@Sakura please summarize this article, thanks uwu.
TLDR:
Memory access time is more accurately modeled as O(N^(1/3)), meaning that as memory size increases, access time increases at a slower rate than previously thought. 
Key Points:
- Memory Access Complexity: Memory access time is O(N^(1/3)), not O(1) as traditionally assumed.
- Theoretical Basis: The delay in memory access is proportional to distance, which affects how quickly data can be read or written.
- Practical Implications: Optimizing algorithms, especially in cryptography, requires considering this new model for memory access.
- Example in Cryptography: Precomputing values can lead to significant performance improvements when memory access is treated as O(N^(1/3)).
- Future of Computing: As we approach limits of general-purpose CPUs, understanding memory access nuances becomes crucial for designing efficient systems.
In-depth summary:
In the article, the author argues that the traditional view of memory access time as O(1) is misleading. Instead, they propose that memory access should be modeled as O(N^(1/3)). This means that if the amount of memory increases by a factor of eight, the time taken to access it will double. The reasoning behind this is based on the physical limitations of how quickly a processor can communicate with memory, which is influenced by the speed of light and the distance between them.
The author provides a theoretical framework for this argument, explaining that in a three-dimensional space, doubling the distance allows for eight times more memory to be accessed, which directly impacts access times. They also touch on practical aspects, noting that different types of memory (like registers, cache, and RAM) behave differently, and that naive assumptions about access times can lead to inefficiencies in algorithm design.
A concrete example is given in the context of cryptography, where the author discusses how precomputing values can optimize performance. By treating memory access as O(N^(1/3)), one can find an optimal size for precomputed tables that balances speed and memory usage, leading to faster computations in cryptographic algorithms. The article concludes by highlighting the importance of understanding these nuances as we explore new computing architectures, especially with the rise of specialized hardware like ASICs and GPUs.
ELI5:
When you want to get something from a big box of toys, it usually takes a certain amount of time to find it. If the box gets bigger, it might take longer to find your toy, but not as much as you might think! Instead of taking a lot longer, it only takes a little longer. This is what the article is saying about computers and how they access memory. If we understand this better, we can make computers work faster, especially for tricky tasks like solving puzzles in cryptography! 
Writers main point:
The primary point the author is making is that memory access time should be viewed as O(N^(1/3)), which has significant implications for algorithm optimization and the future of computing architectures. 