• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

I want to manufacture graphics cards, is nanoelectronics the best thing to study?

Fake

  • Flat

    Votes: 0 0.0%
  • Implants

    Votes: 0 0.0%

  • Total voters
    0

Gizmo j

Platinum Member
And are "microprocessors" relevant? That just means cpu right?

And are graphics cards considered a "integrated circuit"?
 
This is a type of question we have already answered before. The answer is
"If you want to do A, you need to know B, by which time you would no longer ask the question".

Keep studying CS, you'll have your answer way before you are qualified.
 
To be clear, does he want just to make video cards? Or does he want to also design and manufacture GPUs?
 
I would start with looking into what it takes to make a video card, using an existing GPU chip. Only thing, I don't really know where you would buy a GPU chip on it's own, I don't think that's something you can just buy on sites like Digikey or Mouser.

If you want to make an actual GPU, it might be worth looking into FPGA programming. You can do all sorts of stuff with that. A FPGA based GPU would not compete with existing discrete GPUs by any means but could make a fun project either way.

For actual sillicon production I would start with something smaller, like maybe microcontrollers. Or heck even more discrete components like 555 timers and even just transistors. Getting the sillicon fab process right is not easy. With simpler devices running at a lower clock speed you can get away with a larger die process which will be easier to learn.
 
I imagine someone super good with AI could technically do it. Program an AI robot that can 3D print parts for itself, and evolve from there.

This is also how we end up with human killing terminator style machines because once the AI finds out it no longer needs us it will just see us as a threat and eliminate it right away.
 
I imagine someone super good with AI could technically do it. Program an AI robot that can 3D print parts for itself, and evolve from there.

This is also how we end up with human killing terminator style machines because once the AI finds out it no longer needs us it will just see us as a threat and eliminate it right away.
I would to write the AI to manufacture items intended only for Human use. Design it so that its sole purpose is to serve People.
 
I'm well aware that manufacters buy GPUs from Nvidia and AMD.
OP specifically mentioned manufacture "graphic cards", not GPU (graphic processing units). 🙄🙄🙄
606_gtx_680_callouts-11395914.jpg
 
If you want to design computer components - Electrical Engineering.

If you want to manufacture computer components - Mechanical or Industrial Engineering. And learn Mandarin Chinese.
 
Out of curiosity, what is the minimum amount of time it would take someone to learn how to design a basic GPU? Said person is going to school full time.
 
Out of curiosity, what is the minimum amount of time it would take someone to learn how to design a basic GPU? Said person is going to school full time.
You would need to pick up from the top. No such thing as a 'basic' GPU, you'd need to study until AMD can get you to work at the design table.
 
Well a GPU that works.
Are we talking designing gpus for street cred/lulz, or for a living??

I suppose, it wouldnt be completely impossible (for someone who has the tools and is a total nerd) to build a working Voodoo2 or something.
 
Come to think of it, I would start with a CPU. Right now Intel and AMD are basically the main contenders and BOTH have built in backdoors. There is probably a market for an open source CPU architecture where everything is open, and has no royalties to create. Design and build a basic one that can perform decently, try to get the Linux kernel devs involved, and maybe a 3rd party open source community to do the motherboard and chipset and you could potentially create a fully open source computer platform. This would NOT be easy, but if you're serious, it's something to consider.

Before even getting into rolling sillicon I would probably start with a discrete component based computer. Heck, start with a calculator, then work your way up. Don't use microcontrollers, that's cheating. 😛
 
Come to think of it, I would start with a CPU. Right now Intel and AMD are basically the main contenders and BOTH have built in backdoors. There is probably a market for an open source CPU architecture where everything is open, and has no royalties to create. Design and build a basic one that can perform decently, try to get the Linux kernel devs involved, and maybe a 3rd party open source community to do the motherboard and chipset and you could potentially create a fully open source computer platform. This would NOT be easy, but if you're serious, it's something to consider.

Before even getting into rolling sillicon I would probably start with a discrete component based computer. Heck, start with a calculator, then work your way up. Don't use microcontrollers, that's cheating. 😛

https://en.m.wikipedia.org/wiki/OpenSPARC
 

No idea where you would even buy that. Lot of these products never really make it on sites like Tigerdirect etc. There might be a handful of organizations that are part of the project and have them but it's not like the average person can just go out and buy the hardware.

Then again anyone who wants to take on a project like this needs to have a plan to actually make it available to market. That's probably half the battle right there.
 
No idea where you would even buy that. Lot of these products never really make it on sites like Tigerdirect etc. There might be a handful of organizations that are part of the project and have them but it's not like the average person can just go out and buy the hardware.

Then again anyone who wants to take on a project like this needs to have a plan to actually make it available to market. That's probably half the battle right there.


There is RISC-V as well.
https://riscv.org
 
No idea where you would even buy that. Lot of these products never really make it on sites like Tigerdirect etc. There might be a handful of organizations that are part of the project and have them but it's not like the average person can just go out and buy the hardware.

Then again anyone who wants to take on a project like this needs to have a plan to actually make it available to market. That's probably half the battle right there.
I would start with the RISC-V ISA if I wanted to design a CPU. For thing there is already working hardware I can actually buy.
 
Can someone tell me if a gpu/graphics card is a integrated circuit or not?

And is nanoelectronics the best thing to study?
 
Back
Top