Intel Communities : All Content - Processors

DVFS capabilities in latest Intel processors

webadmin@intel.com — Fri, 10 May 2013 00:40:49 GMT

Hi,

I would like to know whether we can change the frequency of the intel multi-core processors individually. I have Intel Core2Duo CPU T5750 which has 2 cores. When I run cpufreq-info in Linux I get to know that both the cores will run in the same hardware frequency. Please find cpufreq-info output below(Relevant data shown)

$ cpufreq-info
*******************
analyzing CPU 0:
driver: acpi-cpufreq
CPUs which run at the same hardware frequency: 0 1
CPUs which need to have their frequency coordinated by software: 0
maximum transition latency: 10.0 us.
hardware limits: 1000 MHz - 2.00 GHz
available frequency steps: 2.00 GHz, 1.67 GHz, 1.33 GHz, 1000 MHz
********************

********************
analyzing CPU 1:
driver: acpi-cpufreq
CPUs which run at the same hardware frequency: 0 1
CPUs which need to have their frequency coordinated by software: 1
maximum transition latency: 10.0 us.
hardware limits: 1000 MHz - 2.00 GHz
available frequency steps: 2.00 GHz, 1.67 GHz, 1.33 GHz, 1000 MHz

Hence by above output I expect both cores to vary its frequency in a synchronous fashion.

But when I tried to check the frequency of each core by using the information in /proc/cpuinfo, I see that the cores are varying its frequency individually. The command I used and its output are below:

$ while true; do grep -e "cpu MHz" /proc/cpuinfo; echo "===";sleep 0.1; done

cpu MHz : 1000.000
cpu MHz : 1000.000
===
cpu MHz : 2000.000
cpu MHz : 1000.000
===
cpu MHz : 1000.000
cpu MHz : 1000.000
===
cpu MHz : 1000.000
cpu MHz : 1000.000
===
cpu MHz : 1000.000
cpu MHz : 2000.000
===
cpu MHz : 1000.000
cpu MHz : 1000.000
===
******************

******************

Hence my questions are:

1) Can someone explain why the cores change the frequency individually even though the cpurfreq-info says that both the cores run at the same hardware frequency? Am I missing something?

2) Does latest intel multi-core chips have per-core DVFS? Kindly let me know.

3) Can voltage and frequency can be changed independent of each other in latest multicore processors using EIST (Enhanced Intel Speed Step Technology)?

Kindly let me know your comments...

Thanks for your time,

Karthik.V.M.

Note:

The cpu frequency can be more neatly obtained using the following command:

$ watch -n 0.1 grep \"cpu MHz\" /proc/cpuinfo

i7 3770K temp

webadmin@intel.com — Wed, 22 May 2013 19:11:15 GMT

I use "thermaltake water 2.0 extreme" water cool kit. But I think I can't take full performance

Room Temp :28 °C

CPU İdle Temp :32 °C

Mainboard : DZ77RE-75K

CPU : i7 3770K (Stock Speed)

PSU : Corsair TX650W

Case : Corsair R500

Thermal compund : Arctic Silver 5

Prime95 5 minutes test results

CPU load temp : 70 °C

Is everything alright?

Core 2 Duo T9300/T9500 processor compatibility

webadmin@intel.com — Wed, 22 May 2013 10:31:16 GMT

I'm new here in the community, so I ask for your understanding and help, if I'm not asking correctly, ok ?

Well, I would like to know, if Core 2 Duo T9300 or T9500 processors, are compatible with an Asus K50IJ laptop. I intended to replace your current Pentium Dual Core T4300, by one T9300 or T9500 ( that I have mentioned above ), to make it faster. I believe that they are compatible, because they have the same socket and FSB, but I do not know if BIOS is compatible. Can you give me your opinion ? My operating system : Windows 7 64 bit, 4 GB DDR2/800.
Thank you, and sorry for my bad english

Fonseca

Why is there 1.65v RAM on Intel's own Memory compatibility spreadsheet.

webadmin@intel.com — Wed, 22 May 2013 09:52:20 GMT

If Intel processors don't support more than 1.5V memory for SB and IB why are there 1.6V and 1.65V RAM kits listed in the Compatible Memory for 1st, 2nd and 3rd Gen Intel Core i5 and i7 processor spreadsheets?

