Radiance Benchmark Test


BENCHMARK RETIRED Due to changes made to the Radiance HEAD release dated 2005-05-25 (and all versions thereafter) and its different handling of the ambient cache for scenes with significant specular components, the bench2 benchmark scene must be retired. A new benchmark scene and a new table of results is currently being created and will appear here shortly. All results tabulated below are still internally self-consistent (see note 3). The work scaling between this scene and bench4 is roughly 1:23.

Go to the new benchmark page.

This section contains instructions for running a sample Radiance test scene and the results from many different computers. The test is not intended to represent all that Radiance can do, nor is it neccessarily representative of the way Radiance is commonly used. It is simply a good scene with reasonable complexity, some procedural texture use, and some interreflection calculation.

The scene within, and the files obtainable below, are the original work of Paul Bourke, who maintains a Benchmarks for Radiance page in Australia with different scenes. Please take a moment to visit it if you can.

Instructions

  1. You must download and build Radiance, or have the Radiance distribution on the machine you wish to benchmark. Download the Radiance distribution from the "download Radiance" page of the Radiance Online site.
  2. Download bench2.tar.gz and use gunzip and tar -xf to create the bench2 directory.
  3. In the doit file are listed several commands. You will need to run each of the commands listed in the order they are listed. Use whatever timing device you have available to determine the CPU time required (not neccessarily the real, elapsed time). On an SGI and other Unixes this can be accomplished by prefacing each command with the time command, like:
    time oconv scene.rad > scene.oct
    Of primary importance is the rpict command, but for the sake of possible interest, try to record times for each of the three commands.
  4. Using whatever means neccessary, compile the information for the other fields listed in the table below. A complete entry is desired, but not required. You can get the version number from the command rpict -version, OS name from uname -a, compiler version usually from [compiler] --version, processor info from cat /proc/cpuinfo, and RAM from running top and reading the amount available.
  5. E-mail the information to me, mstockATumich.edu, and I will add it to the table below.
  6. Choose the guilty pleasure of your choice and indulge.

Results

Note 1: this information is not meant to serve as an exact speed comparison among machines, as unmentioned hardware configurations can have an impact on the speeds reported.

Note 2: rows ordered by rpict time

Note 3: The upgrade to Radiance 3.5 included changes to oconv's default behavior, which explains the differences between oconv timings; but more importantly, Radiance 3.6 includes changes to rpict's default parameter settings. The link to the scene data is for the new version of the benchmark tests. It uses consistent parameters and should be used for new benchmark tests

