Computing desk
< April 16	<< Mar \| April \| May >>	April 18 >

Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

April 17

AOC Slot

In the context of this chassis, what is an ‘’AOC’’ slot? I tried looking it up, but all I got is this disambiguation page, and none of the options seem to apply. Thanks. Rocketshiporion ♫ 10:46, 17 April 2011 (UTC)[reply]

I think this means add-on card. ---— Gadget850 (Ed) ^talk 13:45, 17 April 2011 (UTC)[reply]

hi-speed internet service that does NOT involve a fixed IP address (but not using a proxy)

Dial-up internet access involves a non-fixed IP address. Some high-speed internet access necessitates a fixed IP address. I've heard that at least one type of high-speed access (for an ordinary home computer) does NOT necessitate a fixed IP address (I am not referring to the use of proxies). What is it / are they? TIA63.17.51.171 (talk) 08:47, 17 April 2011 (UTC)[reply]

As a general rule, home broadband connections do not have a fixed IP address.--Phil Holmes (talk) 12:00, 17 April 2011 (UTC)[reply]

More specifically, the internet connection does not require a fixed IP. If your internet provider chooses to give you a fixed IP, there's little that you can do about it. This problem falls into the category of "contact your network administrator," who is in this case your internet provider; realistically, it's unlikely they will change their network management practices to suit your preferences one way or the other, but it ultimately boils down to a decision made by the NOC engineer. Nimur (talk) 16:03, 17 April 2011 (UTC)[reply]

video resolution; crt vs lcd?

ok, this is a question about: can the resolution of a display adapter possibly be affected by the monitor type? here's the story: an HP media center with a PCIexpress video board. (nvidia 9400). not very high end because power supply is only 300 watts, but good enough for nonhd tv. windows 7. anyway. that doesn't matter so much, since this starts with the board getting sick, and removed from the machine when i realize the motherboard has a built in vga port that shuts down when the pciE slot is occupied. so, i run the (dell 15 inch) crt monitor off the builtin port on the motherboard, and it runs in straight vga resolution, only. that seems reasonable, if the builtin port had better resolution, they wouldn't put the pcie video board in, would they? next, i put the flaky video board back, figuring intermittent 1200 pix resolution is better than fulltime VGA. meanwhile, i find a NEC 17 inch LCD monitor at goodwill for $15. Woohoo! then i decide to remove the video board again and run the lcd monitor off the builtin video port, the one that was vga only with the crt. but this combo runs nicely at 1200 pix, true color; and i can even choose 1600 pix if i want. seems just as fast as the pcie board, too. so; where/how/why is this possibly working? and, as a practical matter, why not just abandon the PCIE board entirely, right? Gzuckier (talk) 17:00, 17 April 2011 (UTC)[reply]

Have you installed the driver for the onboard video card? It sounds a lot like it doesn't have a driver installed and thus it falls back on the default mode for safety. Also the reason that you can't get both working at once is because the onboard one is actually PCIe and your motherboard will only let you use either the onboard PCIe or an expansion PCIe video card. The onboard one should be more than capable of running at a decent resolution, but you still need to install the driver (try the support website of whoever makes the computer). If however the onboard one isn't powerful enough (gaming for example) that's when you'd want to use a separate card. ZX81 ^talk 17:13, 17 April 2011 (UTC)[reply]

Yes, just use the built-in board. The only disadvantage I can think of is that some use part of the computer's RAM for graphics, but probably not much. As for why it refused to give a higher res on the smaller monitor, I suspect they have that logic there to avoid high res on a small CRT screen, since, with some monitors, that just makes it fuzzy (due to overlapping pixels). The separate graphics card is probably newer, so doesn't use that same logic, since they assume newer monitors can handle it. StuRat (talk) 22:54, 17 April 2011 (UTC)[reply]

DHL Indian premier league

Hi, Can I get detailed scoreboard of DHL IPL KKR vs Rajasthan royals?Gio2050 (talk) 17:43, 17 April 2011 (UTC)[reply]

Either this or this. I don't understand any of the numbers and terms in the linked pages, though. 118.96.166.45 (talk) 02:21, 18 April 2011 (UTC)[reply]

Connecting computer to TV via HDMI - screen resolution problem

Hi, I have a computer that is connected to the living room TV (a Panasonic) via HDMI. There is no other monitor connected. My problem is that the computer, which is running Windows XP, does not allow me to set the proper resolution for the TV. Both the graphics adapter and the TV should support the 1280x720 resolution, but it cannot be selected - the only available options are 1280x600 and 800x600, both in the "native" Windows dialog box and the custom Intel graphics options dialog box. Do anyone have a suggestion for a solution for this? Things I've thought of:

Setting the resolution directly in the registry (where?)
Installing some "custom" monitor driver (the TV manufacturer does not appear to provide any, currently the "generic" one is used)

Details on the setup is in the collapsed box.