Please help

webadmin@intel.com — Fri, 10 May 2013 18:43:11 GMT

Hi Intel

I have bought Intel Xeon E3-1220 v2 processor.

I am testing Linux PREEMPT-RT on this processor. I created a test to see the time it takes for the OS to handle the clock tick.

My test is as following: I create a thread with highest priority ( can be preempted only by interrupts), whose task is just to do a small calculation in a loop of 45000 times. This task takes 102 µs to be finished.

If a clock tick happens while the task is executing, then this task is interrupted for some time until the interrupt is handled. ( aka: in normal case, the task takes 112 µs and if a tick happens, then the task takes 120 µs, which means that 8 µs are spent by the OS to handle the clock tick interrupt). I did this procedure around 100 times and I saw that each 1 ms I got a clock interrupt as the OS clock tick is running at 1000 Hz. so the normal values should be 102 µs if no tick happens, and 110µs if a clock tick happens.

This was perfect test on Intel core i7-930 @2.8GHz.

If I do the test on processor Xeon E3-1220 v2 @ 3.1 GHz, the task takes shorter time ( 102 µs) as it is faster processor 110 µs if a tick happens.

The strange thing I see on this processor that sometimes I get values around 120 µs, which means that some kind of interrupt happened and preempted the task for 18 µs. If it was clock tick, then it should take 110 µs.

so the processor is generating some kind of interrupts.

If I do the test with loop of 5000 instead of 45000, I do not see anymore such strange values.

Synthesis: if I do a task which is looping 45000 times on intel xeon, I start to get strange interrupts, which is not the case on i7-930.

If I do the test with loop of 5000, the interrupts disappear.

what could be the issue?

Regards

Rendering Workstation

webadmin@intel.com — Tue, 14 May 2013 08:59:03 GMT

Hi all

This is my first post so hope it will be answered, i'll try to make a simple one. Basicly I do renderings of an interior/exterior scene using Win7 64 bit and hp z800 with 2x Xeon E5530 @ 2.40 GHz cpu with NVidia Quadro 3800FX graphics card, and 16gb ram. With 3DS Max Design 2014, Vray2.4 addin it takes up to 10hours of calculation for a complex scene in HQ at 3000x3000 pixels. What I want is to use Dual Xeon E5 cpu with 32 or more ram with a gtx titan or Quadro 4000k graphic card and fasten this process. Has any one got any idea or better suggestions for this issue, or anyone uses a computer like this or better one? Any idea about Xeon phi coprocessor can be a better solution?!? It would be work in a workstation to improve speed in the rendering process?!?!?

Tnx.

Salve a tutti,

questo è il mio primo post e spero che qualcuno possa rispondermi, quindi cercherò di spiegarmi bene. Di media sviluppo renderings di interni ed esterni utilizzando una workstation HP Z800 con sistema operativo Win7 a 64 bit equipaggiato con: 2 processori Xeon E5530 @ 2.40 GHz, scheda video NVidia Quadro 3800FX, 16Gb di ram. Il software di sviluppo che utilizzo è 3DS Max Design 2014, Vray2.4 e per il calcolo di una scena complessa ad alta qualità e risoluzione a 3000x3000 pixels impega di media 10 ore di calcolo. Vorrei passare ad una nuova WS equipaggiata con doppi processore della serie Xeon E5, con 32 o più gb di ram, con scheda grafica GTX Titan o ancor meglio una Quadro 4000k per aumentare la velocità di sviluppo. Qualcuno ha una migliore soluzione o consiglio? Affiancare un coprocessore Xeon Phi potrebbe esser una migliore soluzione?!? Funzionerebbe installato su di una WS per diminuire i tempi di calcolo relativi al processo di render ( CPU based ) ?!?!?

