FAQ about hardware

What is the speed of usb 2.0?
Below is the summazired speed for some common transfer protocol

Protocol : Transfer rate(MB/s)
IrDA-Control : 0.009
Serial : 0.02
Parallel : 1.0
Bluetooth 1.1 : 0.125
Bluetooth 2 : 2 to 12
IEEE 802.11 : 0.25
IEEE 802.11a : 6.75
IEEE 802.11b : 1.375
IEEE 802.11g : 6.75
USB 1.1 : 1.5
USB 2.0 : 60
SCSI 1 : 5
Fast SCSI 2 : 10
Fast Wide SCSI 2 : 20
Ultra SCSI : 20
Ultra Wide SCSI : 40
Ultra2 SCSI : 80
Ultra 160 SCSI : 160
FireWire (IEEE 1394) : 50
FireWire (IEEE 1394b) : 100
Ultra DMA ATA 33 : 33
Ultra DMA ATA 66 : 66
Ultra DMA ATA 100 : 100
Ultra DMA ATA 133 : 133
SATA : 150
SATA 2 : 375
PCI : 132
PCI-X (64bit) : 1.1 GB/s
PCI Express x1 : 250
PCI Express x16 : 4 GB/s
PCI Express x32 : 8 GB/s

Can I use ATI card on SLI mobo?
Yes you can. Note that it will not behave as SLI mode if you plug in 2 of ATI cards.

Can I put a pci-e graphic card in agp slot?
No you can't because the the slot is different.

My mobo only support DDR266. Can i put DDR400 in my mobo?
Yes you can but the DDR400 will become DDR266.

My hard disk is a 80 Gb model, why do I only have 74Gb avaible?
Hard disk manufacturers consider one Kilobyte is equal to 1000 Bytes, one Megabyte is equal to 1000 Kilobytes and one Gigabyte is equal to 1000 Megabytes. This is, however, not true. The real values are one Kilobyte is equal to 1024 Bytes, one Megbyte is equal to 1024 Kilobytes (1048576 Bytes) and so on. So if you're has 80 000 000 000 Bytes, hard disk manufacturers say it is an 80 Gb Hard Disk where as it really has only 74.5 Gb.

For quick reference I have made a simple table:
20 Gb - 18.62 actual Gb
40 Gb - 37.25 actual Gb
60 Gb - 55.87 actual Gb
80 Gb - 74.5 actual Gb
100 Gb - 93.13 actual Gb
120 Gb - 111.75 actual Gb
160 Gb - 149.01 actual Gb

I have two sticks or RAM that are the same model, same speed and same make(i.e.Kingston DDR400 512MB x2). Can i run both as dual-channel?
Yes, you can. Make sure that your mobo supports dual-channel and also the speed of your RAM.

edited by acedriver, credit to everyone who contribute to this thread, some posts need to be deleted..

Plugs and ports

Parallel ports
A parallel port is easy to recognize by its 25 holes.


Serial port
A serial port has 9 or 25 pins.


VGA graphics port
Has 15 holes.


Mouse port

A serial port may look like this.


This is a PS/2 mouse plug which isn't compatible with the serial port.


Keyboard port
Some PCs have a five-hole DIN connection. Others have a PS/2 keyboard port.

Keyboard port, DIN connection


Keyboard plug, DIN


USB

USB Ports


USB plug. This can be plugged in and out while the PC is on.


FireWire


Picture courtesy of ThePC.info

Pentium from Intel .

There's Pentium , Pentium 2 , Pentium 3 and Pentium 4 Proccessor until now . Every batch with different core such as Coppermine , Tualatin , E , C , B , A and bla bla .

Then Athlon is from AMD .

Athlon such as K6-3 , K7 with different core such as Barton , T-bred , Palo ..

Return Merchandise Authorization (RMA)

Chip packaging terms.

BGA = Ball Grid Array
TSOP = Thin Small Outline Package

No more acronym, if you wanna know about PC acronym go to TUCAA: The Ultimate Computer Acronyms Archive

Please only discuss basic computer functions.

