Next-Gen Intel Chip Brings Big Gains For Floating-Point Apps 176
An anonymous reader writes "Tom's Hardware has published a lengthy article and a set of benchmarks on the new "Haswell" CPUs from Intel. It's just a performance preview, but it isn't just more of the same. While it's got the expected 10-15% faster for the same clock speed for integer applications, floating point applications are almost twice as a fast which might be important for digital imaging applications and scientific computing."
The serious performance increase has a few caveats: you have to use either AVX2 or FMA3, and then only in code that takes advantage of vectorization. Floating point operations using AVX or plain old SSE3 see more modest increases in performance (in line with integer performance increases).
Re:Would that improve hashing speeds in, say, Bitc (Score:5, Informative)
Re:Let's see... (Score:5, Informative)
It's a joke. The Intel P5 Pentium FPU had a bug where
4195835/3145727=1.333739068902037589 The correct answer is 1.333820449136241002.
Less rounding of floating point numbers (Score:5, Informative)
While it's got the expected 10-15% faster for the same clock speed for integer applications, floating point applications are almost twice as a fast HTH
Integer and floating point are separately implemented in the hardware, so an improvement to one often doesn't apply to the other. You can add integers by counting on your fingers. To do that with floating point, you have to cut your fingers into fractions of fingers - a very different process.
See: http://en.wikipedia.org/wiki/FMA3 [wikipedia.org]
It's common to have an accumulator like this:
X = X + (Y * Z)
To compute that in floating points, the processor normally does:
A= ROUND(Y*Z) X=ROUND(X+A)
Each ROUND() is necessary because the processor only has 64 bits in which to store the endless digits after the decimal point. FMA can fuse the multiply and the add, getting rid of one rounding step, and the intermediate variable:
X= ROUND( X + (Y*Z) )
That makes it faster. Since integers don't get rounded to the available precision, the optimization doesn't apply to integers. The above processor would do Y*Z, then +X, then round, then X=. A CPU designer can make that faster by including either a "add and multiply" circuit or a "add and round" circuit or a "round and assign' circuit. Any set of operations can be done in two clock cycles, if the maker decides to include a hardware circuit for it.
FMA4 (Score:4, Informative)
Pah. AMD had FMA4 since 2011
Re:Hope it's going in the new Mac Pro (Score:5, Informative)
The Core i7's are consumer-grade processors and are slower than the Xeon's the Mac Pros use
This is completely incorrect. The current Mac Pros use Nehalem based Xeons which are two generations back from the current Ivy Bridge i7s. Xeons may have differences in core count, cache and/or ECC support but their execution units are the same as their desktop equivalents. The base Mac Pro CPU is equivalent to an i7-960 with ECC support. The current Ivy Bridge i7s are a fair bit faster.
Re:Hope it's going in the new Mac Pro (Score:5, Informative)
ECC memory is only marginally slower. Considering error rates and modern memory sizes, it is far past time that it became a standard feature. The extra cost would be totally insignificant if were standard, and not used as an excuse to gouge people on Xeons.