Posts by rubberman

Most of the time, when you install Linux, it already has the drivers you need for your hardware, including video. Linux is NOT Windows! It will interrogate the system on boot and enable the needed drivers at that time. Sometimes you can get better performance with proprietary drivers (nVidia video cards are a good example of that), but don't bother until you really need them.

So,

1. Don't just download and install drivers until you really need them.
2. You don't need to mount a driver on disc. Either the package manager will do that for you, or when it is built from source it can be installed directly onto the system disc. Linux has standard ways to do that.

I have installed Linux in various forms on many systems. Usually, I don't need to deal with hardware drivers at all. This includes servers and laptops.

Well, given that a 2TB disc now costs under $100 USD today, I would say about 2 years... :-) A 2GB disc? You can't even buy one any longer, other that in the format of an SD or USB device! :-)

Which of these are WiFi access points? Regular switches and routers may radiate a little EM radiation, but probably no more than a regular AM/FM radio receiver. WiFi operates in the 2.4, 3.7, and 5 Gigahertz bands. They are low power, and unless they, or their antennas, are right next to you (ie, a couple of inches) they should not be an issue. That said, it seems some people are more sensitive to RF radiation than others. If this is your office, ask to be moved. If it is a personal space, move the gear into a closet or away from your normal location.

FWIW, I have a WiFi access point (2.4 and 5GHz) in my home office where I am right now, about 3 feet from me. I have never felt any effect from it.

1. AWS software is probably not purchasable.
2. As Darren said, AWS provides an infrastructure. You pay for servers and services as used, and the more you use, the less you pay per unit time.

The main advantages to using a cloud service like AWS over your own gear are these (mentioned briefly by Darren):

1. There is no up-front capital equipment cost.
2. There is no need for ongoing power, cooling, and hardware replacement costs.
3. You still probably need your own sysadmins since these are "your" systems, even if virtual ones, but Amazon deals with a lot of the stuff your sysadmins usually would have to deal with, such as checking system health, etc. Your sysadmins still have to deal with system configuration, user accounts/permissions, etc.
4. If you don't need a server any longer, you can shut it down and no longer incurr costs associated with it.
5. You can spin up a bunch of servers on short notice if you suddenly need more processing power.
6. They have good (and inexpensive/reliable) data backup capabilities.

I would suggest that you read the CUDA documentation. Most nVidia GPU's have 100's of CUDA cores, so if done properly, this should not be a problem. How many threads are you going to run?

If you are trying to learn objective-c, are you sure you install the gcc-objc (objective-C) compiler/package on your system? Also, try replacing the line #import <stdio.h> with #include <stdio.h>

Mike does do that well (explain stuff) doesn't he? :-)

You are only using 1 server for 10,000+ users? It is doable - we use about 5000 servers to handle approximately 100M mobile browser users (20,000 per server), but we have multiple processes on each server (8+ cores, 64GB RAM each) to handle the load. Also, of those 100M active users, only about 20% of those are active at any time, since they are spread all over the world. So, 20M users on 5000 servers == 4000 concurrent users per server. At 8 cores per server, that would be about 500 users per core.

So, this is NOT a simple problem. Architecture (and redundancy) is everything! In any case, assuming your client base is not in a single location (geographically speaking), then it is highly unlikely that all of them would try to sync at the same time.

Another critical factor is the data load each user syncing will consume - how big are your network pipes to your peering partner? My guess is that you will saturate your internet connections LONG before you overload your server(s).

Usually laptops have a function key that will turn some keys on the board into a number pad. As for the USB number pad approach, it depends upon whether it requires a specific Windows 8 driver to work. Also, is this an x86 system, or is it using an ARM processor?

The school network probably requires a proxy connection to get past the firewall. What did you do specifically to fix this? FWIW, most browsers (Chrome, Firefox, etc) allow you to tell it to auto-discover the network proxy settings and then use those to connect with internet web sites. I had to do that with Firefox on my work CentOS virtual machine.

If you are using gnome, then it isn't Windows, and Visual Studio is not part of this equation. Please tell us EXACTLY what operating system (and version) you are using for this exercise.

I assume that Lightspeed is your ISP? If it is blocking Facebook from your home system (may be a mistake) then either contact Lightspeed to fix this, or try an open VPN (virtual private network) that will mask your query from lightspeed.

I'm not sure I understand your question (#2). I have been programming Linux/Unix systems since 1982, and led the Y2K analysis/remediation effort for a major software company in 1998 (10 million lines of Unix and Windows code). Also, I have written C++ date/time classes that can handle date/time values up to the heat death of the universe, to the millisecond.

So, please be more explicit (and clear) as to what your concerns are.

P.S. The "heat death" of the universe is hypothetical. Let's just say that my date/time classes handle this up to a REALLY big number. :-)

If you want that code, it is now owned by Applied Materials - the 800lb gorilla of the semiconductor manufacturing equipment industry.

Keep up your efforts Mell. Someday you may become a competent coder! :-) We all learn from our mistakes, and it seems you are learning rapidly! Tongue-in-cheek humor here. Keep posting, and we'll keep trying to help you.