Whenever you see a weird new term, remember:

http://webopedia.com

Pentium 4 willamette 400Mhz FSB (100Mhz Quad Pump) i840 chipset .
Pentium 4A Northwood with 400Mhz (100Mhz Quad Pump)
Pentium 4B Northwood with 533Mhz (133Mhz Quad Pump)
Pentium 4C Northwood with 800Mhz (200Mhz Quad Pump)
Pentium 4E Prescott with 800Mhz (200Mhz Quad Pump) 1Mb L2 Cache , 0.09n micron
Pentium 4EE (Extreme Edition) Gallatin additional 2MB L3 Cache .

the general conception is that the higher is the alphabet the better , mostly right except for the E version with contains optimizations for future software but not current software plus it generates 106 W/mk compared to the Northwood's 82 W/mk

EDIT: Some corrections:

Besides the FSB different between those chip , there's also pipeline stage different , micron built and etc . So every chip have different enchancement . (P.s all of those I dig from my memory , don't know if there's some minor error) .

Intel Pentium 3 if I am not mistaken have Tualatin Core , 133FSB with clock more than 1Ghz . And also their special Coppermine (800EB , 600EB .. etc) with special 133Mhz , compared with P3 100Mhz

THE DEFINITIVE GRAPHIC CARD REFERENCE

Wanted to add something very important too .

User always get confused with Manufacturer's brand and also model .

Let me do a very simple summarization .

Additional knowledge about AMD chips .

AMD chips performance cannot be measured by it clock speed , I mean compare it clock to clock with P4 chip .

For example , comparing a A64 2800+ with a Prescott 2.8Ghz . A64 2800+ running at 1.7Ghz (if i am not mistaken) and P42.8E at 2.8Ghz . Okay , P42.8E looks like have alot of advantages in the clock speed , but that doesn't mean A64 2800+ is slower than that . It's because the chips design , pipeline stage , cache , and the 64bit makes the 64bit processor to be faster than P42.8E at most cases . It's like 20%+ boost in gaming and comparable to any P43Ghz and above , except P4EE .

Even a Barton 3200+ running only at 2.xGhz is faster than P42.8E alot . AMD 3200+ is read as performance equivalent with P43.2Ghz or above , and don't measure the clock to clock performance next time .

BTW , I think this thread's name should be change . PC for dummies seems lame , because newbies sometimes tend to think that they are not dummies , and makes alot of different perspective . Mod ?

Sniper's Centrino FAQ

What is Centrino?

Centrino is a concept/package/ "stable image" promoted by Intel to represent what they think mobile computing should be.

- Powerful
- Light
- Long battery life
- Wireless capability

To be qualified as a Centrino notebook (and use of the Centrino sticker) , the product MUST have all 3 of the following:

UPDATE: Centrino family tree

Carmel : The first Centrino platform
- Intel Pentium M processor (Banias and later Dothan 400FSB core)
- Intel 855GM/GME/PM chipsets (Montara/ Odem)
- Intel Wireless Pro WLAN card (Calexico)

Sonoma: Current Centrino platform
- Intel Pentium M processor (Dothan 400/533FSB)
- Intel 915GM/GMS/PM chipset (Alviso) --> DDR1&2, SATA, beefier video,PCI-E
- Intel Wireless Pro WLAN card (Calexico 2)

Napa: 2006 Centrino platform
- Intel Pentium M single and dual core processor ( Yonah core )
- Intel 945GM/PM chipset (Calistoga)
- Intel 3945ABG WLAN (Golan) --> PCI-E interface.

Later in 2006 --> Merom dual core CPU with EMT64 support.


I have a notebook that has an ATI Mobility Radeon 9700. Is that a Centrino?

Yes, if it's using a 855PM/915PM chipset motherboard which is designed to be used with discrete graphic cards. 855/915GM has integrated (Intel Extreme Graphics 2) graphics. 855GME is the "enhanced" version of the GM that supports the Dothan and DDR333 memories.


Then how come by Acer [place model name] has 855GME but got 9700?

