Amd delivers there performence in their chips by amount of work done clock per clock and improving the operating frequency at the same time. This produces a high volume of work done per cycle and high operating frequencies.
QuantiSpeed architecture which amd is based around produces a nine-issue, superscalar, fully-pipelined core. This provides more pathways to feed application instructions into the execution engines of the core, simply allowing the processor to complete more work in a given clock cycle. This architecture also has even more goodies such as superscalar, fully-pipelined floating point engine, hardware data prefetch, and exclusive, speculative Translation Look-aside Buffers (TLBs). Combined, these features help boost overall productivity and allow a system to boot and load applications quickly.
The AMD XP processor with performance-enhancing cache memory features 64K instruction and 64K data cache for a total of 128K L1 cache. 512K of integrated, on-chip L2 cache for a total of 640K full-speed, on-chip cache.
Socket A infrastructure designs are based on high-performance platforms and are supported by a full line of optimized infrastructure solutions (chipsets, motherboards, BIOS).
The die size is 54.3 million transistors on 101mm2(approx)
Which allows extreme performance, and allows them to keep up with intel 100% of the way