Grazie.

q6600 testing

webadmin@intel.com — Fri, 17 May 2013 11:31:19 GMT

i am having random reboots and i have been trying to cover all my bases my 4 core temps without any real load sit in the 90F-100F range i have within he last 6 months gotten a corsair h60 on sale and i cant tell if thats doing anything but thats another matter .... what i need to know is how to test and understand the results of whatever tests i can dso to see if my processor maybe the cause if trhe restarts .... yesterday i got a new psu from 520w to 750w and still the same i run hwmonitor constantly to watch temp on everything and such i used to also watch the windows performance readings those i understood basically but this may all be getting out of my realm ....

basically i need to cross off things that are not the cause ... what i am presently running with is

Motherboard XFX Nforce 680i LT (Socket 775)

Processor Intel Core 2 Quad Q6600 @ 2.40GHz

Optical Drive PIONEER DVD-RW DVR-215D SCS

HD1 WDC WD50 00AAKS-00A7B SCSI

HD2 WDC WD20 EURS-63S48Y0 SCSI

RAM 4.00GB Dual-Channel DDR2 @ 399MHz (6-6-6-18)

Graphics Card XFX Nforce 680i LT (Socket 775)

and the current read outs are as follows during the time writing this and just speccy and HW monitor open plus webpages

core 0 96F

core 1 100F

core 2 93F

core 3 98F

graphics card 143F

HD 1 118F

HD 2 111F

cputin 109F

auxtin 116F

to be clear i will also be doing more research on the graphics card as well i have been getting these exact cards for like 100 a piece and this is the second after my original they seem to last about a year but i think that might have something to do with the underlined issue here and this random rebooting is a new addition to the same possibly

if anyone has any other information that could be of use in any regard of any of this i would be grateful

processor E5-2620 supports RDRAND?

webadmin@intel.com — Fri, 17 May 2013 17:59:19 GMT

Hello

I have a server platform Intel R1208GZ4GC and E5-2620 processors. I'm trying to use RDRAND, to generate random numbers. To put this code:

int _rdrand16_step(unsigned short *therand) { 
 unsigned char err; 
 asm volatile("rdrand %0 ; setc %1" 
 : "=r" (*therand), "=qm" (err)); 
 return (int) err; 
  }

However, when compiling the error:

{standard input}: Assembler messages:
{standard input}:7191: Error: no such instruction: `rdrand %dx'
gmake: *** [../../main/main.o] Error 1

Operating system FreeBsd 8.3

Help me, please.

Xeon E5630 - Sees Only 16 "CPUs"

webadmin@intel.com — Thu, 16 May 2013 20:54:08 GMT

Gentlemen,

My client has three stations "PowerEdge T610" with 2 processors E5630 (in bios) (Windows E5649). My XenServer (Bare Metal Hypervisor) sees only 16 "CPUs". These 2 processors have 4 cores each and 8 threads for each core.

The problem is that XenServer sees only 16 "CPUs" when he should see 2x4x8 (CPUsxCoresxThreads) 64 "CPUs". Namely, 32 "CPUs" for each processor. Is that correct?

I've Recently upgraded this bios. What could be wrong? This is a limitation of Linux? What can I do for this my client?

Thanks!

Atom N2600 and Memory Timings

webadmin@intel.com — Mon, 13 May 2013 17:07:36 GMT

Hi all,

A few months ago, I bought a 2 GB stick of G.Skill RAM to put in my Gateway LT4004u netbook. It's rated at DDR3/1600 CL9. As I understand it, my Atom N2600 underclocks the RAM to 800 MHz. I was hoping to get CL5 at that speed out of the RAM, but when I started up CPU-Z 1.64 to check, all of the fields were blank. The program can read the SPD of the stick just fine, but it doesn't report any values in the Memory tab (other than its size). Is there any way to see what timings my RAM is running, and even better, change them if needed?

Thanks in advance!