Unverified50% confidenceSolutionExact time
Due to the lack of an FPU, J2ME game developers at the time commonly used fixed-point or pure integer mathematics to implement geometric operations, typically scaling values up by 1024 or 4096 times and storing them as integers.
1
Sources
50%
Confidence
Long-term
Relevance
10/8/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn FP16 format each parameter occupies 2 bytes, while in INT4 format it occupies only 0.5 bytes, theoretically achieving a 4x compression ratio.61% similarUnverifiedFP32 uses 32 bits to represent floating-point numbers, consisting of 1 sign bit, 8 exponent bits, and 23 mantissa bits.60% similarUnverifiedCompute-bound means the hardware's compute units are fully saturated, at which point increasing batch size or optimizing numerical precision (such as using BF16 instead of FP32) becomes meaningful.60% similarUnverifiedDouble quantization further quantizes the FP32 scaling factors shared among every 64 parameters into FP8 format, reducing the quantization overhead per parameter from 0.5 bits to approximately 0.127 bits.59% similarUnverified使用FWHT处理1024维向量仅需约10240次运算，而直接矩阵乘法需约100万次，接近100倍加速59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/990001API
curl https://kongchang.com/api/v1/knowledge/claims/990001MCP
get_claim(id=990001)