Core usage

Hi alI,
i have a questione about cpu usage on my Power Book.
It appears to be using  only odd core so 4 core instead of 8, is it normal ?
Thanks in advance .
MM

update:
Apple technician told me the way Logic uses cores changes depending on it's application.... ugh

Similar Messages

  • Uneven Core usage on dual core processor

    Hi there
    I noticed today whilst doing some video encoding (.TS files to .MP4) that my Mac Mini is not spreading the load over both of the cores in my dual core processor. Whilst Core 1 is showing nearly 100% Core 2 is hardly being used. Also I noticed that when the load on Core 1 dips then Core 2 will peak but as soon as the load goes back to Core 1 it will equally dip on Core 2 so essentially in Activity Monitor it looks like the core usage is a negative image of each other. The total CPU usage never peaked above 54%.
    Could this an indication of a failing CPU or is it a case of poor core management? The software I've been using is VLC Player and also iMagiConverter.
    Cheers
    Jonty

    What I don't understand is why software developers make software (especially video encoding that is all about CPU number crunching) that only utilises 50% of the CPU resources!
    There are any number of possible reasons, from an application using a cross-platform code base that doesn't lend itself well to mulitiprocessor optimization for a particular platform, to inexperienced programmers who don't well understand programming for parallel threads, to developers who just want to crank out an application as quickly as they can (iMagiConverter seems to sort of fall into that category, given the number of applications of a similar nature that the "company" has produced). You'd have to get involved in the project development to know for sure.
    Regards.

  • Manually allocating processor core usage?

    Hi all,
    So I'm not sure if I'm posting this in the right seciton, but here it goes. So my logic board is broken and Activity Moniter shows that kernel_task (the os itself) is using between 500-600% CPU usage (I have 8 cores so this is about 63-75% CPU capacity)
    For reasons I won't get into, I can't fix my computer for about a month. However, there are some higher-performance things I still would like to do. For example, play Dwarf Fortress. Now Dwarf Fortress is not multithreaded, so in theory, if the crazy-high kernel_task usage is restricted to 6 cores, and I can somehow guarentee that 1 core is mostly dedicated to Dwarf Fortress, then I could still play it to a degree. (It's possible that slowdown due thrashing and multi-core issues makes this impossible, but I'm not sure).
    Is there some way I can run the program giving it precedence to one of the cores? Or on the other side, restrict kernel_task to the first 6 cores? I'm pretty sure the program itself has no mechanism for this, so it would have to be something from the mac side. I thought maybe virtualizing a machine and dedicating the cores to that machine might work, but the overhead might make the whole thing slower. I'm a technical user, so any suggestions are welcome.

    kernel_task does not randomly use 600% CPU.  Something is definitely wrong here.
    I don't know that you can assign a "nice" value to kernel_task to de-prioritize its CPU usage.  Some processes you can, but I don't think you can with kernel_task.  Better to find out why it's burning CPU.

  • CPU Core Usage

    Experts,
      What is the optimum core requirements for a SAP server running 7 users ? and How is this derived at? Also the current servers come with dual core, quad core capacities how can I translate this into whats good for SAP Business one v9.0 with the same number of users ?
    Regards,

    Hi Martin,
    First of all the amount of users is not a good measure to determine server hardware requirements.
    A much better measure is the average amount of transactions you expect to enter into the system on a daily basis.
    Very roughly you could count it like this (example of a single sales order):
    Transactions
    One sales order
    1
    The sales order has for example 5 lines
    5
    The corresponding delivery
    1
    The corresponding delivery lines
    5
    The corresponding Invoice
    1
    The corresponding Invoice lines
    5
    Total:
    18
    Next, the CPU has relatively little influence on the performance of your B1 system. Much more important are RAM and hard disks. Please read this article on best practices for the B1 hardware landscape
    Regards,
    Johan

  • Should i Take the more expensive Mac mini (quad core) or dual core

    Hey guys!
    I don't now if I should take the Mac mini with dual core or quad core
    I have never had a Mac so I don't now if the quad core (200$ more) is better than the dual core  :o
    I use it only for standard things (email, web, twitter Mac App Store games)
    But it also should not stuck at bit more powerful apps and run crystal clear !!!
    I want to have the Mac mini for few years !
    Thanks now!