The 855GME supports AGP4X as well and it's easier/cheaper to just make one multipurpose motherboard....disable the integrated graphics and plug in a graphic card for one product while leaving the integrated graphics on for another product.

I have a notebook that has integrated ATI graphics. Is it a Centrino?

No, Centrino is like a marketing package so to speak. Instead of spending 3 times the money on advertising " Pentium M roxs", " 855/915 chipset no regrets" or " Intel Wireless Pro!!! ", Intel just spend one budget saying "Centrino rocks"

Think of it like a McDonalds Value Meal. It's got a burger, fries and coke. If you order a burger, fries and coffee.... it's not a Value Meal.

If so, why must I get a "Centrino"?

Not necessarily but....

We all know that the Pentium M processors rocks especially the Dothan with AMD64 rivalling performance in that only consume 21W of power versus the AMD's 67W and 35W.

There are also low voltage versions that consumes even less.

There are very little alternatives to the 2nd component of "Centrino" - the chipset. Besides Intel's own 855 series, only certain ATI chipsets supports the Pentium M.

Edit: Now there's more socket 479 chipsets available like VIA's PN800 and Nvidia's upcoming C51 in addition to newer ATI chipsets.

AMD's Turion64 is also seen as an alternative now but availability is just starting to trickle in and their cpu pricing is within 10% of the equavalent Pentium-M price.

As for the WLAN card, notebook bioses are not as flexible as desktops and it's always best to install parts that's certified by the manufacturer. Intel's WLAN card is obviously certified by every manufacturer under the blue sky while Cap Ayam may not be in the key parts list.

How come my Centrino got no Wireless Lan?

If it's got the Centrino sticker on the notebook, it should have wireless lan. Else you can sue their a$$ off and complain to Intel.

If it's got a Pentium M sticker on the notebook and the salesman tells you " Ma Car...ini banyak power mar... Centrino 1.5Ghz wor" then I suggest you check the specs of what you're getting and make sure Wireless Lan card is included.

Mmmm... AMD64 or "Centrino"? Which?

In most applications including games, both perform quite similiarly especially if compared with a Dothan.

Until lower voltage version of the A64 hits the market, the average "gaming" Centrino notebooks weigh less than 2.8kg while the average A64 notebook weigh 3.5kg and shorter battery life.

EDIT: See the Turion comment above.

Mmmm... "Centrino" or Celeron-M? Which?

Essentially the Celeron-M ( this is NOT mobile Celeron or Celeron Mobile) is the same darn thing as it's Pentium-M brother based on either the older Banias or current Dothan core. Difference being is half the L2 cache is disabled, 400FSB vs 533FSB and it doesn't support Enhanced Speedstep ( lowering down speed to save batt ).

To keep it from eating into the sales of the more expensive Pentium-M ( the Celeron-M's cpu price is half or less than a Pentium M), the fastest Celeron M at the moment is 1.7Ghz which is less than the slowest commercially available Pentium M @ 1.7Ghz. Clock for clock, performance is within a few single digit percentage difference for most applications.

So, if you're on a budget or spends most of the time plugged onto the AC, the Celeron-M is a viable consideration, in fact a much more viable consideration than the Semprons and others out there. Else, if every single second of batt life is important to you, Pentium M is the one for you.

What is MXM?

In short, MXM is an initiative pushed by Nvidia (ATI also adopted it after nobody gives a damn about its phailed AXIOM) to standardize the form factor of notebook vga's. The MXM form factor itself can be split into the following:

1) MXM1 for lil wussy cards like the 6200, 7300 etc
2) MXM2 for mid range vga's
3) MXM3 for performance range...this would be like the 7900GS etc
4) MXM-HE for the mucho taikor GTX's.

Note: Let's say ur notebook has a MXM3 card, MXM2 and MXM1 should fit but who the heck downgrades their vga.

So, does this mean I can upgrade me lappy?

Generally, that's no.
Manufacturers generally adopts MXM for faster time to market since it's much easier with a reference design. Additionally, notebooks are so optimised that a card from product A may not work with product B .... and the cooling system's different anyways.
Having said so, certain Dell models do allow a vga upgrade within a certain time frame but.....let's say u have a X700 now and 2 years down the road, you're not going to be able to upgrade it.