The conio.h header file provides functions for input/output to console (cmd.exe) windows. You say it isn't working. Please be more specific.

MySQL is a specific implementation of a relational database that uses the SQL data access language. SQL is an ANSI/ISO standard, and is supported (to varying degrees) by database products such as MySQL, Ingres/Postgres, Oracle, DB2, Sybase/MS-SQLServer, et al. It is not a part of any operating system, per se. MySQL is now part of Oracle, and can be downloaded from here: http://dev.mysql.com/downloads/

Time to visit the "gurus" at your local Apple Store... :-)

The AMD 7660 is a very capable chip set. Myself, I prefer the nVidia gear, but it may be overkill for your needs unless you have some requirements for the nVidia GPU math co-processor acceleration for scientific/engineering purposes.

It depends upon the trojan/virus. Some also infect the boot sector of the system drive, so the next time you boot the system, it gets reinfected. I have also seen some that infect libraries (dll's) in the recovery partition, so even if you reinstall the system, it will still mung your system. These days, anti-virus operations are not so much for amateurs... I used to do this (fix infected systems) for a living as a systems consultant, and still do so part time. I charge people $200 USD per hour for the service.

FYI, there are also viruses that will embed themselves in the system flash (bios) memory, and will reinfect you even if the disc is clean! The NSA uses this method, as documented by Edward Snowden and Bruce Schneier, the noted security expert.

Ok, but yes, most start with including iostream because it has the basic stuff for output to console, input from the terminal, etc.

Stop asking us to help you cheat on your school work asignments! Make an effort and post the code here with your errors, and then we may decide to help you. Until then, you are on your own! :-(

Basic C/C++ stuff. RTFM and try, as suggested by Ancient Dragon, a Google search.

Try using Google to find tools to do this. There are a number of them.

Do some reading on how to program in C/C++, work out the algorithms (processes) to do currency conversions from any currency to any other, and write them down in plain language over and over until you are certain you have the processes correct. Then, start writing code that reflects those processes. When you have some code, we can critique it and help you with your errors.

I see javascript and html. Where is the php code?

See the --help output or the man page. In any case, the -t <rootdir> option on the command line will do the trick. IE, php -t xampp\php for Windows, or php -t xampp/php for Linux/Unix. I suspect that the latter will work for both, though the former (backslash) will ONLY work for Windows. Since I ONLY use php on Linux, I am just guessing, but I do have quite a bit of experience munging with php's source code (to fix some serious bugs with their http support). My current set of projects are php/javascript/css intensive, including writing (done) a cell phone web browser emulator.

TopSpeed
FuelConsumption (adjusted for speed) ie, mpg @ 100 mph != mpg @ 50 mph.
PassengerLoad
Cargo (in cubic whatever is appropriate for you)

Cars have a LOT of options. This will change things quite dramatically when all things are considered. FWIW, I was a master mechanic before I became an engineer, and factory trained by Volvo, Fiat, Lancia, Subaru, Toyota, Jeep, Chrysler ... and have the paperwork to prove it.

Good answers Jorge! I'd give you an up-vote but I see you already got a bunch for your replies to this! No sense in spoiling you... :-)

It is all relative to your needs. My needs as a software engineer who works on complex systems is not the same as most people's. Most people will find a desktop that costs well under $1000 USD to be more than they need. Myself, I spend about 5x that for what I need... :-) Of course what I could get for $5000 5 or 6 years ago is MUCH less capable than what the same $5000 can get me today. That said, my personal workstation (a dual-CPU, 8-core, 3GHz, 8GB RAM system), is still pretty kick-ass! Today, that same $5000 I spent on it at the beginning of 2008 would get me a dual-cpu, 16-24 core, 16-32GB RAM system, with MUCH better I/O speed and better graphics. So, same $$ but 2-3x processing capapbility, 2-4x RAM, and MUCH faster overall throughput. Do I need it? Heck, with 8 cores I can build an entire Linux kernel from scratch in 15 minutes (using 8 cores), or transcode a 45 minute HD video in well under 10 minutes (using about 50% of my processing capability).

No. The answer is reduced latency. :-) Latency is the amount of time it takes for device A to send a message to device B. It is more or less directly (inversely) proportional to the speed of the link. IE, if a 1GB link between A and B takes 10ns (nanoseconds) to send a message, then a 10GB link will take (in theory) 1ns (nanosecond). Of course there are other contributing factors to latency, but this is a good example of why "faster is better".

What did you find when you Googled that? In any case, it is a video input card, not a "Security Camera" card.

You have possibly put this file into the Visual Studio configuration file to be compiled, or the problem is that for C++ (especially for pre-compiled headers) it should either not have the .h extension, or should be something like .hh, .hxx, etc.

Jain, we don't do your homework for you. Please make an attempt to solve the problem, and then post your code here. Then we can help you. Your post is not clear to us - or me at least.

If you want us to helpful, please post your code.

Possibly for stuff like attemts to access the management port (not sure what it would be - read the manufacturer's documentation about remote management), port scanning (common, but may be helpful to know). If you can download the logs to your system, post them here so we can look at what is going on.

