Posts by rubberman

I think they are an exception rather than the rule at the moment though.

Possibly. Myself, I depend upon my systems for my business (IT consulting), so except for my laptop, all my systems utilize ECC memory, and I continuously monitor them for RAM/CPU usage, temperature, drive utilization, etc, with alarms to notify me when things start to get "wonky", such as memory utilization starts to hit the swap space. Since I run a variety of operating systems in virtual machines (Windows, Solaris, QNX, Linux) and an Android emulator (not all at the same time), this can help me sort out any number of strange issues.

Mine are fully buffered ECC RAM (Crucial brand) and the motherboard is an Intel S5000XVN workstation/server board. Custom built with dual 3GHz E5450 processors, 8GB (4x2GB) RAM, and 2.5TB of disc cost me about $5000USD when I built it 3 1/2 years ago.

I run Scientific Linux 6 (a clone of RHEL 6) and it has full support for system thermal sensor monitoring (lm_sensors), so I can monitor both RAM and CPU temperatures. Under load one RAM stick would exceed limits, resulting in the OS mapping that RAM out of the system. By moving the sticks I was able to reduce the heat by over 25 degrees centigrade under load - such as when I would run all 8 cores to build a new kernel in 15 minutes rather than 2 hours.

I've had regular RAM sticks overheat on me where use of a heat sink would probably help reduce that. I fixed the problem by moving the sticks to get better airflow, but a sink would probably have worked as well, or better.

Have you considered that possibly the CPU is toast? I would suggest that you contact Gigabyte, the board manufacturer, about sending it in for repair.

You should be able to find the driver on the Toshiba web site, probably under "Support" for the model of laptop you have.

Usually these pretty much just clip on, or use a friction fit. The stuff you found on the sink and chips that they were attached to was probably a thermal transfer paste or tape that helps draw the heat from the chips to the sink where it can be removed by normal heat convection and radiation. You can get this stuff (the thermal paste and tape) from most online electronics stores such as buy.com, newegg, et al. Anyway, don't superglue the sinks to the chips or memory stick! It won't help the thermal transfer characteristics at all, and will make it pretty much impossible to remove!

Anyway, here is a link to one such product on buy.com: http://www.buy.com/prod/startech-com-metal-oxide-thermal-cpu-paste-compound-tube-0-0060-c-w/q/loc/101/10311463.html

Try the Windows version of the open source AV program ClamAV, called ClamWin. ClamAV is used as the core for a number of commercial business-use AV appliances. You can get clamwin here: http://www.clamwin.com

Sometimes, newer is better! :-) Glad that the new card did the trick.

You should be able to install the OS without networking. Is there that option in the installation screens?

Download the driver to a thumb drive. Install the OS and boot into text mode. Log in, connect/mount thumb drive. Install driver. Reboot.

Here is the driver support page for this RALINK device: http://www.realtek.com/downloads/downloadsview.aspx?langid=1&pnid=13&pfid=5&level=5&conn=4&downtypeid=3&getdown=false

Thanks for the link. Is this running the i7 or the i5 CPU?

I'm not familiar with that board. Do you have some info about it, or link to manufacturer/vendor web site with specifications, etc?

Overheating RAM is not an uncommon problem. You need to monitor the temperature, if you can. I don't know about Windows Vista capabilities in that regard. I use a free version of Red Hat Enterprise Linux (Scientific Linux) 6 that has full support for system thermal sensors. Using that, I found that under load one of my memory sticks was overheating and subsequently taken out of service by the operating system. By re-arranging how I had them plugged into the motherboard (getting better airflow), I was able to reduce the temperature by 25-30C, eliminating the need (and co$t) to replace the RAM.

Myself I go with Red Hat Enterprise Linux 6 clones. The best one currently (IMHO) is Scientific Linux 6, maintained by the computer boffins at Fermi National Laboratory and CERN. Check here: www.scientificlinux.org

Login to godaddy.com, click on the "domains" tab. click on your domain name entry. Under the "DNS Manager" section, click on the "launch" link.

Some current trojans are very hard to eliminate because they infect the boot sector of the disc. I have also seen some infect the restore partition libraries, so even if you reinstall the operating system, it will still be infected. This is why I recommend using Linux, and run Windows inside a virtual machine if you must have Windows software.

Virus... Time to clean your system, or reinstall it...

Most any nVidia or AMD video card will run well in both 32-bit or 64-bit environments. Myself, I like nVidia, but that's just my personal preference. I run them on my workstation and laptop in both 32-bit and 64-bit operating systems without problems. As for sub-$100 cards, there are a number of them that are very good in the performance category. Spending more is not necessary to get good game or video performance, unless you are into really high-end software. My old 8800GT nVidia board handles dual HD flat panel displays and can provide full-screen, full-motion video without stutter on both displays at the same time. An equivalent (or better) board today will cost you about $60-$70 USD.

Try a Google search on the term. There is quite a bit of information about this if you would just look for yourself... :-(

Use public get methods to access the member variables. Leave the variables themselves private in order to protect them from outside modification, but with a get methods, such as "const member_type getMemberName() const", you can access the variable data safely, and easily.