Computer Processor Speed RAM OS Radiance
Version
oconv rpict pfilt notes Date Submitter
PC AMD Phenom II X6 1090T 3612 MHz 16 GB Linux 2.6.34.8-68.fc13.x86_64 4.1a
2011-03-24
4.89 80.90 0.11 62 2011-03-25 mstock
PC AMD Opteron 246 1990 MHz 1024 MB FreeBSD 5.3b7-amd64 3.6a
2004-10-06
8.66 164.34 0.14 59 2004-10-07 defuller
PC AMD Opteron 246 1990 MHz 2048 MB Linux 2.4.22-1.2179.nptlsmp 3.6a
2004-05-01
11.64 184.64 0.22 51 2004-05-01 mstock
PC AMD Opteron 246 1990 MHz 2048 MB Linux 2.4.22-1.2179.nptlsmp 3.6a
2004-05-01
11.89 192.76 0.22 50 2004-05-01 mstock
PC AMD Opteron 246 1990 MHz 2048 MB Linux 2.4.22-1.2179.nptlsmp 3.6a
2004-05-01
12.14 207.34 0.24 56 2004-06-01 mstock
PC AMD AthlonXP Barton 2500+ 1980 MHz 1 GB
PC3200
Linux 2.6.10-nitro4 3.6.1 17.8 222.07 0.452 61 2005-02-04 rgs25
PC Intel P4
512kB L2
2993 MHz 512 MB Linux 2.4.26 3.6.1 14.54 222.13 0.29 57 2004-12-02 francesco.anselmo
Apple Mac G5 Dual IBM/Apple G5 2 GHz 1 GB
PC3200
OS X 10.3.3 3.6a
2004-05-21
14.77 251.65 0.50 54 2004-05-21 Giulio
PC AMD AthlonXP Barton 2500+ 1830 MHz 512 MB
PC2700
Linux 2.6.9-1.667 3.6.1 16.897 256.578 0.318 58 2005-01-04 mstock
PC AMD AthlonXP Barton 2500+ 1830 MHz 512 MB
PC2700
Linux 2.4.20-20_29 3.6a
2003-10-24
18.13 258.53 0.35 42 2003-10-29 mstock
Dell Dimension 8250 Intel P4
512kB L2
2533 MHz 512 MB
PC 1066 RDRAM
Linux 2.4.18 3.4.1 12.8 269.52 0.33 33 2002-12-13 chris
Apple Mac G5 IBM/Apple G5 2 GHz 4 GB OS X 10.3.1 3.5 15.26 270.80 0.25 43 2003-11-20 rs.westling
Dell Dimension 8250 Intel P4
512kB L2
2533 MHz 512 MB
PC 1066 RDRAM
Linux 2.4.18 3.4.1 13.18 276.6 0.33 32 2002-12-13 chris
Dell Dimension 8250 Intel P4
512kB L2
2533 MHz 512 MB
PC 1066 RDRAM
Linux 2.4.18 3.4.1 12.8 284.0 0.44 31 2002-12-13 chris
PC AMD AthlonXP 2000+ 1683 MHz 1024 MB Linux 2.4.18-27.8.0 3.5 19.625 286.541 0.398 37 2003-05-15 mstock
Dell Dimension 8250 Intel P4
512kB L2
2533 MHz 512 MB
PC 1066 RDRAM
Linux 2.4.18 3.4.1 12.84 288.0 0.44 30 2002-12-13 chris
Dell Dimension 8250 Intel P4
512kB L2
2533 MHz 512 MB
PC 1066 RDRAM
Linux 2.4.20 3.5 28.33 291.41 0.86 38 2003-05-20 chris
PC AMD AthlonXP 2000+ 1683 MHz 1024 MB Linux 2.4.18-27.8.0 3.5 20.006 294.104 0.406 36 2003-05-15 mstock
Dell Dimension 8250 Intel P4
512kB L2
2533 MHz 512 MB
PC 1066 RDRAM
Linux 2.4.18 3.4.1 12.8 295.2 0.44 29 2002-12-13 chris
Apple PowerMac G5 1.8 GHz 512 MB OS X 10.3.1 3.6a 19.14 297.2 0.45 44 2003-12-10 gward
Dell Dimension 8250 Intel P4
512kB L2
2533 MHz 512 MB
PC 1066 RDRAM
Linux 2.4.18 3.4.1 13.16 302.4 0.32 28 2002-12-13 chris
PC AMD AthlonXP Barton 2500+ 1830 MHz 512 MB
PC2700
Linux 2.4.20-20_29 3.6a
2003-10-24
20.42 310.00 0.37 41 2003-10-29 mstock