    The difference in real performance between dual core
    and quad core machines is highly dependent on the
    apps one uses and how many at the same time.
    If you really only ever have one or perhaps two apps
    open at once and if these apps are like web browsers
    or email, you may never really notice a difference.
    However, if you have many apps open at once or use
    apps that can utilize many cores at the same time (typically
    imaging and video apps), then there will be a significant
    performance increase.
    As far as gaming, by and large the most significant impact
    on game performance is the graphics processing unit (GPU).
    The integrated Intel HD4000 GPU in the Minis can typically
    handle mid range games but will struggle on the very highend
    games.
    Also, if apps are not designed for mutiple core usage, there will be
    now real improvement between 2 or 4 core machines.  As a matter
    of fact, I have a few apps that are faster on my MBP 13" 2.7 GHz
    2 core i7 than on my quad core i7 2.0 GHz Mini Server for this
    very reason.
    So, the bottom line is that for some a quad core machine will
    blow away a dual core machine while for others, there will be
    no noticable difference.

  • ONE Core Live processing

    I am using my rig mainly for live performance. I am quite aware of the single core usage in record/input enabled tracks, I am to understand for latency reasons? My question, is that if I purchase a quad core laptop, doesn't that mean I will theortically get LESS processing power than a duo core, since the power is cut to a quarter instead of half? (assuming the speed and chip the same)
    Does anyone know what it will take to spread the live performance over multiple cores without a hit to latency? Nodes maybe?
    I have done Evan's benchmark, but is there one for live performance???
    Thanks

    Hi
    danois wrote:
    Ahhhh, now maybe you are on to something that I am unaware. I thought that slaving LP to MS required IAC, which you had suggested earlier could pose potential problems.
    The issues with the IAC are that it is easy to get MIDI feedback loops (at Logic's end and at MS). I have read (not personally experienced) that there can be latency issues too???
    danois wrote:
    Are you saying that there is another way to slave LP to MS with two computers, one running LPX (my old MBP) and another running MS3 (the new MBP)? How? via ethernet? Firewire?
    I was thinking of plain, old-school, MIDI Out from MS via External Track to MIDI interface, cable to MIDI Interface on the "Logic Mac". You could MIDI learn Logic keycommands to operate various functions such as Play/Stop etc, or similar, and simply transmit them from Screen Control Mappings in MS, or use an External Instrument Channel to send MIDI Start/Continue/Stop messages??? Never tried it, but should work.
    You could probably do that all on 1 Mac, provided that you were careful to block MIDI Inputs from going to the wrong application.
    danois wrote:
    In regards to running other third party software to process plugins, just too unstable to work live. Not to mention harder to troubleshoot, etc...
    Vienna Ensemble Pro is pretty rock solid, but does need a dongle.
    CCT

  • Why doesn't Pr use my CPU's cores while rendering/exporting without strange workarounds?

    Hi,
    I'm wondering if anyone can help me understand this. I just finished working on a project that utilized a late 2013 maxed out MacBook Pro, a late 2013 maxed out iMac, and a late 2008 maxed out MacBook Pro all using the latest Mavericks OSX and monitoring with iStat Menus. In each of them, when I rendered or exported a timeline, Pr (and Ae and AME for that matter) would not utilize my CPU's cores until I started opening programs and browsing webpages. And if I stopped doing that, the the CC programs would disengage and the core usage would go down to almost nothing.
    This was frustrating to say the least, but I was under a deadline and couldn't see how I could fix it (especially because it happened on three very different computers).
    Now that I have time, I still can't figure it out. Has anyone else experienced these issues? They are intensely annoying because when I set a project to render or export, I can't be having to baby the program to do its job.

    CPU usage depends on the effects you are using, the bandwidth of your system, and several other factors. When you have a machine with a bunch of fast cores and a bunch of memory the bottle neck that slows things down may become the read write speed of your hard drives.
    You can try different memory and processor allocaions in the preferences file. Here are some resources for the optimum settings.