My answer even simpler; can

by default ddr333 & ddr400 will run @ the desire motherboard setting
it is back compatible,but it wont run at its rated speed

VGA memory: What does the "ns" in the specs means?

It's simply the speed the memory chips are rated to operate by the manufacturer.

For example:

1.6ns means rams are rated for 1250Mhz
2.0ns means rams are rated for 1000Mhz.

Calculation:

1000/1.6ns = 625

DDR means 625x2 = 1250Mhz

To find out what your VGA memory are rated, check the markings on the chips itself (no...no way you can tell by software) and check at the ramchip manufacturer's website (samsung, hynix, infineon etc).

Where can I check my HDD rpm??
No manual book.

Differences between Athlon64



if above link fail
http://img72.exs.cx/img72/1821/table9zc.gif

Enjoy

---------------------------------------------------------------------------------------

Q: If my mobo only support max of DDR 266 can i put in DDR 400??
A: Sure, but the DDR400 will only run @ max capacity of mobo = DDR 266

Root Mean Sqaured (RMS - some engineering term) is the amount of power produced by the speakers over a continuous period. This rating is measured in wattage, and higher ratings usually mean better sound quality.

Usually speakers would provide another rating Max Power Output (or something like dat). By right, RMS wattage is what to consider when buying speakers, as it measures how much power the speakers can safely handle for long periods of time. P.M.P.O just measures the maximum power that can be produced by the speakers, of course just for over a split second
In proper speakers, you would usually expect the P.M.P.O rating to be about double the RMS...

Actually, generally speakering, RMS determines the VOLUME more than the SOUND QUALITY. For multimedia speakers, the driver size effects the sound quality more than the RMS. Higher RMS means you'll get a speaker that is more powerful, and can reach higher volume, but it doesnt say much about sound quality. If you're into sound quality, look out of driver size, build material and then only RMS.

Alright, i wanna borrow this thread to ask some questions.

I have been away from hardware forum that i missed a lot of the latest cards feedback.

What i would like to know is the comparison between 6600GT and X700 Pro .
I know 6600GT is faster in most of the benchies but does the quality of 6600GT match X700 pro.

(2) X700Pro has 6 vertex shaders and 6600GT has 3 VS. Does it play a significant role in today's game.

Cheers.

No need to see specs lar. It's generally agreed that the 6600GT tar pau's the X700Pro prompting ATI to paper launch the X700XT which didn't really show up in retail until it's discontinued. ATI's next contender for the 6600GT category will be a scaled down version of the X800.

As for image quality... no difference in IQ these days between the 2 makes.

Overview of Intel & AMD processors

Intel


AMD



Source:www.tomshardware.com

CPU SOCKET TYPE

Socket Type
There are many different slots and sockets for CPUs, the slots correspond with processor types and with processor manufacturers. Newer sockets, those with a three digit number, are named after the number of pins they contain, where as older ones are named in the order of their invention or generation and usually contain a single digit.

Sockets supporting Intel CPUs

* Socket 1 - 80486SX, 80486DX, 80486DX2, 80486DX4
* Socket 2 - 80486SX, 80486DX, 80486DX2, 80486DX4
* Socket 3 - 80486SX, 80486DX, 80486DX2, 80486DX4
* Socket 4 and 5 - early Intel Pentium processors
* Socket 6 - 80486DX4
* Socket 7 - Intel Pentium and Pentium MMX
* Socket 8 - Intel Pentium Pro
* Slot 1 - Intel Pentium II, older Pentium III, and Celeron processors (233 MHz - 1.13 GHz)
* Slot 2 - Intel Xeon processors based on Pentium II/III cores
* Socket 370 - newer Pentium III and Celeron processors (800 MHz - 1.4 GHz)
* Socket 423 - Pentium 4 and Celeron processors (based on the Willamette core)
* Socket 478 - Intel Pentium 4 and Celeron processors
* Socket 603/604 - Intel Xeon processors
* Socket 755 - Intel Pentium 4 and Celeron processors