PC Intel P4
512kB
2.80 GHz 1 GB Linux 2.4.20-8
RH 9.0
3.4.1 pmap 22.53 317.59 0.76 53 2004-05-14 becmkwok
PC AMD
Opteron 142
1600 MHz 3648 MB
PC2100
Linux 2.4.20-8
32-bit mode
3.5 21.70 319.26 0.32 47 2004-01-02 yoichi
PC AMD Opteron 246 1990 MHz 2048 MB Linux 2.4.22-1.2179.nptlsmp 3.6a
2004-05-01
18.35 323.21 0.39 55 2004-06-01 mstock
PC Intel P4
512kB
2.80 GHz 1 GB Linux 2.4.20-8
RH 9.0
3.5 17.53 333.96 0.29 52 2004-05-14 becmkwok
HP tc4100 Intel P3
512kB L2
1400 MHz 1 GB Linux 2.4.18 3.4.1 17.55 333.98 0.5 22 2002-08-23 chris
PC Dual AMD Athlon 1900+ 1600 MHz 1536 MB Linux 2.4.9-31smp 3.4.1 15.61 341.07 0.38 16 2002-05-17 mstock
PC AMD AthlonXP 2000+ 1683 MHz 1024 MB Linux 2.4.18-27.8.0 3.5 22.379 348.305 0.445 35 2003-05-15 mstock
PC
EPOX 8K5A2+
AMD Athlon XP1700+ 1560 MHz 256 MB
PC2100 DDR
Linux 2.4.18-18 3.4.1 18.489 354.373 0.41 26 2002-11-24 darrell
Dell Dimension 8250 Intel P4
512kB L2
2533 MHz 512 MB
PC 1066 RDRAM
Linux 2.4.18 3.4.1 14.55 360.0 0.32 27 2002-12-13 chris
PC AMD AthlonXP 2000+ 1683 MHz 1024 MB Linux 2.4.18-27.8.0 3.5 22.381 369.328 0.418 34 2003-05-15 mstock
PC Dual AMD Athlon 1900+ 1600 MHz 512 MB Linux 2.4.7-10smp 3.4 13.48 383.9 1.8 8 2002-05-13 pillo.pillo
PC AMD Athlon 2000+ 16?? MHz 512 MB Linux 2.4.18 3.4 18.6 384.0 0.5 - 2002-08-06 pillo.pillo
PC Dual AMD Athlon 1900+ 1600 MHz 1536 MB Linux 2.4.9-31smp 3.4 17.42 385.80 0.40 14 2002-05-17 mstock
PC Dual AMD Athlon 1900+ 1600 MHz 1536 MB Linux 2.4.9-31smp 3.4.1 17.21 387.66 0.43 15 2002-05-17 mstock
PC AMD Athlon 1800+ 1539 MHz 256 MB Linux 2.4.18-3 3.4 18.86 405.26 0.44 15 2002-07-14 davep
HP/Compaq AlphaServer ES80 Dual EV7 800 MHz 2 GB Linux 2.4.7 3.4.1 18.87 417.6 0.41 18 2002-06-19 chris
HP/Compaq AlphaServer ES45 Quad EV68c 1000 MHz 4 GB HP Tru64 UNIX 3.4.1 12.9 424.8 0.3 18 2002-05-29 chris
PC AMD Athlon Thunderbird 1400 MHz 512 MB Linux 2.2.19 3.1.20 24.59 445.32 0.64 - 2001-10-10 ohkubo
Dell i8200 Intel P4 1600 MHz 512 MB Linux 2.4.22-1.2115
Fedora Core 1
3.6.1 46.89 452.42 1.34 49 2005-01-27 mstock
HP/Compaq AlphaServer ES45 Quad EV68c 1000 MHz 4 GB HP Tru64 UNIX 3.4.1 15.4 460.8 0.4 19 2002-05-29 chris
Dell i8200 Intel P4 1600 MHz 512 MB Linux 2.6.2-1.156 3.6a
2004-02-09
47.35 463.46 1.33 49 2004-02-09 mstock
Dell i8200 Intel P4 1600 MHz 512 MB Linux 2.4.22-1.2115
Fedora Core 1
3.6a
2004-02-09
46.89 465.78 1.37 49 2004-02-09 mstock
Apple Powerbook G4 1.25 GHz 512 MB OS X 10.3.1 3.5 27 472 0.73 46 2003-12-23 Giulio
PC
HP Vectra
Intel P3 1000 MHz 384 MB Linux 2.4.18 3.4.1 25.66 478.8 0.7 9 2002-05-16 chris
PC AMD Athlon Thunderbird 1400 MHz 512 MB Windows 2000 Desktop Rad 1.0 29.21 479.20 0.53 - 2001-10-10 ohkubo