  • 2012 Mac Pro - balancing cores

    Hi,
    I've been reading up on balancing the core usage in Logic 9 (I have X but won't switch until it's a little more glitch-free) but am running into problems where plugins on busses (or Auxes) drive one core into the red rather than spreading across all 12/24 cores.
    After reading what Apple stated on their page about balancing multi-threaded cores, I still don't seem to have solved the problem! I've done as they've advised (moving 6-8 plugins from 1 Bus or Aux across 2 or 3 daisy chaining them together) but it still just assigns one core to those buses/auxes.
    Am I missing something really obvious here? It's gotten to the stage now where just two UAD plugins are bringing a 2012 Mac Pro (Dual 2.6Ghz 6-core, 24GB RAM, 512GB Mac SSD, 3TB Audio) to it's knees, either on 1 aux, or as shown, across two. There's one other channel (a parallel drum bus) which has around 5 Native plugins on which are being loaded onto the 3rd core (with the 4th core blowing over those two UAD plugins)
    How are people generally sharing the plugin load across cores, as it's getting to the point now of me wondering whether a new 2013 Mac Pro is going to solve the issue, which I'd be gutted about as this Mac is normally such a work horse!
    Ben

    Hi Eriksimon, thanks for your reply!
    I have my Threads set to Automatic, however I had it set to 16 and it made no difference. Sadly LPX is way too buggy and glitch-tastic for me to move over (one example of the repeat regions feature which has *always* worked since version 5 now slowly copies regions out of time... ) even though it's inviting to think that it'll use all the power I have!
    I wasn't aware of the refresh thing upon reopening sessions, but I was working on something today I'd DC'd three weeks ago and again, it fell over, each time at the same place. It's so frustrating!
    Buffer size is 1024 samples, PBR is set to Large
    Thanks again!
    B

  • Multi-core Compiling

    Even though ibberger touched the concept in the idea , I do think that most o people uses LabVIEW under Windows environment. Compiling a FPGA VI happens all in the PC under Windows. I noticed that during this process the compiler uses only one core. Since I'm using a machine with a 4 core processor, the CPU use rarely goes above 25%.  My idea is to update the compiler allowing it to be multicore. The user should have the option to limit the maximum number of cores available to the compiler. This is necessary because the user may want to continue working, while the compiling process is being done in background.

    Here's shortcuts for speeding up single FPGA compiles on both Windows 7 and XP for i7 and i5 systems with Turbo Boost. This only applies to a workstation, not a compile farm.  In my case I was able to reduce the compile time 40-50%. It will at least shave several minutes off the compile.
    Make sure Turbo Boost is enabled in the BIOS if the option is there.
    Windows 7:
    Find the Compile Worker shortcut in the start menu and copy it to the desktop.
    Modify the compile worker target command to add CMD and START commands (spaces and quotes are important):
    C:\Windows\System32\cmd.exe /C START " " /HIGH /AFFINITY 0X01   "C:\Program Files (x86)\National Instruments\FPGA\CompileWorker\CompileWorker.exe"
    Windows XP:
    Download PsExec from the Microsoft website. It will be part of a suite of commands. I used version 1.98. Unzip the directory, then copy PsExec to C:\Windows\System32.
    As above, find the Compile Worker shortcut in the start menu and copy it to the desktop.
    Modify the compile worker target to add the PsExec command:
    C:\WINDOWS\System32\PsExec -high -a 0 -d  "C:\Program Files\National Instruments\FPGA\CompileWorker\CompileWorker.exe"
    If your original Compile Worker command is different, use that. Make sure it is enclosed in quotes.
    To use: double click on the modified shortcut to start the Compile Worker. It will start and remain dormant in the background. Open a project and perform an FPGA build. The compile will be queued to the Compile Worker as normal. CPU usage can be monitored in Task Manager. As the compile progresses, the first core usage should eventually peg at 100%. The synthesis phase should be noticeably faster.
    For Windows 7, Intel has a Turbo Boost monitor tool.
    Why it works: The compiler seems to benefit from higher CPU clock speeds rather than more cores. The scheduler tends to move the process among the different CPU cores, and so the CPU isn't loaded down enough for Turbo Boost to stay on. The above commands force the compile to stay on one core, encouraging Turbo Boost to stay on for the duration of the compile.