Details

Connection: DVI output on the computer via a passive DVI->HDMI adapter to the HDMI input on the TV, audio is run on a separate link, the TV is able to combine video and audio without any problem, the problem is there regardless of whether or not the audio is connected. The connection is several meters long through some walls, for this reason using a VGA cable instead is not an option.
TV: Panasonic TX-L32X10Y, European version; a 720p 32" quite "regular" LCD TV. Allowed resolutions according to manual:

Signal name: 640x480 @60HZ Horizontal frequency: 31.47 kHz Vertical frequency: 60Hz
Signal name: 750/720) /60p Horizontal frequency: 45.00 kHz Vertical frequency: 60Hz
Signal name: 1,125 (1,080) / 60p Horizontal frequency: 67.50 kHz Vertical frequency: 60Hz

(this is exactly how the manual presents it. PC via D-SUB (VGA cable) and "regular" HDMI have more alternatives.) Messing with the "zoom" settings on the TV does not affect the available resolution options on the computer.

Computer: The following is a printout from one of the graphics adapter option pages. I think it covers most of it. The computer is a Dell.

	INTEL(R) EXTREME GRAPHICS 2 REPORT


Report Date:		04/17/2011
Report Time[hr:mm:ss]:	20:18:02
Driver Version:		6.14.10.4396
Operating System:		Windows XP* Professional, Service Pack 3 (5.1.2600)
Default Language:		English
DirectX* Version:		9.0
Physical Memory:		1021 MB
Minimum Graphics Memory:	1 MB
Maximum Graphics Memory:	96 MB
Graphics Memory in Use:	6 MB
Processor:		x86
Processor Speed:		2593 MHZ
Vendor ID:		8086
Device ID:		2572
Device Revision:		02


*   Accelerator Information   *

Accelerator in Use:		Intel(R) 82865G Graphics Controller
Video BIOS:		2972
Current Graphics Mode:	1280 by 600 True Color (60 Hz)



*   Devices Connected to the Graphics Accelerator   *


Active Digital Displays: 1


*   Digital Display   *

Monitor Name:		Plug and Play Monitor
Display Type:		Digital
Gamma Value:		2.20
DDC2 Protocol:		Supported
Maximum Image Size:	Horizontal: Not Available
			Vertical:   Not Available
Monitor Supported Modes:
1280 by 720 (50 Hz)
1280 by 720 (60 Hz)
Display Power Management Support:
	Standby Mode:	Not Supported
	Suspend Mode:	Not Supported
	Active Off Mode: Not Supported

* Other names and brands are the property of their respective owners.

Note that the report explicitly says that the TV supports 1280x720. Still, I am not allowed to select it in Graphics Options, only 1280x600 and 800x600 is available. For 800x600, there's a lot of black around the edges; for 1280x600, the screen is "zoomed" so the edges of the monitor image (like the taskbar) is not visible.

Other: The computer is running Windows XP. More recent versions of Windows are not an option (I have no licence). Linux is probably not an option (some of the video streaming sites I plan to use do not support it, I think)

Thanks for any help! Jørgen (talk) 18:47, 17 April 2011 (UTC)[reply]

Note that just because a certain resolution is supported, that doesn't always mean it's supported via all connectors. It's possible that the 1280×720 res is only supported via VGA. You might want to temporarily move the computer close to the TV, so you can test with a VGA cable to see it that works. StuRat (talk) 22:39, 17 April 2011 (UTC)[reply]

Another option is to go with the 1280×600 res, and adjust the vertical height on the TV (if it is adjustable) to stretch it a bit. Yes, that's not ideal, but you might prefer it over the black bars. StuRat (talk) 22:43, 17 April 2011 (UTC)[reply]

Thank you, but that is sadly not feasible. As said above, the TV explicitly says it supports the resolution via HDMI. And I do not see whether it works via VGA or not would teach me anything new. Yes, I can stretch the TV picture somewhat, but I think it will make for inferior film-viewing quality... Do anyone know how to force the screen resolution on the computer? Jørgen (talk) 14:04, 18 April 2011 (UTC)[reply]

There appears to be something weird about that particular model of TV. According to your TV's manual it would appear to be capable of supporting 720p and 1080p signals received through a HDMI cable (see p.59), yet it only has 1366x768 pixels on the LCD panel (see p.63). I am somewhat unsure how it would display a 1080p image (with 1920x1080 pixels) on the lower resolution screen.

