Jump to content

So how does GPU work. I heard something like that: GPU read CPU instructions and if it says shows a tree it shows a tree.
Is it correct? 

Laptop: Acer V3-772G  CPU: i5 4200M GPU: GT 750M SSD: Crucial MX100 256GB
DesktopCPU: R7 1700x GPU: RTX 2080 SSDSamsung 860 Evo 1TB 

Link to comment
https://linustechtips.com/topic/264854-how-gpu-works/
Share on other sites

Link to post
Share on other sites

Its much more complicated than that :)

 

by much I mean much much much much

. I'm preparing presentation for English class, so I need only basic idea, so every one could undrestand. 

Laptop: Acer V3-772G  CPU: i5 4200M GPU: GT 750M SSD: Crucial MX100 256GB
DesktopCPU: R7 1700x GPU: RTX 2080 SSDSamsung 860 Evo 1TB 

Link to comment
https://linustechtips.com/topic/264854-how-gpu-works/#findComment-3601379
Share on other sites

Link to post
Share on other sites

Here´s the link, if you´re not shy to read you´ll know ;) . That should answer your question in detail.

 

http://en.wikipedia.org/wiki/Graphics_processing_unit

 

Intel i7 7820X (delidded) @ 4.9GHz - MSI X299 M7 ACK + EKWB Fullcover Block - G.Skill Trident Z 32GB @ 3466MHz - nVidia Titan Xp + EKWB Fullcover Block @ 2.1GHz - Samsung 960Pro 2x - WDD Blue 2TB - Seasonic 750W Platinum - modded Corsair 600C - Hardtubed Custom Watercooling

Link to comment
https://linustechtips.com/topic/264854-how-gpu-works/#findComment-3601380
Share on other sites

Link to post
Share on other sites

Well it's the device that renders all of the images on your screen. Boiled down to it's most simple points, it does draw that tree when it's told to. But HOW it draws the tree is very complicated.

Case: Meatbag, humanoid - APU: Human Brain version 1.53 (stock clock) - Storage: 100TB SND (Squishy Neuron Drive) - PSU: a combined 500W of Mitochondrial cells - Optical Drives: 2 Oculi, with corrective lenses.

Link to comment
https://linustechtips.com/topic/264854-how-gpu-works/#findComment-3601381
Share on other sites

Link to post
Share on other sites

Case: Meatbag, humanoid - APU: Human Brain version 1.53 (stock clock) - Storage: 100TB SND (Squishy Neuron Drive) - PSU: a combined 500W of Mitochondrial cells - Optical Drives: 2 Oculi, with corrective lenses.

Link to comment
https://linustechtips.com/topic/264854-how-gpu-works/#findComment-3601384
Share on other sites

Link to post
Share on other sites

27xQWXR.png

Location: Kaunas, Lithuania, Europe, Earth, Solar System, Local Interstellar Cloud, Local Bubble, Gould Belt, Orion Arm, Milky Way, Milky Way subgroup, Local Group, Virgo Supercluster, Laniakea, Pisces–Cetus Supercluster Complex, Observable universe, Universe.

Spoiler

12700, B660M Mortar DDR4, 32GB 3200C16 Viper Steel, 2TB SN570, EVGA Supernova G6 850W, be quiet! 500FX, EVGA 3070Ti FTW3 Ultra.

 

Link to comment
https://linustechtips.com/topic/264854-how-gpu-works/#findComment-3601405
Share on other sites

Link to post
Share on other sites

I don't think this is something that can be explain in a forum post.

There are many different levels on which its operation can be explained.

The most basic would be that a gpu processing binary data across multiple

cores to rasterize an images (really difficult to do). They are preferable for

3d work as these are numerical intensive and require paralleled processing

to solve efficiently. (reason this is done on gpu is because it can excute

~10^3 32bit instructions per cycle, cpu ~10 per cycle, not that a cpu is worse

but it is not as optimize for this kind of tasks.)

Link to comment
https://linustechtips.com/topic/264854-how-gpu-works/#findComment-3601418
Share on other sites

Link to post
Share on other sites

I don't think this is something that can be explain in a forum post.

There are many different levels on which its operation can be explained.

The most basic would be that a gpu processing binary data across multiple

cores to rasterize an images (really difficult to do). They are preferable for

3d work as these are numerical intensive and require paralleled processing

to solve efficiently. (reason this is done on gpu is because it can excute

~10^3 32bit instructions per cycle, cpu ~10 per cycle, not that a cpu is worse

but it is not as optimize for this kind of tasks.)

Remember, OP said it has to be easily explainable to 'newbs' for his class. (They'd have to understand some of it at least) :)

Plus what you said, is kinda said in more laments terms in the Wiki in previous posts.

Not that I disregard your post, I think as accurate as it is, but it would confuse most people with 'overly complicated' wisdom.

Maximums - Asus Z97-K /w i5 4690 Bclk @106.9Mhz * x39 = 4.17Ghz, 8GB of 2600Mhz DDR3,.. Gigabyte GTX970 G1-Gaming @ 1550Mhz

 

Link to comment
https://linustechtips.com/topic/264854-how-gpu-works/#findComment-3601443
Share on other sites

Link to post
Share on other sites

Remember, OP said it has to be easily explainable to 'newbs' for his class. (They'd have to understand some of it at least) :)

Plus what you said, is kinda said in more laments terms in the Wiki in previous posts.

Not that I disregard your post, I think as accurate as it is, but it would confuse most people with 'overly complicated' wisdom.

Truth be told, I have wildly over simplified things. There loads of in-between processes skipped and things reduced,

there is no easy way to explain how gpus work. It is much easier to just explain what they do.

Link to comment
https://linustechtips.com/topic/264854-how-gpu-works/#findComment-3601935
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×