PowerPC Twice As Fast AS Pentium? BYTEmark vs SPEC

| Macintosh vs Windows Index |

Apple uses BYTEmark to bolster it's claim that the new G3 machines are twice as fast as the fastest Pentium based machines. SPEC benchmarks do not show such a wide margin of better performance by the G3 chip. Below we post a comparison between the two benchmark programs. We invite you to tell us what program you think is the more legitimate and why. We will post thoughtful responses on this page.


BYTEmark

The BYTEmark benchmark test suite is used to determine how the processor, its caches and coprocessors influence overall system performance. Its measurements can give us an idea of how well a given system will perform. It is the in-house brogram of Byte magazine and can be freely downloaded from web site.

 The BYTEmark test suite is especially valuable since it lets onedirectly compare computers with different processors and operating systems. The code used in BYTEmark tests simulates some of the real-world operations used by popular office and technical applications.

There are 10 tests in the BYTEmark test suite. Each uses a well-known algorithm to analyze the full spectrum of processor performance in the same way real-world applications do. Some tests concentrate on integer performance; others test floating point capabilities. BYTEmark results are influenced by the type of compiler used

Comparative Benchmark Results

Processor Scores

G3/400 - 148% better performance


G3/300- 143% better performance

Floating Point Scores

G3/400 - 44% better performance

G3/300- 43% better performance

Benchmark Description

Numeric sort Generic integer performance. Should exercise non-sequential performance of cache (or memory if cache is less than 8K). Moves 32-bit longs at a time, so 16-bit processors will be at a disadvantage.

String sort Tests memory-move performance. Should exercise non-sequent ial performance of cache, with added burden that moves are byte-wide and can occur on odd address boundaries. May tax the performance of cell-based processors that must perform additional shift operations to deal with bytes.

Bitfield Exercises "bit twiddling" performance. Travels through memory in a somewhat sequential fashion; different from sorts in that data is merely altered in place. If properly compiled, takes into account 64-bit processors, which should see a boost.

Emulated F.P. Past experience has shown this test to be a good measurement of overall performance.

Fourier Good measure of transcendental and trigonometric performance of FPU. Little array activity, so this test should not be dependent of cache or memory architecture.

Assignment The test moves through large integer arrays in both row-wise and column-wise fashion. Cache/memory with good sequential performance should see a boost (memory is altered in place -- no moving as in a sort operation). Processing is done in 32-bit chunks -- no advantage given to 64-bit processors.

Huffman A combination of byte operations, bit twiddling, and overall integer manipulation. Should be a good general measurement.

IDEA Moves through data sequentally in 16-bit chunks. Should provide a good indication of raw speed.

Neural Net Small-array floating-point test heavily dependent on the exponential function; less dependent on overall FPU performance. Small arrays, so cache/memory architecture should not come into play.

LU decomp. A floating-point test that moves through arrays in both row-wise and column-wise fashion. Exercises only fundamental math operations (+, -, *, /).


SPEC

SPEC, the Standard Performance Evaluation Corporation, is a non-profit corporation formed to "establish, maintain and endorse a standardized set of relevant benchmarks that can be applied to the newest generation of high-performance computers"

SSPEC95 is composed of two suites of benchmarks:

SPEC CINT95: a set of eight compute-intensive integer/non-floating point benchmarks SPEC CFP95: a set of 10 compute-intensive floating point benchmarks

These are intended to provide a measure of compute-intensive performance of the processor, memory hierarchy and compiler components (the 'C' in CINT95 and CFP95) of a computer system. SPEC95 results are influenced by the type of compiler used



Comparative Benchmark Results

Processor Scores


G3/400 - 19% better performance


G3/300- 22% better performance

Floating Point Scores

G3/400 - 2% worse performance

G3/300- 17% better performance



CINT95 Benchmarks

Benchmark Application Area Specific Task
099.go Game playing; artificial intelligence Plays the game Go against itself.
124.m88ksim Simulation Simulates the Motorola 88100 processor running Dhrystone and a memory test program.
126.gcc Programming & compilation Compiles pre-processed source into optimized SPARC assembly code.
129.compress Compression Compresses large text files (about 16MB) using adaptive Limpel-Ziv coding.
130.li Language interpreter Lisp interpreter.
132.ijpeg Imaging Performs jpeg image compression with various parameters.
134.perl Shell interpreter Performs text and numeric manipulations (anagrams/prime number factoring).
147.vortex Database Builds and manipulates three interrelated databases.

CFP95 Benchmarks

Benchmark Application Area Specific Task
101.tomcatv Fluid Dynamics / Geometric Translation Generation of a two-dimensional boundary-fitted coordinate system around general geometric domains.
102.swim Weather Prediction Solves shallow water equations using finite difference approximations. (The only single precision benchmark in CFP95.)
103.su2cor Quantum Physics Masses of elementary particles are computed in the Quark-Gluon theory.
104.hydro2d Astrophysics Hydrodynamical Navier Stokes equations are used to compute galactic jets.
107.mgrid Electromagnetism Calculation of a 3D potential field.
110.applu Fluid Dynamics/Math Solves matrix system with pivoting.
125.turb3d Simulation Simulates turbulence in a cubic area.
141.apsi Weather Predication Calculates statistics on temperature and pollutants in a grid.
145.fpppp Chemistry Performs multi-electron derivatives.
146.wave Electromagnetics Solve's Maxwell's equations on a cartesian mesh.

Which benchmark program do you trust more and why? - Let us know

BYTEmark vs SPEC - Readers bite back and let us know what benchmark program they think is more effective

For Free Macintosh Hardware - Click Here

Internal Links

  • PowerPC Chip performance compared
  • Check out the Links Page for links to sites with more Processor information
  • More Windows vs Mac Comparisons
  • External Links

  • PC Magazine takes on the iMac's performance and BYTEmark claims
  • SPEC vs BYTEmark - which is the better, more accurate benchmark program
  • PentiumII vs the G3 / G4 - MaKiDo compares these two chips
  • Motorola - detailed information about the various PowerPC chips
  • BYTE Magazine review of G3 and Pentium II
  • BYTE Magazine review of the G3/266
  • BYTEmark faq
  • PowerPC - source data for the benchmarks above
  • Intel - source data for the benchmarks above
  • SPEC 95 FAQ - all your questions about SPEC 95 answered
  • BYTEmark documentation
  •