Dell i8200 Intel P4 1600 MHz 512 MB Linux 2.4.22-1.2115
Fedora Core 1
3.5 47.09 485.67 1.35 49 2004-02-09 mstock
PC
HP Vectra
Intel P3 1000 MHz 256 MB Linux 2.4.20 3.5 39.01 496.04 0.92 39 2003-05-15 chris
Macintosh PMac 1.42 GHz . OS X 10.2.6 3.5 26.1 512.01 0.56 40 2003-06-04 gward
PC
HP Vectra
Intel P3 1000 MHz 384 MB Linux 2.4.9 3.4.1 27.6 518.4 0.74 4 2002-04-15 chris
PC
HP Vectra
Intel P3 1000 MHz 384 MB Linux 2.4.9 3.4.1 27.46 525.6 0.76 5 2002-04-15 chris
Dell i8200 Intel P4 1600 MHz 512 MB Linux 2.4.18-10 3.4.1 19.87 557.79 0.59 16 2002-09-22 mstock
Dell i8200 Intel P4 1600 MHz 512 MB Linux 2.4.22-1.2115
Fedora Core 1
3.6a
2004-02-09
51.24 563.08 1.52 48 2004-02-09 mstock
Apple Powerbook G4 1.25 GHz 512 MB OS X 10.3.1 3.5 30 569 0.81 45 2003-12-23 Giulio
Dell i8200 Intel P4 1600 MHz 512 MB Linux 2.4.22-1.2115
Fedora Core 1
3.5 51.72 596.28 1.51 48 2004-02-09 mstock
PC
HP Vectra
Intel P3 1000 MHz 384 MB Linux 2.4.9 3.4.1 30.31 601.2 0.78 3 2002-04-15 chris
PC
ASUS A7A266
AMD Athlon Thunderbird 900 MHz 526 MB DDR Linux 2.4.7 3.4 29.71 647.88 0.77 - 2002-02-22 mjmeier
PC
ABIT KT7
AMD Athlon Thunderbird 900 MHz 256 MB Linux 2.2.17 3.1.20 25.95 694.57 0.96 - 2000-11-24 steve
Apple Powerbook Apple G4 867 MHz 640 MB OS X 10.3.7 3.6.1
2005-01-18
43.98 769.32 0.89 60 2005-01-18 grobe
Apple Powerbook Apple G4 867 MHz 640 MB OS X 10.3.7 3.6.1 44.30 792.14 0.88 60 2005-01-18 grobe
PC
Dell OP GX400
Intel P4 1400 MHz 256 MB Linux 2.2.19 3.1.20 31.28 794.57 0.74 - 2001-09-21 markusn
SunBlade 1500 _ ? MHz 1024 MB SunOS 5.8 3.6a 32.97 807.54 1.02 2003-11-12 mstock
Apple Powerbook Apple G4 867 MHz 640 MB OS X 10.3.7 3.6.0 43.76 853.88 0.92 2005-01-18 grobe
PC Intel P3 733 MHz 256 MB RDRAM Linux 2.2.17 3.1.20 41.23 875.54 1.06 - 2001-01-28 markmp
Compaq ProLiant DL590/64 Quad Itanium 800 MHz 4 GB Linux 2.4.7 3.4.1 55.36 1094.4 1.59 20 2002-06-18 chris
PC
Dell XPS T500
Intel P3 500 MHz 512 MB Linux 2.4.7-10 3.4.1 54.08 1100.47 1.49 13 2002-05-17 mstock
Mac Titanium Powerbook PMac G4 667 MHz 512 MB OS X 10.1.4 3.4 44.30 1102.18 1.12 - 2002-04-29 pillo.pillo
Compaq ProLiant DL590/64 Quad Itanium 800 MHz 4 GB Linux 2.4.7 3.4.1 55.7 1209.6 1.59 21 2002-06-18 chris
PC
ASUS P3B-F
Intel P3 500 MHz 128 MB Linux 2.2.14 3.1.20 62.50 1256.74 1.66 - 2000-08-23 yoichi
Mac PMac G4 450 MHz 1.2 GB OS X 10.1.1 3.4b 50.1 1286.65 1.25 - 2002-02-21 pillo.pillo
Mac iBook G3 500 MHz 384 MB OS X 10.1.2 3.4b 48.96 1307.80 1.24 - 2002-01-02 pillo.pillo
PC
Dell XPS T500
Intel P3 500 MHz 256 MB Linux 2.2.12-20 3.1.20 64.04 1368.86 1.90 - 2000-02-18 mstock
PC
Dell XPS T500