I do know I have no such problems connecting my PC to my Panasonic TV using a HDMI cable, but my TV has the full 1920x1080 pixels and my PC has a HDMI output port. Even so, I think your problem is more likely to do with the PC's display driver - I do wonder why it opts for "plug and play monitor" and restricts you to just 2 resolutions? - or it could be something to do with you converting DVI to HDMI.

One hint: if you do get it working you will need to turn off the "picture overscan" feature (mentioned on p.29 of the manual) or it will crop an annoying few pixels from around the edge. Astronaut (talk) 17:14, 18 April 2011 (UTC)[reply]

Thanks! I think the TV, if fed a 1080p signal, just downscales it - if I set my TV set-top box or DVD player to 1080p output the TV still works fine. I think you are right in suspecting the display driver. Another annoying quirk is that it has to recognize the TV at start-up time - if the TV is not on and connected when I turn on the computer it won't send any signal when I turn the TV on later. Jørgen (talk) 20:29, 18 April 2011 (UTC)[reply]

"Address of property" in C#?

At work, I have to deal with "data container" classes in C# that have many properties to contain values, and the code that handles these properties makes pretty much similar checks for every property, with the only variations being the name and type of the property. So I got an idea to write a reusable method for the checks, but I don't know how to do that in C#. The idea is, I think, best expressed in C code:

#include <stdio.h>
struct foo {
  int a;
  int b;
};
 void setvalue(int *address, int value) {
  *address=value;
}
int main(void) {
  struct foo testfoo;
  setvalue(&testfoo.a, 1);
  setvalue(&testfoo.b, 2);
  printf("%d %d\n", testfoo.a, testfoo.b);
}

This code prints out "1 2". You'll notice there is not a single direct assignment to testfoo.a or testfoo.b in the code, all assigning is done in the setvalue() function. In analogue with the real-world code, setvalue() is more complex. Is there a way to do such a thing in C#? In C#, the "data container" classes are more like this:

class Foo {
  public int A { get; set; }
  public int B { get; set; }
}

Is there a way to store the "properties themselves" A and B in some variable, rather than accessing their values? JIP | Talk 19:03, 17 April 2011 (UTC)[reply]

I'm not sure I understand what you're trying to do. If you use fields, ref does what you want. If you use properties, ideally this kind of checking should happen in the set method; that's what it's for. The notion of a property reference does make sense; it would simply be the get and set methods (as delegates) packaged together. But there's no syntax for that. I don't think C# even lets you refer to these methods (without calling them), even though they're just ordinary methods at the .NET level. You could do something like setvalue((x) => { testfoo.a = x; }, 1);, but it's ugly. You could also consider testfoo.a = checkvalue(1);, where checkvalue either returns its argument or throws an exception. -- BenRG (talk) 10:59, 18 April 2011 (UTC)[reply]

The setting of the property values is a bit complex. The values are read from a different object, where they reside in a name-based dictionary, not as separate fields. There is separate validity checking for each field. What this means in practice is that each assignment goes something like this:

if (source.IsValid("A")) {
  destination.A = source.GetValue<int>("A");
}
if (source.IsValid("B")) {
  destination.B = source.GetValue<int>("B");
}

I would like to make a reusable method so I could do something like this (not valid C#):

void SetValueToFoo<T>(Foo foo, Source source, Property prop, string name) {
  if (source.IsValid(name)) {
    foo.prop = source.GetValue<T>(name);
  }
}
SetValueToFoo<int>(destination, source, A, "A");
SetValueToFoo<int>(destination, source, B, "B");

But I don't know how that is possible in C#. JIP | Talk 19:15, 19 April 2011 (UTC)[reply]

I think that should be possible if you utilize reflection. That should allow you to set a property of source based on the property's name as a string. --Cybercobra (talk) 13:22, 22 April 2011 (UTC)[reply]

Hardware random number generators

I am interested in the range of prices at which USB hardware random number generators are available. What are the cheapest and most expensive ones that people know of? Thanks! ╟─Treasury Tag►consulate─╢ 22:18, 17 April 2011 (UTC) [reply]

Side-discussion about the true nature of random numbers. The question was about the price of USB RNGs. ╟─Treasury Tag►consulate─╢ 17:10, 19 April 2011 (UTC)[reply]

The following discussion has been closed. Please do not modify it.