Sockets supporting AMD CPUs
# Socket 7 - AMD 80486, K5, and K6 processors
# SuperSocket7 - AMD K6, K6-2, and K6-3 processors
# Slot A - older AMD Athlon and Duron processors
# Socket 462 (or Socket A) - newer AMD Athlon, Athlon XP, and Duron processors
# Socket 754 - lower end AMD Athlon 64 with single-channel memory support
# Socket 939 - AMD Athlon 64 and AMD Athlon FX with dual-channel memory support
# Socket 940 - AMD Opteron and early AMD Athlon FX processors

BIOS Flashing - How to flash a motherboard BIOS

What is the BIOS?

The BIOS is a piece of software that is permanently stored in a chip on the motherboard. When you first turn on a computer, the BIOS program is initiated. It performs a hardware check and makes sure that certain crucial pieces of hardware are present and functioning properly. For example, it makes sure there is a video card, a CPU, memory, and a power supply providing proper voltages. If there is a problem, it will inform you via beep codes. If everything is fine, it will display a startup screen, the first thing you see on your monitor. It then proceeds to do a memory check, discover other hardware, configure devices as needed, identify the boot drive, and then hand over control to the boot sector on the boot drive, which launches the operating system.

What does "flashing the BIOS" mean?

This BIOS software is in some regards like other software. There are newer versions available that contain updates, enhanced, features, new features, and bug fixes. It is possible to install newer BIOS versions and upgrade, but the process is different from upgrading regular software that's installed on your computer. The BIOS software is not stored on the hard drive, but in a chip on the motherboard. To erase the software on the chip and program it with a newer software version you need to use a special program called a flash utility, and the process of performing the upgrade is called flashing the BIOS.

Why would you need to flash the BIOS?

The most popular reasons that cause people to flash the BIOS are...

* Support for newer processors - The BIOS allows the motherboard to accept processors up to a certain speed. Since you got the computer, the top speed of the processor type your computer accepts has increased. You decide to replace your processor with a faster one, but the BIOS does not recognize it or does not have the right settings. Upgrading the BIOS can resolve this problem.
* Support for bigger hard drives - The BIOS allows the motherboard to accept hard drives up to a certain size. Since you got the computer, the maximum size for hard drives has increased. You decide to install a newer and bigger hard drive, but the BIOS does not recognize it or only recognizes part of the drive. Upgrading the BIOS can resolve this problem.
* Bug fixes - The BIOS contains certain features and options, but one of them does not function properly or not at all. Since you got the computer, the manufacturer has recognized and corrected the problem with a new BIOS version. Upgrading the BIOS can resolve this problem.

The hazards of BIOS flashing

As described earlier, the BIOS is a crucial component of your computer since it is always the first program that runs when you turn the machine on. If the BIOS does not run, the machine cannot boot. Therefore it is very important that a BIOS upgrade is performed properly. If it fails, it can permanently ruin the BIOS, resulting in your computer becoming a door stop. Common reasons for BIOS flashes gone bad are power failure during the flashing process, and flashing the BIOS with an incorrect BIOS version. Because of these dangers you do not flash a BIOS just for shits and giggles, but only if there is a very good reason.

How to fix a ruined BIOS chip

If a flash has gone horribly wrong and the computer fails to boot altogether, the only solution is to replace the BIOS chip on the motherboard with a new one that contains a healthy and correct BIOS. To obtain a new BIOS chip, contact your motherboard manufacturer and give them the model of your motherboard. For a relatively small price ($10-15 plus shipping) they may mail you a new BIOS chip. To install the new chip, you open up the computer after taking appropriate anti-static precautions, identify the BIOS chip on the motherboard with help of the manual, carefully pull the old chip out of its socket, and insert the new chip in its place. Be sure to carefully follow the manufacturer's installation instructions.

However, be aware of the fact that some manufacturers will not sell individual BIOS chips, and some older motherboards have non-replacable BIOS chips. In those two cases you will be forced to replace your motherboard, which is expensive and somewhat of a hassle.

