanskas71]

@@nathanflett6427 you're dumb.