|  When AMD delayed its 64bit processor, code named Clawhammer, it was thought that AMD would have nothing to compete with Intel and its mainstream processors until September. Barton would have been overlooked had it not been for this delay, but since then the market has changed. Barton has now become AMD’s effort to keep Intel at bay, until it has other products to come into the mainstream. Intel has of late taken an aggressive approach, bringing in hyperthreading and transitioning to newer chipsets and standards faster than originally planned.
AMD’s Barton CPU does not differentiate too much from prior Athlon processors; the only real addition is 256KB added on to its on die L2 cache. The Barton processor comes with 512 KB of 16-way set associative L2-cache, as well as 128 KB L1 cache. Essentially Barton’s only advantage over other Athlon XPs is 256KB of cache. It is possible for the older Athlon XPs to outperform the Barton as well, as the Barton Athlon XP 3000+ is clocked at 2,170 MHz, while the Athlon XP 2800+ is clocked at 2,250 MHz. Cache can actually help a lot in many applications. The principle behind cache is that a processor’s cache will store the most common information, and when an application needs that information it will first check the L1 cache, and then L2 cache, and then L3 cache (if applicable), and then go to Ram. Ram is fast, but no where near as fast as cache. Barton’s transistor count is significantly higher than prior Athlon XPs due to its on die cache upgrade. Both the Pentium4 and Barton XP come in at about 55 Million transistors. Like other newer Athlons, Barton contains a 333MHz front side bus, which will give it some increase in performance against older Athlons. Factures: - Nine-issue superpipelined, superscalar x86 processor microarchitecture designed for high performance
- Multiple parallel x86 instruction decoders
- Three out-of-order, superscalar, fully pipelined floating point execution units, which execute x87 (floating point), MMX™ and 3DNow!™ instructions
- Three out-of-order, superscalar, pipelined integer units
- Three out-of-order, superscalar, pipelined address calculation units
- 72-entry instruction control unit
- Advanced hardware data prefetch
- Exclusive and speculative Translation Look-aside Buffers
- Advanced dynamic branch prediction
- 21 original 3DNow!™ instructions—the first technology enabling superscalar SIMD
- 19 additional instructions to enable improved integer math calculations for speech or video encoding and improved data movement for Internet plug-ins and other streaming applications
- 5 DSP instructions to improve soft modem, soft ADSL, Dolby Digital surround sound, and MP3 applications
- 52 SSE instructions with SIMD integer and floating point additions offer excellent compatibility with Intel’s SSE technology
- Compatible with Windows® XP, Windows 2000, Windows ME, and Windows 98 operating systems
The Athlon XP 3000+ has power requirements and thermal output, very similar to that of an Athlon XP 2700+. The more astute among you will also notice that the 3000+ also has the same clock speed as the 2700+, 2.16GHz (13x166MHz). AMD had to change their naming convention scale to reflect the performance increases the larger cache brings to the Athlon XP. Something very interesting to note is that even with approximately 16.7 million more transistors, the Athlon XP 3000+ operates at the same core voltage as the 2700+, yet its typical Thermal Power output and typical current draw is lower. It seems AMD has further refined their .13 micron copper manufacturing process, which should allow the "Barton" to operate at lower temperatures in most circumstances. |
