From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?ISO-8859-2?Q?Rafa=B3_Bilski?= Subject: Re: Power measurements Date: Mon, 25 Jun 2007 19:55:23 +0200 Message-ID: <4680018B.9010902@interia.pl> References: <467F633B.3000909@interia.pl> <1182762651706@dmwebmail.belize.chezphil.org> Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Return-path: In-Reply-To: <1182762651706@dmwebmail.belize.chezphil.org> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: cpufreq-bounces@lists.linux.org.uk Errors-To: cpufreq-bounces+glkc-cpufreq=gmane.org+glkc-cpufreq=gmane.org@lists.linux.org.uk Content-Type: text/plain; charset="iso-8859-9" To: Phil Endecott Cc: cpufreq@lists.linux.org.uk >>>> % dmesg | grep eps=20 >>> >>> Thanks, yes, I found this after I posted my last message. Apparently it >>> should use 860 mV at 1.2 GHz and 844 mV at 400 MHz. >>> >> Most C7-Eden CPU's have same voltage on high and low frequency. Looks >> like Your's has different voltages. If You don't see 844mV at 400MHz >> then probably VID pins aren't connected to VRM on motherboard. >> Mode MHz Voltage (mV) Power (W) >> P0 1200 860 7 >> P1 400 844 3 >=20 > May I ask where these numbers come from? ACPI tables. > [...] >> You should probably poke some registers in northbridge. I don't know >> much about newer VIA chipsets, but older (CLE266) had many bits which >> could improve power savings - dynamic clock stop to different chipset >> components, ACPI C3 can put memory to sleep and so on. Unfortunatly >> these bits are disabled by default and are never touched by BIOS. >=20 > Is there any documentation available for any of this? No. All datasheets are NDA protected. However You can find in Google=20 datasheets for older chips. > Hmmm, I really should learn something about ACPI. Well, ACPI C3 isn't available anyway due to VIA's policy. Check=20 "/proc/acpi/processor/CPU0/power" file for supported C states. There is=20 almost no difference in power consumption between ACPI C1 and C2. >> [...] >=20 > Here is nbench for my Jetway J7F2 with VIA C7 @ 1.2G with the 'ondemand' > governor: Asrock P4S61 with P4-M @ 1,2GHz (12x100MHz) @ 1,3V. Without cpufreq. Note that P4-M is starting in "battery optimized mode" which means x12=20 multiplier. I'm not able to change multiplier to x17 nor enable "Hyper=20 Threading" because Asrock says: "sorry, this motherboard isn't=20 supporting P4-M". All I can do is to change FSB to 133MHz and have=20 1,6GHz (12x133MHz). > BYTEmark* Native Mode Benchmark ver. 2 (10/95) > Index-split by Andrew D. Balsa (11/97) > Linux/Unix* port by Uwe F. Mayer (12/96,11/97) >=20 > TEST : Iterations/sec. : Old Index : New Index > : : Pentium 90* : AMD K6/233* > --------------------:------------------:-------------:------------ > NUMERIC SORT : 354.96 : 9.10 : 2.99 NUMERIC SORT : 425.52 : 10.91 : 3.58 > STRING SORT : 39.736 : 17.76 : 2.75 STRING SORT : 25.959 : 11.60 : 1.80 > BITFIELD : 1.1172e+08 : 19.16 : 4.00 BITFIELD : 1.4342e+08 : 24.60 : 5.14 > FP EMULATION : 46.664 : 22.39 : 5.17 FP EMULATION : 42.897 : 20.58 : 4.75 > FOURIER : 2729.8 : 3.10 : 1.74 FOURIER : 6879.9 : 7.82 : 4.39 > ASSIGNMENT : 8.4321 : 32.09 : 8.32 ASSIGNMENT : 11.601 : 44.14 : 11.45 > IDEA : 1042.5 : 15.94 : 4.73 IDEA : 1214.2 : 18.57 : 5.51 > HUFFMAN : 525.7 : 14.58 : 4.66 HUFFMAN : 465.84 : 12.92 : 4.13 > NEURAL NET : 4.7302 : 7.60 : 3.20 NEURAL NET : 10.503 : 16.87 : 7.10 > LU DECOMPOSITION : 298.8 : 15.48 : 11.18 LU DECOMPOSITION : 459.6 : 23.81 : 17.19 > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3DORIGINAL BYTEMARK > RESULTS=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D > INTEGER INDEX : 17.572 > FLOATING-POINT INDEX: 7.148 > Baseline (MSDOS*) : Pentium* 90, 256 KB L2-cache, Watcom* compiler 10.0 INTEGER INDEX : 18.267 FLOATING-POINT INDEX: 14.648 > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3DLINUX DATA > BELOW=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D > CPU : CentaurHauls VIA Esther processor 1200MHz 1200MHz > L2 Cache : 128 KB > OS : Linux 2.6.21-1-686 > C compiler : gcc version 4.1.3 20070601 (prerelease) (Debian > 4.1.2-12) > libc : libc-2.5.so > MEMORY INDEX : 4.507 > INTEGER INDEX : 4.295 > FLOATING-POINT INDEX: 3.964 > Baseline (LINUX) : AMD K6/233*, 512 KB L2-cache, gcc 2.7.2.3, > libc-5.4.38 > * Trademarks are property of their respective holder. CPU : GenuineIntel Mobile Intel(R) Pentium(R) 4 - M CPU 1.7= 0GHz 1203MHz L2 Cache : 512 KB OS : Linux 2.6.22-rc4 C compiler : 4.1.2 libc : MEMORY INDEX : 4.727 INTEGER INDEX : 4.436 FLOATING-POINT INDEX: 8.124 >=20 > [...] >=20 > How does this compare with your Intel chip? Intel seems to have some weak points. Sometimes it is faster. But at 1,2GHz= =20 it needs *26,5W* and *10,7W* in ACPI C1. At 1,7GHz it needs *38,9W*. IMHO=20 VIA wins this round. I wish that C7-M for Socket478 would exists... >=20 > Regards, >=20 > Phil. Regards Rafa=B3 ---------------------------------------------------------------------- Ile masz w domu niepotrzebnych rzeczy? Wymien sie z sasiadami >> http://link.interia.pl/f1a93