+- +-
Say hello if visiting :) by Gecko
11 Jan 2023, 07:43:05 pm

Seti is down again by Mike
09 Aug 2017, 10:02:44 am

Some considerations regarding OpenCL MultiBeam app tuning from algorithm view by Raistmer
11 Dec 2016, 06:30:56 am

Loading APU to the limit: performance considerations by Mike
05 Nov 2016, 06:49:26 am

Better sleep on Windows - new round by Raistmer
26 Aug 2016, 02:02:31 pm

Author Topic: SETI MB CUDA for Linux  (Read 387526 times)

Offline riofl

  • Knight o' The Round Table
  • ***
  • Posts: 240
Re: SETI MB CUDA for Linux
« Reply #570 on: 19 Jan 2010, 11:41:37 pm »
i have been looking around and found that the G0 stepping is a lower voltage version of the previous B3 stepping so yes that VID may be right

intel says this about the VID voltage range of the G0

VID Voltage Range   0.85V-1.5V

Offline riofl

  • Knight o' The Round Table
  • ***
  • Posts: 240
Re: SETI MB CUDA for Linux
« Reply #571 on: 19 Jan 2010, 11:55:27 pm »
my friend has an asus rampage formula mobo with his q6600 G0 running at 3.0ghz also and his VID says 1.22 so i guess that is what it is. i wont be able to look at my bios until i can reboot which will be sometime tomorrow if this one application finishes its work before tomorrow night.

Offline sunu

  • Alpha Tester
  • Knight who says 'Ni!'
  • ***
  • Posts: 771
Re: SETI MB CUDA for Linux
« Reply #572 on: 20 Jan 2010, 06:21:10 am »
Lets do something else. Post again the above values with 0% cpu utilization (idle, nothing runs) and 100%.

Offline riofl

  • Knight o' The Round Table
  • ***
  • Posts: 240
Re: SETI MB CUDA for Linux
« Reply #573 on: 20 Jan 2010, 12:23:42 pm »
Lets do something else. Post again the above values with 0% cpu utilization (idle, nothing runs) and 100%.

with boinc set to use 100% cpu on 3 cpus the 4th is always reserved for cuda and my desktop. cuda is set for 0.35cpus

sensors
coretemp-isa-0000     
Adapter: ISA adapter
Core 0:      +55.0°C  (high = +84.0°C, crit = +100.0°C)

coretemp-isa-0001
Adapter: ISA adapter
Core 1:      +50.0°C  (high = +84.0°C, crit = +100.0°C)

coretemp-isa-0002
Adapter: ISA adapter
Core 2:      +50.0°C  (high = +84.0°C, crit = +100.0°C)

coretemp-isa-0003
Adapter: ISA adapter
Core 3:      +50.0°C  (high = +84.0°C, crit = +100.0°C)

it8718-isa-0290
Adapter: ISA adapter
in0:         +1.22 V  (min =  +0.00 V, max =  +4.08 V)
in1:         +1.89 V  (min =  +0.00 V, max =  +4.08 V)
in2:         +3.22 V  (min =  +0.00 V, max =  +4.08 V)
in3:         +2.94 V  (min =  +0.00 V, max =  +4.08 V)
in4:         +1.87 V  (min =  +0.00 V, max =  +4.08 V)
in5:         +0.08 V  (min =  +0.00 V, max =  +4.08 V)
in6:         +1.04 V  (min =  +0.00 V, max =  +4.08 V)
in7:         +2.96 V  (min =  +0.00 V, max =  +4.08 V)
in8:         +3.30 V
fan1:       2288 RPM  (min =    0 RPM)
fan2:       2115 RPM  (min =    0 RPM)
fan3:       1406 RPM  (min =    0 RPM)
fan4:       1415 RPM  (min =    0 RPM)
temp1:       +39.0°C  (low  = +127.0°C, high = +127.0°C)  sensor = thermistor
temp2:       +43.0°C  (low  = +127.0°C, high = +127.0°C)  sensor = thermal diode
temp3:        -2.0°C  (low  = +127.0°C, high = +127.0°C)  sensor = thermistor
cpu0_vid:   +1.219 V


with boinc shut down and time given for everything to cool off average cpu usage is 0 to 1% peaking at 5% momentarily

sensors
coretemp-isa-0000     
Adapter: ISA adapter
Core 0:      +40.0°C  (high = +84.0°C, crit = +100.0°C)

coretemp-isa-0001
Adapter: ISA adapter
Core 1:      +39.0°C  (high = +84.0°C, crit = +100.0°C)

coretemp-isa-0002
Adapter: ISA adapter
Core 2:      +38.0°C  (high = +84.0°C, crit = +100.0°C)

coretemp-isa-0003
Adapter: ISA adapter
Core 3:      +38.0°C  (high = +84.0°C, crit = +100.0°C)

