What's new
Carbonite

South Africa's Top Online Tech Classifieds!
Register a free account today to become a member! (No Under 18's)
Home of C.U.D.

Graphics rendering question...sort of?

BARGUILD

Orange Expert
VIP Supporter
Rating - 100%
76   0   0
Joined
Apr 6, 2016
Messages
2,577
Reaction score
294
Points
4,535
Location
Randburg, Johannesburg
Alright, not even sure the title is correct but I am going to ask away.

When a computer renders a game, and say it renders an arm and you see the edge of the arm, it is rendered or created in polygons and then turned into blocks/pixels. Correct me if this logic is incorrect please.

When the computer "comes up" with said image, does it "come up" with the image broken up into polygons/blocks/pixels? Or is the image "an arm" which is then cut into polygons/blocks/pixels?

Finally, was planning on tweeting this at Lex Fridman to see if he would help me out but I'll ask here anyway.

A- Do we know how the brains processing of images works?
B- If so, and assuming it is different to that of a computer, what is the fundamental difference and why can a computer not be built to draw imagery this way?
C- THE ACTUAL QUESTION (lolsadness.jpeg). Is the answer to question B like trying to answer the question of "How does the entire brain work?" and thus essentially unanswerable right now? Essentially being able to understand the entire brain in order to build a computer that could simply draw imagery in the same way.
 
I would love to think about this and respond, but my mental energy is too low at the moment. Watching the thread though and will try and catch up tomorrow.

But keep in mind that our brains essentially receive 'pre-rendered' content and then construct imagery from that. Very little of what we dream or imagine is "made up" by our brains, but constructs of what we have seen, a mixture of neural pathways. Also, I do not believe that we know for sure how our brains 'render' images, on a technical level, as even some who have been born blind have explained images that they have 'seen'.

I'll have to research this, very interesting topic, thanks for bringing it up.
 
I don't believe our brains render anything. The eyes work like a sort of window and our brains look for similarities in what is seen in the window. Super short and sweet but this is how I understand it.

We aren't processing images like a computer does, a computer builds the images out of code where half the image can be missing and then shows as a crappy jagged solid colour sort of thing.
Our eyes will pick up light wavelengths they can, I think 3, and then the brain takes each "Image" and just looks for similarities.

My take, prob wrong but I think that's how it works. Either way I'm interested in this thread.
 
I would love to think about this and respond, but my mental energy is too low at the moment. Watching the thread though and will try and catch up tomorrow.

But keep in mind that our brains essentially receive 'pre-rendered' content and then construct imagery from that. Very little of what we dream or imagine is "made up" by our brains, but constructs of what we have seen, a mixture of neural pathways. Also, I do not believe that we know for sure how our brains 'render' images, on a technical level, as even some who have been born blind have explained images that they have 'seen'.

I'll have to research this, very interesting topic, thanks for bringing it up.
Based on your second paragraph, I am now thinking that I may have misworded part of my questioning. And now I am not even sure how to ask what I am wondering. When the computer receives the pre-rendered image, it then cuts it up for you to "see" the image. Similarly, the information received by the eyes needs to be cut up/built up/understood? by the brain for you to perceive it. Can you replicate this process on silicon or something else, now? If not, is the reason because the brain is not understood enough? Is the answer to that question really the answer to understanding the brain or can it be cut up like that and different processes replicated?

Lol, I should not have put this into words at the end of a work day ffs. I will sit and think about it and put something properly in writing tomorrow that makes more sense I think.
 
I don't believe our brains render anything. The eyes work like a sort of window and our brains look for similarities in what is seen in the window. Super short and sweet but this is how I understand it.

We aren't processing images like a computer does, a computer builds the images out of code where half the image can be missing and then shows as a crappy jagged solid colour sort of thing.
Our eyes will pick up light wavelengths they can, I think 3, and then the brain takes each "Image" and just looks for similarities.

My take, prob wrong but I think that's how it works. Either way I'm interested in this thread.
I think I remember reading a paper somewhere suggesting that certain neurons in the brain 'process' different parts of images at different rates. Doesn't sound too much like a 'window'. Let me see if I can find that paper and properly read it again and link it here.
 
I don't believe our brains render anything. The eyes work like a sort of window and our brains look for similarities in what is seen in the window. Super short and sweet but this is how I understand it.

We aren't processing images like a computer does, a computer builds the images out of code where half the image can be missing and then shows as a crappy jagged solid colour sort of thing.
Our eyes will pick up light wavelengths they can, I think 3, and then the brain takes each "Image" and just looks for similarities.

My take, prob wrong but I think that's how it works. Either way I'm interested in this thread.
Just to follow up on this, I think I may be wrong.


That suggests that processing occurs almost as a single unit, but it does show that processing of the image occurs. So maybe not on a single neuron level, but processing does seem to be occurring. Definitely based on prior information and context, so I guess this means my question would probably require the idea of intelligence, and the brain, and maybe even consciousness, to be solved first.


That is somewhat tangential, it deals specifically with motion and not 'building' the image, but is also probably integrated as a single unit considering the previous article.

There was another article I definitely read somewhere that had circuit diagrams of individual processing of individual neurons and it had to do with sight. Maybe I am misremembering, can't seem to find it now.
 
So maybe the ultimate answer to my question is...









...yes, you probably need to solve the brain, and intelligence, in order to replicate just the visual processing of humans.
 
Me: How many drugs did you use today?
BARGUILD: Yes.

What we see is partially what our eyes perceive, partially our brain filling in the blanks (eg, close one eye - you don't see your blind spot where the optic nerve leaves the eye as your brain fills in that missing information. Your brain does such a good job of this that you can't see your blind spot even if you try (unlike your nose) without an image such as a black spot on a white piece of paper which disappears when hitting the blind spot (which your eye will then fill in as white - so you still don't really see the blind spot as such).
 

Users who are viewing this thread

Latest posts

Back
Top Bottom