  • I7 Question - Hyperthreading virtual cores

    So I have a new macbook pro with the 2.66 GHz Intel Core i7 processor and I have a question. I don't consider myself crazy knowledgable with Macs so I'm not entirely sure what's going on
    The guys behind the genius bar told me there was a way to push my processor to do 8 virtual cores instead of 2 using the developer tools. I don't really know my way around the code so I have no idea what does what. I've been looking through forums and such and I can only find people talking about disabling hyper-threading.
    I went into the CPUpalette and only see an option for disabling the already existing extra 2 cores.
    I ran the Mac OS X install DVD and installed the developer tools.
    If anyone could guide me through activating the extra virtual cores, that'd be sweet. Or just tell me if it isn't possible to do more than what's already activated.

    I'm not familiar with the hardware, but the i7 supports what is called "Turbo Mode." As I understand this mode is automatic. Depending upon core usage elsewhere processor turbo charging will occur as needed. Thus, it's possible to have all 8 v-cores running at normal speed when fully loaded or fewer threads at higher speeds.
    The Developer Tools contains a preference pane in the Extras folder that you might install, but I do not believe that "Turbo Mode" is user controllable since it's always turned on.

  • Problems with P965 Neo and E4500 Core 2 Duo

    Hope someone can help.
    I have a gadget for Vista that shows core usage. The second core does not register at all when running a performance test. Also, when I look under "My Computer" at the Properties it shows:
    Processor:    Intel(R) Core(TM)2 Duo CPU   E4500 @ 2.20GHz  1.10GHz
    Also, my Windows Experience Index for my processor is only 4.3, seems low for the processor.
    Makes me wonder if the second core is toast? Why would it show 1.10GHz? Looking at my BIOS version, I believe it's at 1.5 (I've never updated it since it came out of the package) but the Live Update Online utility says there are no newer versions available. I tried to download the BIOS update off MSI's website, but when I click on the .exe file, it flashes and then goes away. So fast, that I can't even read or see much of anything. The instructions that were included do not cover Vista, and say I have to get a 98/XP boot disk??? When has that ever been necessary for a BIOS update?
    Anyway, any help would be appreciated.

    Hello and Welcome to the Forum!
    BIOS-Updates
    BIOS Updates should be done in pure DOS mode.  Flashing from a Windows Environment is very risky and should be avoided if possible.  Flash your BIOS only if you are aware of the risks of BIOS Flashing.  Updating the BIOS should be the last resort in troubleshooting.  If you are sure you want to flash a different BIOS use the Forum's USB-Flashing.  It is easy to use and prepares the Flash Drive automatically for a DOS-Flash.  You can find it here:
    https://forum-en.msi.com/index.php?topic=108079.0
    Quote
    The second core does not register at all
    Is your CPU properly recognized in BIOS and during POST?
    In BIOS, make sure the following options are properly set:
    ACPI = Enabled
    (IO)APIC = Enabled
    MPS Table Version = 1.4
    If you are sure these options are as shown above, force Vista to re-detect proper HAL during startup.  You have to find Vista's Advanced Boot Options dialog box in order to do that:

  • Ideas for lastest chipsets and computing hardware

    i have many ideas for silicone based builds and i would like input from everyone if you wish.
    discrete graphics are great and i believe that both Intel and AMD have made great recent leaps in this architecture and i give them both respect.
    i believe low cas is the way to go for memory to assist in optimal multi-core usage and rendering capability.
    3-d technology looks amazing to me although i have yet to try it out myself as well as a lot of the newest touch screens.
    i would like to see 4-8 system threads with high physics per thread and as low of frequency as possible for less power draw after all if we are going to be innovative we should be energy efficient at the same time.
    my point of view is from a gamers and over clocking perspective with little understanding of the hard work that goes into the programming and server end of the computing industry.