it8718-isa-0290
Adapter: ISA adapter
in0:         +1.23 V  (min =  +0.00 V, max =  +4.08 V)
in1:         +1.89 V  (min =  +0.00 V, max =  +4.08 V)
in2:         +3.23 V  (min =  +0.00 V, max =  +4.08 V)
in3:         +2.96 V  (min =  +0.00 V, max =  +4.08 V)
in4:         +0.69 V  (min =  +0.00 V, max =  +4.08 V)
in5:         +0.08 V  (min =  +0.00 V, max =  +4.08 V)
in6:         +0.42 V  (min =  +0.00 V, max =  +4.08 V)
in7:         +2.98 V  (min =  +0.00 V, max =  +4.08 V)
in8:         +3.30 V
fan1:       1917 RPM  (min =    0 RPM)
fan2:       2115 RPM  (min =    0 RPM)
fan3:       1406 RPM  (min =    0 RPM)
fan4:       1415 RPM  (min =    0 RPM)
temp1:       +38.0°C  (low  = +127.0°C, high = +127.0°C)  sensor = thermistor
temp2:       +33.0°C  (low  = +127.0°C, high = +127.0°C)  sensor = thermal diode
temp3:        -2.0°C  (low  = +127.0°C, high = +127.0°C)  sensor = thermistor
cpu0_vid:   +1.219 V

Offline riofl

  • Knight o' The Round Table
  • ***
  • Posts: 240
Re: SETI MB CUDA for Linux
« Reply #574 on: 20 Jan 2010, 12:25:25 pm »
if i truly need 0% i will have to wait till i can reboot and then kill off X and stop all my background processes like storegrid backups, openfire jabber server etc and then leave only the server os running.


b0b3r

  • Guest
Re: SETI MB CUDA for Linux
« Reply #575 on: 20 Jan 2010, 04:49:29 pm »
Hello

Can you also include output of "nvidia-smi -lsa" command when gpus are on load?

Offline riofl

  • Knight o' The Round Table
  • ***
  • Posts: 240
Re: SETI MB CUDA for Linux
« Reply #576 on: 20 Jan 2010, 07:31:25 pm »
Hello

Can you also include output of "nvidia-smi -lsa" command when gpus are on load?


i didnt even know that existed... so it is a temp utility... my devices 0 and 1  run between 57 and 66c, most commonly running around 64c  as reported by gkrellm. the nvidia-smi output is:

nvidia-smi -lsa

==============NVSMI LOG==============


Timestamp                       : Wed Jan 20 19:26:05 2010

GPU 0:
        Product Name            : GeForce GTX 285
        Serial                  : 3169719755757
        PCI ID                  : 5e310de
        Temperature             : 63 C
GPU 1:
        Product Name            : Tesla C1060
        Serial                  : 837485170935
        PCI ID                  : 5e710de
        Temperature             : 61 C

and yes they are running full-bore constantly. i keep both fans set at 100% since i cannot change the card bios settings permanently to make the auto more aggessive. wish they had something like cmos that you could change the temp ranges vs fan speeds permanently without always having to fiddle with the driver. for a harder test the device 0 runs 68-71c after 5 minutes of glxgears running.

i try to keep the ambient temp in my office between 19 and 22c to help cool the systems



« Last Edit: 20 Jan 2010, 07:46:46 pm by riofl »

Offline sunu

  • Alpha Tester
  • Knight who says 'Ni!'
  • ***
  • Posts: 771
Re: SETI MB CUDA for Linux
« Reply #577 on: 20 Jan 2010, 09:32:41 pm »
Well, you never know, seems like cpu0_vid is the VID after all. Still, I'd like to see it from windows too.

With this VID and this motherboard I think you can easily go to3.6GHz if you have a decent cooler.

Offline riofl

  • Knight o' The Round Table
  • ***
  • Posts: 240
Re: SETI MB CUDA for Linux
« Reply #578 on: 20 Jan 2010, 10:51:35 pm »
Well, you never know, seems like cpu0_vid is the VID after all. Still, I'd like to see it from windows too.

With this VID and this motherboard I think you can easily go to3.6GHz if you have a decent cooler.

i know i can.. i ran it for a week at 3.6ghz but temps were running higher than i like. cpus were always in the mid to high 60s.  71c is the max operating temp before the cpu should go into lower power mode so i chose the highest rating keeping the cpu temps at a decent level and that was 3.0ghz.. at this level they never momentarily peak higher than 65c on a very hot florida day. i always keep my systems as cool as is practical.. heat kills..  thats why i put up with a cooler than i would like ambient environment.. it helps everything.

this weekend will be the earliest i can down the system long enough to put the windows drive in and run some program.. i hear cpuz is supposed to be a winner in measurements so ill download that and make a cd for the weekend unless you have another favorite i should use.


« Last Edit: 20 Jan 2010, 10:58:28 pm by riofl »

Offline riofl

  • Knight o' The Round Table
  • ***
  • Posts: 240
Re: SETI MB CUDA for Linux
« Reply #579 on: 20 Jan 2010, 10:57:17 pm »
Well, you never know, seems like cpu0_vid is the VID after all. Still, I'd like to see it from windows too.

With this VID and this motherboard I think you can easily go to3.6GHz if you have a decent cooler.

forgot... the cooler is a Zalman CNPS 9700 NT and if you want to see what that is doing 'fan1' in the sensors report is the cpu cooler fan. i tried it on full throttle but it wasnt worth making it work that hard. not enough appreciable difference in cooling so i leave it on auto in the motherboard.