Well, some folks work for the Department of Procrastination, which is situated next door to the Department of Redundancy Department... :-)

I would agree with Mike...7 that Kdevelop is one of the more (or less) interfering IDE's. I still prefer to stay "primitive"... :-)

2GB of RAM to run Vista plus the Android emulator is not enough, unless you have a big swap file, and then it will run like molasses. You need at least 4GB for this code combination, IMO.

Personally, IDE's generally get in my way. I prefer a good programming editor, Makefile, debugger user. Each can run in its own window and don't interfere with each other. I have used a lot of different IDE's including Visual Studio, Eclipse, Qt-developer, and others (some that no longer exist). All have their good and bad points, and all eventually get in the way of software development. The only real GUI-based development tool that I currently use is Sparx Enterprise Architect, a full-featured UML modeling and model-driven-design tool that can reverse engineer source code to generate UML models of existing code, or generate code from models. I spend most of my development time there, then generate code - build, test, fix, and then pull the code changes back into the model for further analysis and simulation.

JNI is used to access C/C++ and other machine code components from Java, not the other way around. That said, I think that there are ways to access Java code from C/C++ applications, but it has been a lonnnng time since I read anything about that. Anyway, my Google search on "calling java code from C" provided some interesting links. Here are a few:

http://java.sys-con.com/node/45840
http://www.codeproject.com/KB/cpp/CJniJava.aspx
http://answers.google.com/answers/threadview/id/730502.html

I was having a similar problem on my small office network, and it turned out to be a switch that was starting to fail. Interestingly enough, the errors were all on received packets. Outgoing ones were fine, which is what made this difficult to diagnose. Replacing the switch fixed the problem.

Unless you're doing this to learn about the encodings, I'd recommend using a library that does it for you. I usually recommend ICU.

Also, since it is an open source library, you should be able to get the source code for the code page conversion routines and inspect/analyze them in order to understand what the algorithms are that they use. There may also be documentation on those, but I didn't look into the site far enough to determine that. Good reference Narue! Thanks! :-)

you rubber man,you have switched the threads i guess lol.anyway,
what are the maximum heat for components? RAM,Hard,CPU,Chip and GPU ?
thanks

Well, the thread was about whether to leave a system on, or shut it off more or less frequently. I thought some info about monitoring system health would be in order so people can determine whether or not running continuously may or may not be a good idea. If after leaving it one for an extended period and things start to overheat, then shut it down. If things "keep their cool", then don't bother. Starting/stopping drives is harder on them than running continuously, unless they are overheating because of inadequate air flow.

As for proper temperatures, it depends. My cores run between 20C and 50C or so. My RAM runs between 60C and 80C. My drives vary, depending upon where they are installed and what make/model they are. My 1.5TB Seagate drives run hot, up to about about 44C in a vertical dock (ambient air flow - no fan). They start to complain above 45C. My internal 500GB Seagate drives run about 25C. My 2TB WD drives in a RAID enclosure run about 32C. And my 1.5TB Seagate drives in a similar enclosure run about 40C or so.

Modern operating systems can monitor a number of components on the computer that have thermal sensors. For example, on my Scientific Linux 6 (a clone of Red Hat Enterprise Linux 6) system, I can monitor the temperature of each CPU core (all 8), each memory stick, and all my hard drives. The CPUs and RAM have hardware sensors that the OS can access, and the disc drives all have what is called SMART technology on the disc controllers that monitor a lot of items such as temperature, number of bad sectors that have been remapped with good ones, number of read/write errors, number of times the drive has been started/stopped, how long it has been powered on (total), etc. With the appropriate monitoring software, your system can tell you exactly what is going on. For example, I had an overheating memory stick last year that would start to flake out under load, such as when I would build a new kernel. I was able to fix that by changing the slots the memory was in to get better air flow, and the problem has gone away since I was able to reduce the temperature by about 25 degrees C under load. It would start to fail at 105C, and now I can keep it around 80C under load. Without the monitoring software to see exactly what was going on, I would not have known what to do, other than replace all the memory (8GB in 4 sticks).

As for …

Have you tried running chkdsk on the volume? Or does Windows not see the partitions? What does it look like when you run the disk manager tool?

Another question, if I do get a new graphics card, what's most important, memory, speed, power usage or what & if it's important I am not a gamer, I process huge video files, frequently a frame at a time, video editing.

The 430GT card should do nicely for that. Memory would be the biggest boost here. If you are doing a lot of rendering, then you might want to go with a higher performance board, but one of those would cost quite a bit more, up to $500 or more.

Duhhhh, why didn't I think of that, must be that Al's hammer kickin' in!!! Anyhow, super thanks!!!11 Any thoughts on the explorer.exe spikin' when the display stalls?

Difficult to say. There are too many factors to account for. Possibly a retry loop or something of the sort. Not having explorer's source to analyze, that is just a SWAG.