    In before Mike!
    Unfortunately your post is off topic here, in the TechNet Site Feedback forum, because it is not Feedback about the TechNet Website or Subscription.  This is a standard response I’ve written up in advance to help many people (thousands, really.)
    who post their question in this forum in error, but please don’t ignore it.  The links I share below I’ve collected to help you get right where you need to go with your issue.
    For technical issues with Microsoft products that you would run into as an
    end user of those products, one great source of info and help is
    http://answers.microsoft.com, which has sections for Windows, Hotmail, Office, IE, and other products. Office related forums are also here:
    http://office.microsoft.com/en-us/support/contact-us-FX103894077.aspx
    For Technical issues with Microsoft products that you might have as an
    IT professional (like technical installation issues, or other IT issues), you should head to the TechNet Discussion forums at
    http://social.technet.microsoft.com/forums/en-us, and search for your product name.
    For issues with products you might have as a Developer (like how to talk to APIs, what version of software do what, or other developer issues), you should head to the MSDN discussion forums at
    http://social.msdn.microsoft.com/forums/en-us, and search for your product or issue.
    If you’re asking a question particularly about one of the Microsoft Dynamics products, a great place to start is here:
    http://community.dynamics.com/
    If you really think your issue is related to the subscription or the TechNet Website, and I screwed up, I apologize!  Please repost your question to the discussion forum and include much more detail about your problem, that could include screenshots
    of the issue (do not include subscription information or product keys in your screenshots!), and/or links to the problem you’re seeing. 
    If you really had no idea where to post this question but you still posted it here, you still shouldn’t have because we have a forum just for you!  It’s called the Where is the forum for…? forum and it’s here:
    http://social.msdn.microsoft.com/forums/en-us/whatforum/
    Moving to off topic. 
    Thanks, Mike
    MSDN and TechNet Subscriptions Support

  • Logic GUI not in sync with audio

    A freshly-installed Logic 8 on a clean Mac Pro 2.26 (x8) results in badly synched graphics.
    The meters in all of my plug-ins do not sync up with the audio -- doesn't matter which. More specifically, the graphics are coming before the corresponding audio.
    This occurs no matter what the latency sample value is set at -- 128 or 1024.
    This is definitely a Logic problem, not a plug-in problem, because when I duplicate the settings in GarageBand, the graphics are in sync. Even a basic one audio track with a 2-bar drum loop isn't synched.
    Any suggestions?

    Todd Schnitzer wrote:
    The sync is messed up no matter how many plug-ins are running, UAD or otherwise.
    I tried restarting, quitting and starting again, different sessions, etc, to no avail. Hopefully, Logic 8.0.3 (or 8.1) will address this and other issues (64-bit, multi-core usage, etc).
    if you have uad plugin on the master or in the AuX the GUI if the track will not in sync.. but the meter of the master out is the only one in sync
    This is the same on any system!
    Is not related with mac pro 2.26 x8!
    G

  • Screen Flicker When Using Slide Bars in Develop Module

    Image below shows 6 core usage during edit of an image using LR 4.2RC. Image import was running in the background.  This only occurs when I use the develop module sliders (it doesn't matter which slider). 

    At this point, I think I will have to go back to LR 3.x.  Lightroom 4 is so unstable that I cannot get it to work.  I constantly get not responding messages and screen flickering in the develop module.
    I have done everything I can find and nothing seems to work.  It is either an incompatibility with the AMD video drivers (best guess) or there is something seriously wrong with the code (second guess), however LR 3.2 does not have this issue.
    Lightroom version: 4.2 RC [844780]
    Operating system: Windows 7 Home Premium Edition
    Version: 6.1 [7601]
    Application architecture: x64
    System architecture: x64
    Physical processor count: 6
    Processor speed: 2.8 GHz
    Built-in memory: 16381.5 MB
    Real memory available to Lightroom: 16381.5 MB
    Real memory used by Lightroom: 603.5 MB (3.6%)
    Virtual memory used by Lightroom: 621.6 MB
    Memory cache size: 191.8 MB
    System DPI setting: 96 DPI
    Desktop composition enabled: Yes
    Displays: 1) 1920x1080, 2) 1280x1024

  • Compressor 4.1 Crop and Size appears to be broken

    anyone who can check this, much appreciated before I send this in as a bug
    try to make a conversion..set any crop and then force the size back to it's original size and start job and it should fail telling you missing codec.
    seems to be happening with anything but uncompressed files...it was fine in 4-4.07
    this just seems like a really bad rethinking of the whole application. it wasn't made 64 bit and the interface changes shouldn't be breaking basic functionality (crop and size, multi core usage etc)

    Thanks Russ
    Can you try these steps?
    1 your original xdcam
    2. set a crop of 8 to top and bottom
    manually set size to original (it will adjust down with e crop, so enter the size manually)
    submit and let me know if it completes.
    if it's working, then it look like it's prores specific (which is bad if you're doing roundtrip from fcp)

Maybe you are looking for