Offline sunu

  • Alpha Tester
  • Knight who says 'Ni!'
  • ***
  • Posts: 771
Re: SETI MB CUDA for Linux
« Reply #580 on: 20 Jan 2010, 11:31:29 pm »
Don't bother with windows but if you're going to do it anyway or for future reference:

CPU-Z from http://www.cpuid.com/cpuz.php Among other things it will show the vcore.
Core Temp http://www.alcpu.com/CoreTemp/ Most importantly it shows the VID of the processor.
Prime95 http://www.mersenne.org/freesoft/ For stability testing and to load the processor 100%.

As for the cooler, it's decent. Personally I don't like my fans autothrottled and don't connect them on the motherboard. I connect them directly to the PSU cables for 100% speed always.
« Last Edit: 20 Jan 2010, 11:36:30 pm by sunu »

Offline riofl

  • Knight o' The Round Table
  • ***
  • Posts: 240
Re: SETI MB CUDA for Linux
« Reply #581 on: 20 Jan 2010, 11:37:31 pm »
ok ill prep a cd with those things on it.. ill probably do it anyway just to see how this prime prg loads the cpu and now i am also curious what the windows utilities say in addition to my bios.


Offline sunu

  • Alpha Tester
  • Knight who says 'Ni!'
  • ***
  • Posts: 771
Re: SETI MB CUDA for Linux
« Reply #582 on: 20 Jan 2010, 11:49:18 pm »
Ok. Be sure to check how the vcore changes between 0% and 100% cpu utilization. And we'll see if lm_sensors reports the correct VID.

In prime95 you'll need to start 4 threads for your quad core. It has three tests: small FFTs, large FFTs and a mixed mode. Choose small FFTs.

What voltage did you use when you're running at 3.6 GHz?
« Last Edit: 20 Jan 2010, 11:52:25 pm by sunu »

Offline riofl

  • Knight o' The Round Table
  • ***
  • Posts: 240
Re: SETI MB CUDA for Linux
« Reply #583 on: 21 Jan 2010, 05:08:01 am »
Don't bother with windows but if you're going to do it anyway or for future reference:

CPU-Z from http://www.cpuid.com/cpuz.php Among other things it will show the vcore.
Core Temp http://www.alcpu.com/CoreTemp/ Most importantly it shows the VID of the processor.
Prime95 http://www.mersenne.org/freesoft/ For stability testing and to load the processor 100%.

As for the cooler, it's decent. Personally I don't like my fans autothrottled and don't connect them on the motherboard. I connect them directly to the PSU cables for 100% speed always.

i have found that high airflow is not always better. the air can move so fast it reduces the ability of the heatsink/air interface to transfer maximum heat to the air.. for instance if i run my rear fans at 100% the components, especially those under the heatpipe on the mobo actually run hotter. i have found that 80% is a nice starting value that usually is very close. the exceptions are gpu fan/coolers the stock coolers are efficient even at 100% fan which tells me they did not put high enough rpm fans into the cooler.
« Last Edit: 21 Jan 2010, 06:05:39 am by riofl »

Offline riofl

  • Knight o' The Round Table
  • ***
  • Posts: 240
Re: SETI MB CUDA for Linux
« Reply #584 on: 21 Jan 2010, 05:24:58 am »
Ok. Be sure to check how the vcore changes between 0% and 100% cpu utilization. And we'll see if lm_sensors reports the correct VID.

In prime95 you'll need to start 4 threads for your quad core. It has three tests: small FFTs, large FFTs and a mixed mode. Choose small FFTs.

What voltage did you use when you're running at 3.6 GHz?

ok.


if i remember correctly (it has been almost 2 yrs since i did this) i believe i upped the VID to 1.3 v or so. i had found that slight voltage changes had a big difference in stability and nothing gained after a certain voltage. i tried 1.35v and only got more heat as a reward. i think i ran it at 1.28 and it was fine for almost an entire day then went south but at 1.3 it was good.,  i know of one person running the cpu at 4.0ghz with water cooling and i think he is running his  VID around 1.4v i dont remember exactly but i believe it was 1.38v. i do know he is pushing the limit and it is not 100% stable but it is good enough for him. he said he thinks he needs 1.43 or 1.45 but he has reached the safe limit of his cooling system. it is one of the cheaper ones. still that is quite a jump. i am not much of a believer in overclocking as it does not fit with my philosophy of 'if it aint broke dont fix it' and the mfgrs set their specs for a reason.

but since i always really wanted a 3.0 ghz machine but at the time could not afford the processor, and this can safely go to that speed i decided to go against my beliefs so that i had the best of both worlds for myself. :)



 

Welcome, Guest.
Please login or register.
 
 
 
Forgot your password?
Members
Total Members: 97
Latest: ToeBee
New This Month: 0
New This Week: 0
New Today: 0
Stats
Total Posts: 59559
Total Topics: 1672
Most Online Today: 23
Most Online Ever: 983
(20 Jan 2020, 03:17:55 pm)
Users Online
Members: 0
Guests: 17
Total: 17
Powered by EzPortal