Enhancing graphics, understanding the basics.

Computer Graphics is a complicated subject. That requires a deeper study into geometry, algorithms, and performance. The difficulty seems to be further increased by the state of the technology, and the advancement of it. As GPUs (Graphic Processing Units) gain speed in advancement the CPU is left behind. Thus, despite this difficulty, we are still relying on the GPU as a fine tuned tool for solving difficult math problems within milliseconds.

Despite the challenge, my brief understanding is that as a computer ages, new GPUs or dedicated cards are created to withstand the demands of heavier loads and more complicated calculations.

Let us consider Occam's Razor, the simplest answer tending to be the right one, for the sake of making it easier for me to understand and hopefully explain my results or brief conclusions to others. Since the design of such a processing unit is based on future reducing costs, to hopefully create and continue a trickle-down effect that reaches the general mass. We must start with high performance devices and break the limits of what the device is capable of. I am not sure at what level the costs are consider for the consumer, though it seems from a design standpoint that cost breakdown is based on expected consumer demand as well as meeting baseline performance standards. Where does the cost come from? Who decides the worth/value of a GPU? Is the cost solely based on the production of parts? Does the cost weigh more heavily on the materials used to build the GPU?

Also, let us consider, the causal gamer, I do not want to get involved in the differences between casual gamer and hardcore gamer, as where that line is draw is questionable, and even so now that the general population of gamers playing different games is increasing. Here is fact not stating any correlation or even causation that may affect the increase in the diversity of gamers who potentially will use a dedicated card, such as a GPU. Though provides to some insight into a growing community. The Entertainment Software Association states,

"Esports has the potential to disrupt traditional sports viewership. According to technology strategist Michael Wolf, by 2020 esports will make up roughly 10 percent of US sports viewing, reaching 88 million fans in the US and 495 million fans world-wide, and exceeding baseball, basketball, or hockey. "

From: https://www.theesa.com/, https://email.theesa.com/t/ViewEmail/y/0E6E19169D787869/29830D811AABECCD05263A35EB2CBB57

eSports is growing commodity in itself, as such game tournaments like the Heroes of the Dorm are being broadcasted on ESPN2. I would expect that those that want to get into the game at a more competitive level will want to get involve. To be honest, watching this community grow just like any other is exciting, and promotes teamwork, dedication, and willingness to spend hours in front of a screen (which in all or most cases is relatively unhealthy, a sacrifice dedicated gamers take care to make). 

Back to cost though, which goes with saying the cost of high performance gaming computers is let us say ambiguously high with the GPU being largely part of it. Ambiguously, as that the consumer can build a gaming computer at relative low cost, and get there "bang for their buck" with some time and effort. However, there is noticeable limit, where standards are much higher, and you can not simply use a day-to-day laptop or regular task-oriented PC to compete at such a level. Though, once this bar is meant, the belief is that relatively long term costs will be better, as the user will not have to submit to a large price tag for a pre-built machine every few years.

Seemingly, at buyer level things get less complicated, because as your purchasing parts for computer, the thoughts come about, is the GPU running the correct amount teraflops? Or is the memory matching correctly with the CPU? These questions seem to diminish as manufacturers seem to make design decisions simple with advertising. Also, the supporters or critics of a particular GPU, helps eliminate the confusion, between what is good or bad in understanding a GPU's capabilities. The internet is flourishing with youtubers that give excellent explanations of the product (a GPU), and at a more personable level. I believe this does more with how much time and effort that one would care to make learning about what they are purchasing.

So, final thought, GPUs are excellent devices that require exceptional engineers and architects to fit performance needs of a growing community. The best way to simplify the understanding of these GPUs is to rate and compare them by performance, the numbers seem to be clear from there. This is much more simpler then working with one particular GPU, and studying its capabilities based on mathematical concepts.