Intel P3 500 MHz 512 MB Linux 2.4.7-10 3.1.20 60.95 1369.85 1.68 - 2002-02-20 mstock
PC
Dell XPS T500
Intel P3 500 MHz 512 MB Linux 2.4.7-10 3.4 63.66 1383.44 1.70 - 2002-02-25 mstock
HP Brio Intel P2 350 MHz 256 MB Linux 2.4.18 3.4.1 78.31 1447.2 1.79 23 2002-08-07 chris
Sun
Ultra 5/400
UltraSPARC IIi 400 MHz 256 MB Solaris 8 3.4.1 51.4 1652.4 - 7 2002-04-15 chris
HP Brio Intel P2 350 MHz 256 MB Linux 2.4.18 3.4.1 84.13 1659.6 1.93 24 2002-08-07 chris
Sun
Ultra 5/400
UltraSPARC IIi 400 MHz 256 MB Solaris 8 3.4.1 55.2 1724.4 - 12 2002-05-07 chris
PC
Dell XPS T500
Intel P3 500 MHz 128 MB Linux 2.2.5-22 3.1.11 72.36 1787.78 1.88 - 1999-08-06 mstock
Digital
PWS 500a
alpha 21164 500 MHz 512 MB Linux 2.2.17 3.1.20 84.5 1842.6 1.9 2 2001-08-25 grobe
Sun
Ultra 5/440
UltraSPARC IIi 440 MHz 256 MB Solaris 8 3.4.1 56.98 1974.16 2.13 2002-05-25 mstock
Sun
Ultra 5/400
UltraSPARC IIi 400 MHz 256 MB Solaris 8 3.4.1 56.8 1976.4 - 11 2002-05-06 chris
laptop AMD K6-3+ 380 MHz 128 MB Linux 2.4.19 3.4 109 2051 3.25 2003-02-10 grobe
Sun
Ultra 5/400
UltraSPARC IIi 400 MHz 256 MB Solaris 8 3.4.1 56.5 2055.6 - 10 2002-05-06 chris
Sun
Ultra 5/400
UltraSPARC IIi 400 MHz 256 MB Solaris 8 3.4.1 56.47 2239.2 - 6 2002-04-15 chris
Digital
PWS 500a
alpha 21164 500 MHz 512 MB Linux 2.2.17 3.1.20 84. 2340. 2.5 1 2001-08-05 grobe
PC
pcchips vxpro+
K6-II 3dnow 415 MHz 96 MB Linux 2.4.4 3.1.20 124. 2490. 3. - 2001-08-05 grobe
PC (2) Intel P2 333 MHz 256 MB Linux 2.0.36 SMP 3.1.11 68.45 2681.71 2.37 - 1999-03-03 rspier
PC
Tyan S1692D
Intel P2 233 MHz 512 MB Linux 2.2.14 SMP 3.1.20 140.44 2791.35 3.46 - 2000-08-23 yoichi
PC
GIGABYTE GA-BX2000
Intel P2 233 MHz 512 MB Linux 2.2.14 SMP 3.1.20 143.64 2854.37 3.51 - 2000-08-23 yoichi
Sun E450 (4) 300 MHz MB - - 147.0 - - - 2000-02-18 p.griffiths
PC Intel P2 233 MHz 512 MB Windows NT 4.0 Desktop Rad 1.0 192.64 3143.53 3.60 - 2000-08-23 yoichi
HP 9000/782/C200 HP PA-8200 200 MHz 128 MB HP-UX B.10.20 3.1.4 119.13 5076.15 3.12 - 1999-11-12 markusn
alphaserver
1000 4/266
alpha 21064A 266 MHz 128 MB FreeBSD-4.7R 3.4.1 245.27 5854.05 3.51 25 2002-11-10 cjm
SGI (2) MIPS R10000 ? 256 MB IRIX 6.4 3.1.11 85.20 5962.40 2.74 - 1999-03-05 rspier
PC Pentium Pro MMX 233 MHz 128 MB Linux 2.0.36 3.1.8 187.25 6432.85 5.05 - 1999-01-23 lrsmith
PC AMD K5-133+ 100 MHz 64 MB Linux 2.4.7-10 3.4.1 309.80 7826.99 9.11 17 2002-05-22 mstock
Sun SPARC Ultra 30 UltraSPARC-II 248 MHz 128 MB Solaris 2.5.1 3.1.8 105.0 7978.0 4.0 - 1999-01-21 lrsmith
SGI O2 MIPS R5000, IP32 180 MHz 192 MB IRIX 6.3 3.1.5 140.72 9541.87 4.34 - 1999-01-20 mstock
PC AMD K5-133+ 100 MHz 64 MB Linux 2.0.36 3.1.8 291.11 12836.39 9.19 - 1999-01-22 mstock