If you have a modern Linux, you can use /dev/random to create hardware (non-pseudo) random numbers. If you need random numbers for security or cryptographic purposes, read the detailed descriptions for your particular Linux and hardware configurations. Some esoteric Linux distributions replace or alias /dev/random with /dev/urandom, which is a software, pseudo random number (which may or may not be hardware-seeded, depending on your system). Again, if you aren't sure of the difference, feel free to ask followups or clarify your question. Ultimately, you need to decide whether a thermal noise meter or a radio-receiver is actually more random than any other set of hardware-seeded random numbers. There's a critical and subtle distinction between "uniformly distributed" and "unpredictable." Similarly, there is a difference between a hardware-based random number and a hardware-seeded pseudorandom generator. Nimur (talk) 02:19, 18 April 2011 (UTC)[reply]

That didn't answer the question, though which was about USB hardware random number generators and the prices at which they retail. ╟─Treasury Tag►pikuach nefesh─╢ 08:09, 18 April 2011 (UTC)[reply]

The point is that a modern computer has a hardware random number generator in it already, so you can get one for zero dollars. Nimur (talk) 13:31, 18 April 2011 (UTC)[reply]

Is it a USB hardware random number generator? No. What's the matter with you? ╟─Treasury Tag►Speaker─╢ 07:17, 19 April 2011 (UTC)[reply]

I'm not sure /dev/random counts as a "hardware generator". It's essentially a software generator whose entropy pool can be configured to use various hardware sources as inputs (Typically mice and keyboards and whatnot, but also the more dedicated hardware devices the poster is probably asking about.) APL (talk) 15:20, 18 April 2011 (UTC)[reply]

Why is diode shot noise any more random than mouse input? Both are based on non-software interaction between the computer and the physical world. Nimur (talk) 15:57, 18 April 2011 (UTC)[reply]

I didn't say it was. My point was only that /dev/random itself is a software RNG, that can (optionally, but by default) use some hardware elements. (Potentially even including purpose-built equipment of the sort I assumed the question was about.)

You might claim that the mouse,keyboard, and IDE harddrives themselves form a hardware random number generator to which /dev/random is only the interface. But that's clearly not true. None of those hardware items generate acceptable random numbers. Their only value is as entropy for an actual software-based RNG. The question is about hardware-based RNGs.

So I stand by my comment that, despite it's usefulness, /dev/random does not count as a "hardware random number generator"

However, if you want me to play devil's advocate, I'd say that mice and keyboards generate little to no entropy when there is no user present. Most of /dev/random's commonly used entropy sources fall victim to potential problems along those lines. A default Linux install will typically depend on mouse,keyboard, and IDE. On a server that doesn't access the disk very often the entropy pool can often run dry.

Also the level of entropy from such devices may be significantly less than they appear. (For example, An unattended mouse may continue to report movement, but in a highly repetitive manner.) APL (talk) 18:16, 18 April 2011 (UTC)[reply]

Yes, but as I mentioned before, a modern CPU has a hardware random number generator in it. Here's the Intel vPro implementation: Advanced Security Features of Intel® vPro™ Technology. Nimur (talk) 17:02, 19 April 2011 (UTC)[reply]

There are some prices at comparison of hardware random number generators. Prices in that table seem to be more-or-less correlated with throughput. Gandalf61 (talk) 14:29, 18 April 2011 (UTC)[reply]

The most straightforward thing to do is use an RNG-equipped smart card (US$5-10) and a USB card reader (start around $10), or a security token containing basically the equivalent. Depending on your application, you may actually want a host security module, which costs a lot more. Some Intel chipsets used to have built-in RNG's but I don't know if they are still made. It's not clear to me who (if anyone) uses those high speed RNG's in the "comparison" article. Usually you just need enough physically random bits to seed a CSPRNG and freshen it up once in a while. 69.111.194.167 (talk) 17:38, 19 April 2011 (UTC)[reply]

The Comparison of hardware random number generators article neglects to mention a much cheaper (albeit) slower device. And since all numbers are binary anyway, this is probably cheaper still! Mitch Ames (talk) 11:41, 22 April 2011 (UTC)[reply]