How do Graphics Cards Work? Part 3 #gpu #videogame #technology #engineering #gpucomputing

A GPU (Graphics Processing Unit) works in tandem with its GDDR (Graphics Double Data Rate) memory to process and render graphics at high speeds. The GPU consists of thousands of cores optimized for parallel processing, enabling it to handle multiple tasks simultaneously, such as rendering pixels or running AI computations. GDDR memory, located on the GPU card, is specifically designed for high bandwidth and low latency, allowing it to transfer large amounts of data quickly to and from the GPU. The GPU fetches data, such as textures, vertex information, or computational tasks, from GDDR memory, processes it, and outputs the results to the display or main memory. This seamless interaction, facilitated by high-speed memory buses, ensures that graphics and computational tasks are processed efficiently and in real time.

[Aliforoutan-cqtf]

Wow i loved the graphics and explanation

[manjosh]

I can't believe I am watching this instead of studying. Its just too good to watch.

[icegiant]

You forgot that before the final pixels get to your screen, they must be intertwingled, which is a complex yet important process.

[demure.SHTUT]

TLDR:

GPU: High volume simple problems
CPU: Low volume complex problems.

[Aoedz_XV]

Ohh, thats why modern game needs a really big vram,

[Alexifeu]

This video is so cool. Hell Yeah. I love the detailed explanation and the analogies. This is some quality stuff. I learn more than in school in a minute.

[techrazor]

Branch education is the best learning channel.❤

[edwelndiobel]

I loved the class on SSD now i understand completely how they work. Super simple!

[HollywoodGaming-fyvt]

Glad to see these guys getting the recognition they deserve. These videos are impeccably reseached and put together.

[Smash_ter]

Going to put this here, typically a GPU bus width per module depends on the type of module and how it's placed on the card. The card shown here is based on the 3090, and at the time of the 3090's release Micron was only able to make 1 GB modules of GDDR6X, hence why it had 24 VRAM modules, and on the 3090 the VRAM chips were put on the same spot on both sides of the PCB so it acts like 1 VRAM module.

[Thomas-svii]

The visuals of this video made the content even more interesting 👏 ✨️ cool video!

[justafreakable]

My god this is such a good explanation

[mscking]

This is best explaining I ever hear and see about cpu to know about what cpu do

[the_devilsadvocate]

I love when Byakuya Kuchiki explains gaming things to me.

[muhazreen]

Finally, channel dedicated explaining microscopic lightning witchcarft❤❤❤❤

[gregmarsters]

As a one time graphics programmer I can tell you the ever increasing volume of data bandwidth is more about low efficent algorithms than anything else.

[marlinhicks]

Branch education shows up in my feed. It’s a good day

[KobaMy]

This education was more than nowadays in classrooms …

[somerandomjuice]

This reminds me of a quality cable tv service 😭 i love it

[alsarudin]

That’s why the Ryden x3d series of chips help it load faster because it basically stores the data making it more easy for the guy to get