Originally posted by: screech
fx51 = 2.2, 53 = 2.4, 55 = 2.6, 57 = 2.8
4000+ = 2.4
i didn't know they even made a fx51 for socket 939. fx53 came in both 939 and 940 i think, but i only remember a socket 940 (slower registered memory, basically) fx51. also i think the fx51 is about 600 (it was on zipzoomfly anyway) dollars while the 4000+ is about $483ish, and more overclockable (albeit no unlocket multipliers).
multiplier (multi) x FSB = CPU clock speed. example, the fx51 is 11 x 200 = 2.2 gigaherts. for the 4000+ it is 12 x 200 = 2.4.
so put simply: even without OCing the 4000+ is better. if you want to OC it would probbaly go farther then the fx51 as well, due to the san diego core.....
Actually, the FX-51 was Socket 940 ONLY (130nm process) and need Registered Memory. The FX-53 (130nm process) Clawhammer core was the first Socket 939 (can use Non-Registered memory) FX-chip, followed about three months later by the FX-55 (also 130nm on Clawhammer core) but using Strained Silicon got a spped bump to 2.6GHz. Several months ago a 90nm process FX-55 was released with the San Diego core (still at 2.6GHz default speed). The newest FX chip (the FX-57) is built on the 90nm process, uses DSL (Dual Stress Liner) Strained Silicon, runs 2.8GHz default speed, and has the San Diego core).Originally posted by: GuitarDaddy
The FX51 comes in both s939 and s940. The main difference with any FX processor is it has fully unlocked multis (up and down), while A64's have top locked multis. With A64's you can only lower the CPU multi from the factory preset default.
For instance I beleive both come factory set at 200mhz x 12 = 2400mhz, with the 4000+ you can use any multi 12x or below, with the FX51 you can also use 13x,14x,15x, ect..
The only advantage an FX has is the ability to overclock without raising the HTT, leaving your ram at stock speeds without using a memory divider. But the FX51 is the much older Clawhammer core, where the 4000+ has the newest flagship core the San Diego