Thanks for clarifying that. As for nVidia cards, there are a bunch of good ones for $100USD or less. You might want to visit the nVidia web site to see what they have these days. My 8800GT card is getting a bit old, but performs very well and was under $100 when I bought it over 3 years ago. The 430GT card seems pretty equivalent in performance, but with more memory than the 8800GT, and pricing it on the nvidia site shows prices in the $60-$70 range. It is also pretty power efficient, drawing a max 49 watts.

Pseudo code is just a step by step description of how to perform a computation, along with appropriate explanations of what each step does to fulfill the requirements. There is no magic here, just work so get started!

Sum: for (i = 1; i <= n; i++)
{
// compute value for i
// add value to total
}

Personally, I doubt it is the GPU that is at issue here. How much RAM do you have? If your complex processing needs have consumed all your physical RAM and it is hitting the swap file, then this sort of problem becomes common. So, monitor your system performance and memory usage before you make any assumptions about the video board. I would expect other indications if that were the problem, such as a corrupted display.

The problem is possibly one of overheating. Make sure there is adequate air flow over the motherboard and RAM. Used parts are not advisable for a new PC. You don't know what sort of stress they have experienced in the past. Note that this can also be a power supply problem.

The original post wasn't clear as to whether they were talking about laptops or desktop/server systems. Laptops are generally configured to sleep/hibernate when they are on battery power after some period in order to conserve battery. Workstations can also be configured to hibernate similarly, or on demand, and that does allow you to restart the system in the same context that it was in when hibernated. It does shut down the machine, but saves the context in a hibernation file, and when restarted, reads the hibernation file and restores the system to that context. Sleep is different in that the machine is still powered up enough to keep memory, stack, etc. running enough to save the system context. On a laptop on battery power, the battery will eventually run down, unlike hibernation which actually physically shuts off the system.

Get and read Andrew Tanenbaum's Minix book for a good guide to building a minimal OS. FWIW, that was (I believe) where Linus Torvalds started when he built Linux.

http://en.wikipedia.org/wiki/Andrew_S._Tanenbaum

I think I understand what you are trying to get at. The T1...T4 elements are threads (label axis clearly please), and 1...28 are the sequence of steps that occur. With that clarified, we can move forward.

One observation - T2 unlocks D3 at step 21 after writing it at step 7, and then commits finally at step 22. This would not occur in real life unless you were rolling back the transaction, so get rid of step 21.

S1 - t1.d1.lock()
S2 - t1.d1.read()
s3 - t2.d3.lock()
s4 - t2.d3.read()
s5 - t4.d3.lock()
s6 - t4.d3.read() At this point t4 has read consistency established on d3. Any changes by t2 or others will not be visible until it relinquishes the lock.
s7 - t2.d3.write() T2 now has a write (exclusive) lock on d3
s8 - t3.d2.lock()
s9 - t3.d2.read()
s10 - t3.d2.unlock() No other locks exist on d2 at this point, so d2 is available
s11 - t1.d2.lock()
s12 - t1.d2.read()
s13 - t2.d2.unlock() See s10.
S14 - t4.d1.read()
S15 - t4.d4.read()
S16 - t4.d4.write() This should establish a lock on d4, implicitly.
S17 - t1.d1.write()
S18 - t2.D4.lock() This will queue t2 on D4 until t4 commits or rolls back. As a result T2's steps 19-22 will also wait until T4 commits. Likewise T3's steps 24-27 will queue. T2 and T3 are now gated on T4's actions and will wait …

You might be able to find some diagnostics software on the manufacturer's web site that can determine if there are component-level problems. The problem probably isn't the hdd, but may be some other IC on the motherboard, or other peripheral chip.

I see no closing brace in main(). Also, post your code inside code blocks and indent appropriately, please. It makes it a lot easier to read, and fix.

The putw() function writes the raw integer data to the output stream. If you want it viewable by human eyes, then use printf("%d ", num) instead. Also, you might find this reads better:

for (num = getw(fp); num != EOF; num = getw(fp))
{
    printf("%d ", num);
}

Run diagnostics on the system. It sounds like some component(s) is starting to fail. Since it is likely out of warranty, you may be on the hook for repair/replacement.

It is best to leave workstations and servers running, unless you are going to leave them just idling for extended periods of time. IE; turn off the workstation when going away for the weekend. Servers should probably not be powered down except for maintenance. In any case, power-up is the most stressful on system components. If you do leave your system on for extended periods of time, monitor component temperature. I left my workstation on when going away one weekend, and when I got back I found that the fan on one of my external disc arrays had failed, cooking the drives and my data as a result... sigh. Since I keep temperature monitoring software running, I would have noticed were I around that the drives were getting "toasty" before they failed, and could have fixed the situation. So the lesson there is - going away for more than a day, then shut down the system unless I need to access it remotely (via VPN).

If you open a USB port and get a file descriptor, that can be used just like a socket in select() calls.

Use the select() call to indicate when there is data ready to be read on a socket/file-descriptor.

Windows does support the POSIX interfaces, but writing code that works equally on Windows as on Unix is not simple, especially if you want to write GUI applications. In such a case, you would be best served by using a platform neutral SDK such as Qt. That provides a means to write C++ code that will run on any number of different operating systems without recoding - just recompilation.