While BIOS flashing requires some care and preparation, it is a relatively simple process if done right. Let's go over the steps of properly flashing a BIOS using a real-world example, the motherboard BIOS in my test machine.

Note: Please keep in mind that the following steps are an example to help you understand how the process works. However, your motherboard and BIOS version are most likely different from mine and you will have to adjust your steps, download files, and command line switches accordingly. Please exercise caution when flashing your BIOS and stop and ask for help if you are not certain.

1 - Identify your motherboard and BIOS

The first step is to find out what motherboard and BIOS you have. We have a comprehensive article on how to identify your motherboard in our how-to section, you can find the article here: http://www.pcnineoneone.com/howto/moboid1.html. It also covers information on identifying your BIOS.

While there are a lot of motherboard manufacturers, there are two main BIOS companies out there: AMI (American Megatrends) and Award. Most motherboards use one of these two BIOS types. Both BIOS types display the BIOS identifier number on the main BIOS screen you see when you start the computer.

Tip: Push the Pause/Break key on your keyboard after the BIOS screen appears. It will pause the BIOS program and give you time to look over the screen to identify and document the BIOS number. Press Enter to continue.

Tip: There is a program available that is able to identify the BIOS from within Windows. It is called BIOS Agent and can be downloaded here: http://www.esupport.com/bioswiz/



In the example used for this article the BIOS is an Award BIOS. The identifier string is located at the bottom and looks like this: 11/08/1999-i440BX-W83977-2A69KA1UC-PL

This gives us the following information:

* 11/08/1999 is the date of the BIOS
* i440BX indicates that the motherboard is using the Intel 440BX chipset
* 2A69K also identifies the chipset as an Intel 440BX/ZX
* A1 indicates that the motherboard was made by Abit
* UC identifies the board as the model BE-6
* PL is the BIOS version

Now that we know the motherboard model and the BIOS version, the next step is a visit to the Abit website to check for a later BIOS version.

2 - Download the updated BIOS file and flash utility

Sure enough, there have been three new BIOS releases for this particular motherboard. According to the release notes they fixed three bugs, improved IRQ assignment, and offered support for faster processors. The most recent BIOS version is TH.



Note: You do not have to flash every single update, only the most recent one. It contains all previous updates.

Download the BIOS file to the hard drive. It is a self-extracting executable. After double-clicking it extracts the actual BIOS BIN file, in our case called BE6_TH.BIN, which contains the new BIOS software.

In order to perform the flash, we also need the Award flash utility. This program performs the task of erasing the old BIOS from the chip and uploading the new BIOS. We download this file called awdflash.exe to the hard drive as well.

3 - Prepare a clean DOS boot disk

A BIOS flash needs to be performed from a true DOS environment. The only way to get a clean plain DOS environment is by booting from a DOS boot floppy. You can download a DrDOS boot disk from http://www.bootdisk.com/. Execute the downloaded file to create a DOS boot floppy. Please use a good quality new floppy disk to minimize the risk of the BIOS flash failing due to corrupted files on the floppy.

Note: Yes, there are now programs that supposedly allow flashing certain BIOS's from Windows. Yes, there are ways to perform the flash from the hard drive instead of floppy disk in certain situations. However, these situations do not apply to every single computer. The method of booting and flashing from a floppy disk described here is by far the most common one.

Once the DOS boot disk has been created, copy both the BIN file as well as the flash utility to the floppy disk.



Be sure to reboot with the disk at least once before the flash process to confirm the disk works properly.

4 - RTFM - Read the freakin' manual!

Before you go any further, make sure you read the manual, specifically the instructions for BIOS flashing. It will explain how to use the flash utility and all the command line switches it uses. It is important that you understand what command to run to perform a proper BIOS flash.

