How Do Vector Graphics Work?

Vector graphics allow you to rescale images to any size without losing quality! How does that work?

Thanks to Pearl Auto for supporting Tech Quickie. Go to to get FREE two-day shipping!


Join the community:


25 thoughts on “How Do Vector Graphics Work?

  • 2:45 – This is an incorrect statement. The reason vectorized graphics aren't used is precisely because such graphics systems have far worse compression, absolutely horrible compression.

    Besides, JPG is a sort of vectorized algorithm, it's moving data into the frequency domain from the spatial domain. As you zoom in however, what you see becomes increasingly meaningless as high frequencies (i.e. detail) has been discarded to make the image a reasonable size.

    This guy isn't as educated as he presents himself.

  • But why aren't (at least) desktop operating systems laden with vector graphics? Like all the toolbars, icons, all visual gui elements, etc.- why aren't they vector at this point?

  • vector images are made of paths or formulas but these must also contain pixels thats why they are visible on screen right ??
    any one explain please

  • Looks like the link to the back=up camera doesn't work so you likely are not getting your referral and they are probably getting free advertising on your dime & time.

  • While all of this was technically correct.. it was terribly explained!

    I've been using vector graphic programs for over 20 years now. NeoPaint on the AtariST was my first bitmap creator. 1 floppy disk per image! Even if the space in the image was blank, every square white pixel had a value. So it was 700KB for a tiny image.

    Later, when I got my first PC, in the 90s I bought Corel Suite, which included Corel Draw – a whole pack of CDs with vector clipart.

    This is totally off-topic but..
    One thing I realised (and subsequrntly got me thinking about the way that AI works, for some reason)
    Is that, I intuitively knew that there was no zoom or resize limit to vector images as soon as I started working with them because I realised that there was no 'absolute' information stored within the image relating to point positions, lines or curves. They're all described using mathematics, rather than a pixel grid with 'absolute' values.

    I could see what the limitations were too.. Things like complicated colour fades – gradients that weren't perfect, or had anything other than point to point colours. An edge to edge colour change, with each edge having a defined colour is something that a computer could calculate and fill in. Anything else would be far more complicated. Textures as well – you'd be going back to pixels again. It's best to think about it like a computerised version of layers of cut-out cardboard with a few additional interesting features like gradients and perfect mathematical curves..

    Anyway – the leap from seeing the difference between bitmap and vector to lossless quality resizing was something that just occured to me straight away.. without even knowing that vectors were even saved in an original co-ordinate maths system instead of a bitmap export.

    It got me thinking.. I don't know how human brains make certain leaps of logic. Sometimes its a realisation, other times you have to work on a problem for an answer to come.

    How would AI make these leaps? Are computers destined to always follow exact orders.
    Another thought.. will there be 'literal' disasters when operators make mistakes due to wordplay or a slip of the tongue etc..

    Do AI neural networks need a creativity node.. Imagine all of the knowledge and understanding that computers have for AI problem solving, being linked together by a 'creative' network, that takes whats known and combines it with other knowns and using a check and improve feedback algorithm it could check its own creativity, see what works, what doesn't… all in order to 'think', to create, to imagine..

    Until we get anywhere close to that, we will never have computers that are inspired, creative or 'think'..
    They only ever do what they are told, follow the instructions, learn the way it has been told to learn..
    Yeah, just because it's complicated and computer scientists are baffled more and more by their own creations and their complexity – yes, it worked, but the unexpected way in which it found the answer confounds the creators.. It's still not creating anything truly new.

    I am totally impressed and amazed by the state of AI even now. But at this pace, we're 50 years off anything close to brain-like.. 100 from anything human like.

  • Speaking of enhance you should cover the ways Ai is now enhancing images taking low res raster images and adding details that we're not even captured accurately orย editing motion in video to take a bad dancer or even just random movements of one person and mapping them to mimic the movements of a professional dancers video.

  • $500 FOR A CAMERA THAT GOES BEHIND MY CAR, BULL****. I mean like seriously, I see how good the pitch idea is, but the marketing team is messing around when they price the item at 500

Comments are closed.