CPU Speed?

I am running openSuSE 11.1 (KDE 3.5). I had to rebuild my system and just dropped the old Drives and with its with its data and OS intact in the new system and it is playing great but I was curious about my processor speed. I used the command “cat /proc/cpuinfo” to find out what speed my CPU was currently running. (AMD Phenom II X4 955 3.2 gHz) It showed 800mHz. I then launched a YouTube Flash video and a local video file at the same time. It’s still showing 800mHz while those are running when I run “cat /proc/cpuinfo” I would expect the CPU to increase it’s speed as Flash is a pig and the h264 video is being rendered by the CPUs. Any Ideas?

flamebait@linux-v5xu:~> cat /proc/cpuinfo
processor : 0
vendor_id : AuthenticAMD
cpu family : 16
model : 4
model name : AMD Phenom™ II X4 955 Processor
stepping : 2
cpu MHz : 800.000
cache size : 512 KB
physical id : 0
siblings : 4
core id : 0
cpu cores : 4
apicid : 0
initial apicid : 0
fpu : yes
fpu_exception : yes
cpuid level : 5
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc rep_good nopl pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt
bogomips : 6428.38
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 48 bits physical, 48 bits virtual
power management: ts ttp tm stc 100mhzsteps hwpstate

processor : 1
vendor_id : AuthenticAMD
cpu family : 16
model : 4
model name : AMD Phenom™ II X4 955 Processor
stepping : 2
cpu MHz : 800.000
cache size : 512 KB
physical id : 0
siblings : 4
core id : 1
cpu cores : 4
apicid : 1
initial apicid : 1
fpu : yes
fpu_exception : yes
cpuid level : 5
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc rep_good nopl pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt
bogomips : 6428.48
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 48 bits physical, 48 bits virtual
power management: ts ttp tm stc 100mhzsteps hwpstate

processor : 2
vendor_id : AuthenticAMD
cpu family : 16
model : 4
model name : AMD Phenom™ II X4 955 Processor
stepping : 2
cpu MHz : 800.000
cache size : 512 KB
physical id : 0
siblings : 4
core id : 2
cpu cores : 4
apicid : 2
initial apicid : 2
fpu : yes
fpu_exception : yes
cpuid level : 5
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc rep_good nopl pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt
bogomips : 6428.42
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 48 bits physical, 48 bits virtual
power management: ts ttp tm stc 100mhzsteps hwpstate

processor : 3
vendor_id : AuthenticAMD
cpu family : 16
model : 4
model name : AMD Phenom™ II X4 955 Processor
stepping : 2
cpu MHz : 800.000
cache size : 512 KB
physical id : 0
siblings : 4
core id : 3
cpu cores : 4
apicid : 3
initial apicid : 3
fpu : yes
fpu_exception : yes
cpuid level : 5
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc rep_good nopl pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt
bogomips : 6428.53
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 48 bits physical, 48 bits virtual
power management: ts ttp tm stc 100mhzsteps hwpstate

flamebait@linux-v5xu:~> cat /proc/cpuinfo
processor : 0
vendor_id : AuthenticAMD
cpu family : 16
model : 4
model name : AMD Phenom™ II X4 955 Processor
stepping : 2
cpu MHz : 800.000
cache size : 512 KB
physical id : 0
siblings : 4
core id : 0
cpu cores : 4
apicid : 0
initial apicid : 0
fpu : yes
fpu_exception : yes
cpuid level : 5
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc rep_good nopl pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt
bogomips : 6428.38
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 48 bits physical, 48 bits virtual
power management: ts ttp tm stc 100mhzsteps hwpstate

processor : 1
vendor_id : AuthenticAMD
cpu family : 16
model : 4
model name : AMD Phenom™ II X4 955 Processor
stepping : 2
cpu MHz : 800.000
cache size : 512 KB
physical id : 0
siblings : 4
core id : 1
cpu cores : 4
apicid : 1
initial apicid : 1
fpu : yes
fpu_exception : yes
cpuid level : 5
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc rep_good nopl pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt
bogomips : 6428.48
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 48 bits physical, 48 bits virtual
power management: ts ttp tm stc 100mhzsteps hwpstate

processor : 2
vendor_id : AuthenticAMD
cpu family : 16
model : 4
model name : AMD Phenom™ II X4 955 Processor
stepping : 2
cpu MHz : 800.000
cache size : 512 KB
physical id : 0
siblings : 4
core id : 2
cpu cores : 4
apicid : 2
initial apicid : 2
fpu : yes
fpu_exception : yes
cpuid level : 5
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc rep_good nopl pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt
bogomips : 6428.42
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 48 bits physical, 48 bits virtual
power management: ts ttp tm stc 100mhzsteps hwpstate

processor : 3
vendor_id : AuthenticAMD
cpu family : 16
model : 4
model name : AMD Phenom™ II X4 955 Processor
stepping : 2
cpu MHz : 800.000
cache size : 512 KB
physical id : 0
siblings : 4
core id : 3
cpu cores : 4
apicid : 3
initial apicid : 3
fpu : yes
fpu_exception : yes
cpuid level : 5
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc rep_good nopl pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt
bogomips : 6428.53
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 48 bits physical, 48 bits virtual
power management: ts ttp tm stc 100mhzsteps hwpstate