In this example I decided to use the following command after reading the manual:

AWDFLASH BE6_TH.BIN /PY BE6_PL.BIN /SY /CD /CP /CC

Here's what this command does in this case:

* Awdflash is the name of the utility
* BE6_TH.BIN is the name of the new BIOS BIN file to be uploaded
* /PY is the command to program the flash memory
* BE6_PL.BIN is the name of the file as which the old BIOS will be backed up to the floppy disk
* /SY is the command to back up the old BIOS to the floppy
* /CD is the command to clear the DMI data after programming
* /CP is the command to clear the Plug'n'Play data after programming
* /CC is the command to clear the CMOS data after programming

The reason I am clearing all the existing data is because I want to start with a nice clean slate and verify or configure all BIOS settings after the flash.

Again, read the instructions carefully, make sure you understand them, prepare the command line you will use for the flash process, and write it down on a piece of paper.

5 - Last preparations

As mentioned before, the most common reasons for a failed BIOS flash are uploading the wrong file and having power dips or failures during the flash causing the computer to reboot during the procedure, wrecking the BIOS for good.

You protect yourself against the incorrect BIN file by carefully double-checking your work, making sure you correctly identified your motherboard and BIOS, and downloaded the appropriate BIOS update file.

To make sure that a power failure doesn't ruin your BIOS, plug your computer into a UPS (Uninterruptable Power Supply) that will provide battery power in case of power failure. More on UPS in this article: http://www.pcnineoneone.com/howto/ups1.html

Reboot your machine and press the appropriate key to enter your BIOS. Typical keys to access the BIOS are (by manufacturer):

AMI - F1, Del, Ctrl-S
AST - Ctrl-Alt-Esc
Award - F1, F2, Del, Ctrl-Alt-Enter
Compaq - F10
Phoenix - F2, Ctrl-Alt-Esc, Ctrl-Alt-R, Ctrl-Alt-Ins, Ctrl-Alt-Q, Ctrl-Alt-F1, Ctrl-Alt-+, Ctrl-Alt--
IBM - PS1 Ctrl-Alt-I

Now go through every single page and carefully document the current BIOS settings. This step is very important because after the BIOS flash some or all of the settings will be gone and you will have to reconfigure it manually.

Tip: If you have a digital camera, use it to take pictures of each BIOS screen. It's a lot quicker than writing down each setting by hand.

While you're in here, take a look at the boot sequence and make sure the floppy disk drive is set as the first boot device.

6 - Performing the BIOS flash

Now that we're thoroughly prepared, it's time for the actual procedure.

Insert the boot floppy you prepared and restart your computer. You will end up at a DOS prompt for the floppy drive A:\.

At the A:\ prompt, type the command line you prepared earlier to run the flash utility and program the BIOS.



Since you already provided all the details via switches, the flash program should run without interruption. First it backs up the current BIOS to a file on the floppy disk.



hen it programs the BIOS chip with the new BIOS.


Done!

The process is very quick and uneventful. It took less than a minute to flash this BIOS, no user input required, and it went smoothly.

Now it is time to reboot. Watching the screen after the reboot we can now see the updated BIOS version TH in the BIOS identifier string at the bottom.

Immediately go into the BIOS after the reboot, do not let the operating system boot yet. In the BIOS, go through each screen and configure it with the proper settings that you documented earlier. Double-check all the settings for accuracy. Misconfiguring the BIOS can lead to system instability or hardware failure, so be thorough.

Once the BIOS is configured properly, reboot one more time, and this time let the operating system boot as usual. Everything should come up normally. It is possible that Windows might announce that it found new hardware and starts installing drivers. That's ok, it is a result of the Plug'n'Play data being cleared during the BIOS flash. Since all your hardware was already installed before the flash and Windows has all the drivers, it should automatically install them without incident. You might be prompted for one last reboot after this is done.