@Mike2k
Generally, you are spot-on with regard to the old nVidia drivers, but I have been VERY happy with them since 2008 (CentOS 5.x), including on my nVidia laptop (a Dell D630) as well as my personal workstation (8800GT card). I am running Scientific Linux 6.x on both now, and there are ABSOLUTELY no issues running the proprietary drivers, and I do a lot of graphics intensive work on them.

Recently Linus Torvalds gave nVidia a big "thumbs-up" for their recent efforts to be more supportive of the open source Nouveau driver, instead of the big middle finger he gave them last year! Hopefully they will continue in this direction.

FWIW, the Nouveau drivers have improved significantly as a result of nVidia's contributions to Nouveau. I recently upgraded my workstation to SL 6.5 and the Nouveau driver was MUCH better than the old ones. It still had some irritating glitches, so I replaced it with the current nVidia GeForce driver - no glitches and still performs much better than Nouveau. If it weren't for the small, but irritating glitches (such as issues with cursor tracking between multiple screens), I would have kept the Nouveau driver.

Ok. Totally different thing. The 32-bit system is the print server. Just install the compatible 64-bit driver from HP on the Win8 machine and point your print stream to the 32-bit server.

Usually there are text files in the package that tell you what to do. Many have a configure command that will check your system for needed dependencies and create a Makefile that you can use to build the source into the appropriate bits and pieces, which you can then install with the command "make install". Some don't go that way, but have scripts which do the "right" thing.

Of course, this is mostly a Linux thing. What OS are you running? Windows is another kettle of fish... :-)

I agree with AD and Mike2k about using Qt. It is a cross-platform C++ API that lets you really write once and deploy on multiple system types from Windoze, to Linux, to Unix (BSD et al), to some mobile operating systems, etc. Of course, once you use Windows API's in your code directly, that compatibility goes right out the window (so to speak)!

@NardCake
Some of the issues with nVidia cards is the default Linux use of the open source Nouveau nVidia driver, which while significantly improved as of late, is far behind the proprietary Linux drivers in terms of performance and capability, and in the past it was pretty pathetic.

Yeah, if this wasn't mentioned, I think that currently about 95% of the top 500 super computers in the world are running Linux - you can't get much more "big iron" that that! A lot of them are also running scads of nVidia's tesla GPUs for parallel floating point performance.

What kind of VPN server are you trying to connect with?

First observation is that you need to move the sqlite3* db; declaration into the class member variable list, otherwise it will not be accessible outside of the DbInit() method. Also, I would create a set of constructors that call DbInit(), and a virtual destructor that disconnects from the database.

Try the jar command? It is part of the java development tool set.

Lambdabeta's answer is a good example of applying the KISS (Keep It Short and Simple) to code. IE, keep it as simple as possible, but no simpler than necessary. Too simple (short) usually makes stuff hard to understand. Also, don't hesitate to add comments to the code, especially stating what your intention was when you wrote it. If someone else has to come along and fix bugs or enhance it, that can be critically useful. Comments only increase the size of the source file, but do nothing to increase the size or impact the performance of the object and executable files.

Good luck Kent! :-) Sometimes this stuff gets my eyeballs rolling, and I've been doing it for 30+ years! Deceptikon's answer was to the point. Couldn't state it better myself.

Sigh - C++ for Dummies. The problem with that is that dummies are not going to be able to write decent C++ code, no matter how hard they try! After 22 years as a professional C++ developer, I can write "decent" C++ code, and have done (and still do) some pretty impressive stuff, but am I the "best" in the business? Not a chance! I still have to go back to the manual on a regular basis... :-) Best reference? I personally prefer Ellis and Stroustrup's ARM (Annotated Reference Manual - ANSI version). It is short, but dense, and covers the basics very well. That said, C++ has undergone some major changes in the past few years (C++11, C++14). Here is a link to Stroustrup's web site where he discusses C++11 and such - as the author/inventor of C++ he is immenently qualified to comment on changes in the language.

http://www.stroustrup.com/arm.html

What API's does Ilog provide for Java? I haven't done anything with Ilog since the 1990's. I still have the documentation though. :-)

Suzie999 was being nice to you. Usually we just say that we don't do your homework for you, but if you try and post your code here, we will usually try to point you in the right direction. So, get to work, and make an effort!

1. What is the Kinect's physical interface to the PC - USB?
2. What do you mean by "Is it possible to run applications on computers Kinect ARM architecture?"? Are you saying that the Kinect runs an ARM processor and you want to run apps on it directly?
   3&4. Yes, you can develop/run C# applications on Linux with the mono environment. As for install/use the "wrapper" I can't say since I don't run C# in a Linux environment.
3. MS allowing use of Kinect on a different OS? Probably not. I think you are on your own. If the Kinect has an ARM architecture, then it is likely that the OS is MS's RT version of Windows. In any case, it is highly unlikely that they wan't the kinect to be used by anything but their own tools.

All that being said, I'd be happy to find out I am wrong about most of that.