contractless/blockchain
2
1
Fork 0
Code Issues 1 Pull Requests Actions Packages Projects Releases Wiki Activity

Update Proof of Time

contractless 2024-10-23 20:28:22 +00:00
parent 02e4cc0148
commit 349eaae725
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
Proof-of-Time.md

@ -27,9 +27,9 @@ RandomX is a PoW concept that was created by Monero. It is specifically designed
RandomX is a memory-intensive and compute-intensive PoW algorithm that leverages features of general-purpose CPUs to perform mining tasks efficiently. The core idea behind RandomX is to design an algorithm that can fully utilize the strengths of modern CPUs, such as branch prediction, out-of-order execution, and memory access patterns. This it difficult to optimize these processes on GPUs and ASICs. It achieves this by relying on random code execution and memory-hard operations, which require large amounts of RAM and frequent memory access. This makes it prohibitively expensive and inefficient for specialized hardware to outperform CPUs in the mining process. RandomX is a memory-intensive and compute-intensive PoW algorithm that leverages features of general-purpose CPUs to perform mining tasks efficiently. The core idea behind RandomX is to design an algorithm that can fully utilize the strengths of modern CPUs, such as branch prediction, out-of-order execution, and memory access patterns. This it difficult to optimize these processes on GPUs and ASICs. It achieves this by relying on random code execution and memory-hard operations, which require large amounts of RAM and frequent memory access. This makes it prohibitively expensive and inefficient for specialized hardware to outperform CPUs in the mining process.
The key difference between Monero's RandomX and traditional PoW algorithms, like Bitcoin’s SHA-256 or Litecoin’s Scrypt, lies in the way these algorithms interact with hardware. In traditional PoW, the computational workload is focused on repetitive hashing operations, which can be parallelized easily, giving GPUs and ASICs a significant edge. The key difference between Monero's RandomX and traditional PoW algorithms, like Bitcoin's SHA-256 or Litecoin's Scrypt, lies in the way these algorithms interact with hardware. In traditional PoW, the computational workload is focused on repetitive hashing operations, which can be parallelized easily, giving GPUs and ASICs a significant edge.
In contrast, RandomX is deliberately designed to take advantage of the sequential processing capabilities of CPUs. The algorithm dynamically generates random sequences of instructions and requires large amounts of memory, limiting the degree of parallelism that GPUs can achieve. GPUs, which excel at performing many simple tasks in parallel, struggle with RandomX’s sequential and memory-intensive workload. In contrast, RandomX is deliberately designed to take advantage of the sequential processing capabilities of CPUs. The algorithm dynamically generates random sequences of instructions and requires large amounts of memory, limiting the degree of parallelism that GPUs can achieve. GPUs, which excel at performing many simple tasks in parallel, struggle with RandomX's sequential and memory-intensive workload.
RandomX has an inverse effect compared to traditional PoW algorithms. While traditional PoW leads to mining centralization around GPUs and ASIC hardware, RandomX shifts the focus to CPUs and memory, making CPU mining more efficient. However, like other PoW algorithms, it still favors one type of hardware—CPUs—over GPUs and ASICs. To date, no PoW algorithm has been able to balance mining equally across all hardware types, as each algorithm inevitably favors either CPUs or specialized hardware like GPUs and ASICs. RandomX has an inverse effect compared to traditional PoW algorithms. While traditional PoW leads to mining centralization around GPUs and ASIC hardware, RandomX shifts the focus to CPUs and memory, making CPU mining more efficient. However, like other PoW algorithms, it still favors one type of hardware—CPUs—over GPUs and ASICs. To date, no PoW algorithm has been able to balance mining equally across all hardware types, as each algorithm inevitably favors either CPUs or specialized hardware like GPUs and ASICs.