Notes

(1) Using gcc
(2) Using Compaq compiler
(3) Using gcc 3.0.4 with -O2 (Radiance default)
(4) Using gcc 3.0.4 with -march=i686 -malign-double -mno-ieee-fp -O3 -ffast-math -fomit-frame-pointer -fstrict-aliasing
(5) Using Intel C 5.0.1 with -tpp6 -xK -O3 -ipo -ipo_obj
(6) Using Sun WorkShop 6 2000/04/07 C 5.1 with -O (Radiance default)
(7) Using Sun WorkShop 6 2000/04/07 C 5.1 with -fast
(8) Oconv probably used both processors; rpict used one; Using rad/rpiece on dual processors returned rendering times from 240 (20 to 25 divisions) to 266 (100 divisions) Author has PDF summarizing results
(9) Using gcc 3.1 with -march=pentium3 -malign-double -mno-ieee-fp -mfpmath=sse -O3 -ffast-math -fomit-frame-pointer -fstrict-aliasing
(10) Using gcc 2.95.2 with -O2
(11) Using gcc 2.95.2 with -O3 -ffast-math -fomit-frame-pointer
(12) Using gcc 3.2 20020506 (experimental) with -mcpu=ultrasparc -O3 -ffast-math -fomit-frame-pointer -fstrict-aliasing
(13) Using gcc 2.96 with -march=i686 -malign-double -mno-ieee-fp -O3 -ffast-math -fomit-frame-pointer -fstrict-aliasing -DSORT_THRESH=0x7fffffffffffffff -DMC
(14) Using precompiled x86 binaries
(15) Using gcc 2.96, default compiler options
(16) Using gcc 2.96 with -march=i686 -malign-double -mno-ieee-fp -O3 -ffast-math -fomit-frame-pointer -fstrict-aliasing
(17) Using gcc 2.96 with -march=i586 -malign-double -mno-ieee-fp -O3 -ffast-math -fomit-frame-pointer -fstrict-aliasing
(18) Using Compaq ccc -arch host -fast on a loaded machine
(19) Using Compaq cc -O on a loaded machine
(20) Using Intel C 5 -O3 -ipo -ipo_obj on a loaded machine
(21) Using gcc 2.96 20000731 -O3 -ffast-math -fomit-frame-pointer -fstrict-aliasing on a loaded machine
(22) Using gcc 3.3 20020823 (experimental), with -march=pentium3 -malign-double -mno-ieee-fp -mfpmath=sse -O3 -ffast-math -fomit-frame-pointer -fstrict-aliasing -ftracer
(23) Using gcc 3.1.1, with -march=pentium2 -malign-double -mno-ieee-fp -O2 -finline-functions -ffast-math -fomit-frame-pointer -fstrict-aliasing
(24) Using gcc 3.1.1, with -O2
(25) Using compaq-cc, with -O4
(26) Using gcc 3.2-7
(27) Using gcc 3.2
(28) Using gcc 3.2 with -march=pentium4 -O3 -ffast-math -fomit-frame-pointer
(29) Using Intel C 7.0 with -xW -O3 -ip
(30) Using Intel C 7.0 with -xW -O3 -ipo -ipo_obj
(31) Using Intel C 7.0 with -xW -O3 -ipo -ipo_obj -fno-alias
(32) Using gcc 3.2 with -march=pentium4 -malign-double -mno-ieee-fp -mfpmath=sse -O3 -ffast-math -fomit-frame-pointer -fstrict-aliasing
(33) Using gcc 3.3 20021213 (experimental) with -march=pentium4 -malign-double -mno-ieee-fp -mfpmath=sse -O3 -ffast-math -fomit-frame-pointer -fstrict-aliasing -ftracer
(34) Using the precompiled binaries from http://www.dream.unipa.it/dream/pub/dot/anselmo/radiance/packages/, specifically, radiancestatic-3.5.0-2.i386.rpm
(35) Using gcc (GCC) 3.2 20020903 (Red Hat Linux 8.0 3.2-7) with default options