FlameBait adjusted his/her AFDB on Mon 12 July 2010 03:46 to write:

>
> I am running openSuSE 11.1 (KDE 3.5). I had to rebuild my system and
> just dropped the old Drives and with its with its data and OS intact in
> the new system and it is playing great but I was curious about my
> processor speed. I used the command “cat /proc/cpuinfo” to find out what
> speed my CPU was currently running. (AMD Phenom II X4 955 3.2 gHz) It
> showed 800mHz. I then launched a YouTube Flash video and a local video
> file at the same time. It’s still showing 800mHz while those are running
> when I run “cat /proc/cpuinfo” I would expect the CPU to increase it’s
> speed as Flash is a pig and the h264 video is being rendered by the
> CPUs. Any Ideas?

What was the actual CPU usage before and during the playback?

Also did they play OK?


Mark
Caveat emptor
Nullus in verba
Nil illegitimi carborundum

I am not sure of the percentage of CPU resources used.

They played fine. I did an further experiment and was running several IRC sessions, my email client, running a torrent, using Skype (voice only), playing a flash video from YouTube and running a h264 video at the same time and then ran the cat /proc/cpuinfo command again and got the same results. It didn’t skip a beat though. 800mHz may be the minimum speed it steps down to from 3.2gHz and not the actual speed it’s running. I think it would have choked somewhere on all it was doing otherwise.

The system is way faster than it was with my dual core Athlon that ran 2.8gHz all the time. (there was no stepping on that CPU version)

Very interesting.

cpu MHz as 1000.00
for OpesSuse 11.1/AMD X2 64 6000+ and AMD 64 3000+/OS 11.2.

But on 11.1 when I run hwinfo, cpu MHz for /proc/cpuinfo shows 3000.000 which is correct.

My guess is /proc/cpuinfo has a bug reporting AMD cpu MHz.

tararpharazon wrote:

>
> Very interesting.
>
> cpu MHz as 1000.00
> for OpesSuse 11.1/AMD X2 64 6000+ and AMD 64 3000+/OS 11.2.
>
> But on 11.1 when I run hwinfo, cpu MHz for /proc/cpuinfo shows
> 3000.000 which is correct.

Exactly, it depends on the load at the time.

> My guess is /proc/cpuinfo has a bug reporting AMD cpu MHz.

I doubt it.


Per Jessen, Zürich (28.2°C)
http://en.opensuse.org/User:Pjessen

On Mon, 2010-07-12 at 11:36 +0000, tararpharazon wrote:
> Very interesting.
>
> cpu MHz as 1000.00
> for OpesSuse 11.1/AMD X2 64 6000+ and AMD 64 3000+/OS 11.2.
>
> But on 11.1 when I run hwinfo, cpu MHz for /proc/cpuinfo shows 3000.000
> which is correct.
>
> My guess is /proc/cpuinfo has a bug reporting AMD cpu MHz.
>
>

Actually, my testing found that the only good source of data regarding
CPU was out of sysfs.

/sys/devices/system/cpu/cpu*/cpufreq/*_freq

But because this is sysfs, the location, etc of the above may vary from
kernel to kernel.

cjcox I tried that and it doesn’t work.

flamebait@linux-v5xu:~> /sys/devices/system/cpu/cpu*/cpufreq/_freq
bash: /sys/devices/system/cpu/cpu0/cpufreq/cpuinfo_cur_freq: Permission denied
flamebait@linux-v5xu:~> su
Password:
linux-v5xu:/home/flamebait # /sys/devices/system/cpu/cpu
/cpufreq/*_freq
bash: /sys/devices/system/cpu/cpu0/cpufreq/cpuinfo_cur_freq: Permission denied

FlameBait wrote:
> Permission denied

first, “/sys/devices/system/cpu/cpu*/cpufreq/*_freq” is not an
executable, instead it is a location…a file that you can look inside…

open a text editor (like KWrite) as root:

go Alt+F2 to get a run dialog, then type in “kdesu kwrite” and hit
enter, give root pass and then nav to

/sys/devices/system/cpu/cpu[a number]/cpufreq/[something]_freq

here [a number] = 0, and [something] is one of these:

cpuinfo_cur
cpuinfo_max
cpuinfo_min
scaling_max
scaling_min

pick the one you wanna look at and open it…

CAREFUL, do not change or save the file you open…


Enjoy Packman’s benefits? Show your appreciation by donating at
http://se.unixheads.org/?donate

DenverD
CAVEAT: http://is.gd/bpoMD [posted via NNTP w/openSUSE 10.3]

Very good. I understand now. I’ll check it later under a load. Right now it’s throttled down to 800mHz when I looked under no real load (only Xchat, Firefox and Thunderbird running) when I look at the different cores.
How I checked was to open a terminal, log in as root, launch konqueror from CLI and navigate to file location running Konqueror as root.