(36) Using gcc (GCC) 3.2 20020903 with -march=i686 -malign-double -mno-ieee-fp -ffast-math -fomit-frame-pointer -fstrict-aliasing
(37) Using gcc (GCC) 3.2 20020903 with -march=i686 -malign-double -mno-ieee-fp -O3 -ffast-math -fomit-frame-pointer -fstrict-aliasing
(38) Using gcc 3.3 with -march=pentium4 -malign-double -mno-ieee-fp -mfpmath=sse -O3 -fomit-frame-pointer -ffast-math -ftracer
(39) Using gcc 3.3 with -march=pentium3 -malign-double -mno-ieee-fp -O3 -fomit-frame-pointer -ffast-math -ftracer
(40) Using gcc 3.1 with -O2, oconv was utilizing both processors, wall clock time was 23.08; rpict and pfilt used one processor
(41) gcc (GCC) 3.2.2 20030222, -DMC
(42) gcc (GCC) 3.2.2 20030222, -DMC -march=athlon-xp -ffast-math -mno-ieee-fp
(43) gcc 3.3, -fast
(44) gcc, -fast
(45) gcc 3.3, default
(46) gcc 3.3, -fast -mcpu=7450
(47) using statically-linked libraries: radbin_lnx_static.tar.gz
(48) gcc (GCC) 3.3.2 20031022, -DMC
(49) gcc (GCC) 3.3.2 20031022, -DMC -O3 -march=pentium4 -malign-double -mno-ieee-fp -ffast-math
(50) gcc (GCC) 3.3.2 20031022, 1 processor working on a 2-procesor system, -O2 -m64
(51) gcc (GCC) 3.3.2 20031022, 1 processor working on a 2-procesor system, -O3 -m64 -ffast-math -mno-ieee-fp -funroll-loops; running rpiece on 2 processor (1 machine, 8x8 divisions) time was 108s (108s,108s); 4 processor (over 2 machines, 8x8 divisions) time was 61s (58s,60s,61s,61s)
(52) using statically-linked libraries: radbin_lnx_static.tar.gz
(53) using the precompiled binaries from http://www.dream.unipa.it/dream/pub/dot/anselmo/radiance/packages/, specifically radiancepmap-3.4.1-2.i386.rpm
(54) 1 processor working on a 2-procesor system, -fast instead of -O2; rpiece on 2 processors, 20x20 grid, did rpict in 180.69 s
(55) gcc (GCC) 3.3.2 20031022, 1 processor working on a 2-procesor system, -m64
(56) gcc (GCC) 3.3.2 20031022, 1 processor working on a 2-procesor system, -m64-O
(57) gcc 3.3.4, -DMC -march=pentium4 -malign-double -mno-ieee-fp -mfpmath=sse -O3 -ffast-math -fomit-frame-pointer -fstrict-aliasing -ftracer; with a second 2254 MHz P4 node, rpiece (4x4 divisions) returned 156.92 for the render
(58) gcc (GCC) 3.4.2 20041017 (Red Hat 3.4.2-6.fc3), -DMC -march=athlon-xp -ffast-math -mno-ieee-fp
(59) gcc version 3.4.2 [FreeBSD] 20040728, -O3 -m64 -ffast-math -mno-ieee-fp -funroll-loops
(60) GCC 3.3, -O2 -DSPEED=200 -mcpu=7400 -mtune=7400 -malign-natural -funroll-loops -fstrict-aliasing -fsched-interblock -falign-loops=16 -falign-jumps=16 -falign-functions=16 -falign-jumps-max-skip=15 -falign-loops-max-skip=15 -ffast-math -mpowerpc-gpopt -fstrict-aliasing
(61) gcc version 3.4.3 20050110, -DMC -march=athlon-xp -O3 -mfpmath=sse,387 -pipe -fomit-frame-pointer -ffast-math -malign-double -mno-ieee-fp -DSPEED=200
(62) gcc version 4.4.5 20101112 (Red Hat 4.4.5-2) (GCC), -O2 -march=native -ffast-math -funroll-loops


Page last modified 2011-03-25
Mark J. Stock, PhD