Thanks for the URL to donate to Packaman. A donation has been made.

FlameBait wrote:
> How I checked was to open a terminal, log in as root, launch konqueror
> from CLI and navigate to file location running Konqueror as root.

good idea using konqueror (instead of a text editor) as it offers
little chance of accidentally corrupting…

> Thanks for the URL to donate to Packaman. A donation has been made.

good going…


Enjoy Packman’s benefits? Show your appreciation by donating at
http://se.unixheads.org/?donate

DenverD
CAVEAT: http://is.gd/bpoMD [posted via NNTP w/openSUSE 10.3]

Have you tried looking at clock speeds with ksysguard? It gives nice graphs (although it does smooth the data somewhat…a running average, rather than an absolute instantaneous value) and is capable of much, much more than it is configured to do by default.

Watch for a while and then start a few heavy users of cpu; you should see the effect on clocks and the memory used.

markone that actually worked quite well. I was able to see the speed bump up on the different cores once I set up a workspace for their speed.

This CPU is hardly taxed under my usage.

I still have a worry in the back of my brain about CPU throttling back due to heat. I worry that I put the proper amount of thermal transfer as I had to put Arctic Silver Céramique on the stock AMD heat sink as my first motherboard was DOA. It seems my fear are unfounded so far.

FlameBait wrote:
> I worry that I put the proper amount of thermal transfer

the main thing to know about thermal goop is don’t over do it…more
is not better…

the amount the size of a grain of rice is plenty…UNCOOKED rice, not
big and fluffy…spread as evenly and thinly as you can…i use (like)
a business card to move to scrape it evenly side to side…


Enjoy Packman’s benefits? Show your appreciation by donating at
http://se.unixheads.org/?donate

DenverD
CAVEAT: http://is.gd/bpoMD [posted via NNTP w/openSUSE 10.3]

The proper amount of thermal paste is not very much…

Basically, it fills what would otherwise be air voids between the not-flat-at-the-atomic-level surfaces. With any luck, the retention pressure is enough to squeeze out excess paste, but that does lead to excess paste (excess conductive paste) being squeezed out of the sides, and that can be a disaster, if it squeezes into the wrong places.

I use the ‘credit card’ technique, using an old card to spread material at right angles to any machining marks on the surfaces (so that it fills into the valleys), but you’ll find all sorts of things recommended.

…about CPU throttling back due to heat…

At the risk of opening another can of worms, monitor the temperature, then. Conky?

This CPU is hardly taxed under my usage.

Well, yes, it seems much more likely that this is being caused by a low level of load (good!) than a high level of temperature (bad!), but I am not going to be able to tell which is which from here. Looking at the numbers now, when it is known to be ok, so you are prepared for some future time, when you have suspicions, is key.

On Wed, 2010-07-14 at 06:15 +0000, DenverD wrote:
> FlameBait wrote:
> > How I checked was to open a terminal, log in as root, launch konqueror
> > from CLI and navigate to file location running Konqueror as root.
>
> good idea using konqueror (instead of a text editor) as it offers
> little chance of accidentally corrupting…
>
> > Thanks for the URL to donate to Packaman. A donation has been made.
>
> good going…
>

Us old people just use cat :slight_smile:

cjcox wrote:
> Us old people just use cat :slight_smile:

10-4 good buddy.


Enjoy Packman’s benefits? Show your appreciation by donating at
http://se.unixheads.org/?donate

DenverD
CAVEAT: http://is.gd/bpoMD [posted via NNTP w/openSUSE 10.3]

“Céramique does not contain any metal or other electrically conductive materials. It is a pure electrical insulator, neither electrically conductive nor capacitive.”

Arctic Silver Incorporated - Céramique

I used more than a rice grain and gave up trying to tint the heat sink the second time it flipped over and landed thermal sink down on my work bench. I just let the excess squish out.

I haven’t got the thermal sensors working yet. I have done “sensors-detect”

FlameBait wrote:
> “Céramique does not contain any metal or other electrically conductive
> materials. It is a pure electrical insulator, neither electrically
> conductive nor capacitive.”

who in this thread said it conducted anything other than heat?


Enjoy Packman’s benefits? Show your appreciation by donating at
http://se.unixheads.org/?donate

DenverD
CAVEAT: http://is.gd/bpoMD [posted via NNTP w/openSUSE 10.3]

markone commented about conductive paste being a bad deal.

Still trying to figure out why I don’t have thermal sensors. “Sensors” must not exist for this CPU?

If sensors is correctly installed did you start with

sensors-detect