You've successfully flashed your BIOS

As you can see, flashing the BIOS is not as scary or mysterious as it may sound. It is a process consisting of a simple series of logical steps. Performing these steps with care, patience, and preparation will minimize the risks of the procedure.

source:BIOS Flashing Guide

one question: should we simplify the name of PCI-express, we should put it PCI-E or PCI-X(which shuttle xpc system advertisement use pci-x for its pci-express)?

http://www.wi-fi.org/OpenSection/glossary.asp?TID=2

source:Wi-Fi Alliance
i duno wheter it is correct onot to post here but i cant find a place in Network,Broadband thread to post it up...so plz move it to if mod feel not suitable...
thanks in advanced

Expansion Card Explained (From PCI up to PCI-e)

With the assumption the same amount of transistors are used. If one goes lower nm fabrication, it allows more tansistors to be added too in the quest for higher speed stably (Utilising multi-pliers). But this will cause an increase in temepratures.

the difference between SATA and PATA

PATA


PATA has been the bog standard for hard drive interfaces, it transfer information by mean of parrallel method, meaning it sends out multiple signals at once, each carrying a small amount of data. the problem with the PATA method is that as signals travel faster (signal becomes shorter) the problem is that the multiple wire engage in a banter called crosstalk, meaning the electromagnetic energy jumps from here to there distorting data, that's why we see, ATA 133 as the limit, because if we had to put in anymore grounding wires we'd just go mad with fat cables.

-uses a much broader connector, around 2 and half inches wide
-has 80 connector wires (for ATA 100-133) and 40 (for ATA-66) you might wanna use fatter cables for better transmission and shielded ones if your PC is an interference powerhouse
-Has jumpers near the connector (you need to assign a drive the designation master and slave) there can only be one MASTER and one SLAVE
-Usually abit cheaper than SATA models
-Compatible will almost all PC's made in the past 15 years
-Requires no special drivers to make it work

SATA

SATA is a relatively new standard worked up by major hard disk companies like seagate, western digital, and intel. it was designed to combat the major flaws of PATA by utilizing newer and better technologies, like tighter CRC, an optimised buffer system and more. first the fact that it's serial based kill two birds with one stone, firstly the crosstalk problem it transfers data one bit at a time, but does it far more times per cycle to make up for the lack of multiple wires, it also means those ugly grounding wires and all can go, meaning a neater, smaller and easier to install connector. SATA is also hot pluggable, that means you can take you SATA hdd out of the computer even when it's running. and the master and slave thing, forget about it!

-Small Connector
-Has transfer rates of up to 300MBps utilizing the same good ol cabling
-No master of slave (emancipation act mah!)
-Hot pluggability
-Easier to manage cables
-Not so good with extreme overclockers, the SATA serial system is quite sensitive to FSB fluctuations
-Requires Drivers to unlock it's RAID potential (for ICH5R), Requires drivers to even use the drives (non Intel Based boards, AMD chipsets, may vary)
-Older PCs might not have a SATA connector
-Requires the use of a SATA power socket for 100% native units (seagate)

NOTE; Most SATA hard disks these days are just PATA hard drives given bridge chips (meaning the originally talk in PATA, but it's converted to SATA talk) these kinds of drives are called non-native drives, but don't look down on them, they are backward compatible with the 4pin MOLEX and their performance is only slightly lower, note the world's faster non-enterprise drive the Raptor is one such drive. if you mant native drives, seagate has the best of the bunch (the only bunch infact) and those are quite the performers themselves.

Hope this guide helps, and send you money to me, and you first born daughter or son while you're at it.

Empire23's guide to hard disk and scuhzy madness, plus a load of bull shiate about HDD's

today i'll just post on the more advanced and complex parts of a hard disk because i'm a big showoff, remember i'm not here to help you

Spindle Speed! Take me to warp baby!

ok let's imagine you were a static object, and that things came to you, that means the slower it is, the slower it gets to you and you can't move a bloody inch! same with hard drive arms ( the read and write thinggy ), so the obvious answer is to make the stuff come to you faster. Same applies to HDDs except you are the arm and the platter has the stuff on it. Consumer HDD's come on Flavours of 5400 RPM , 7200 RPM and 10000 RPM. simply here the faster the better, but let's not jump on the speed bandwagon yet, remember the IBM Desk *Death* star a few years back? they jumped on the 7200 RPM bandwagon a little too early and their drives died on them, same thing with the first batches of the 10000 RPM Raptors. Speed is a good thing if backed by sound engineering.

Stripe and Strip are the same, but clusters aren't!

What i can say here is that the mininmum space any amount of imformation can occupy is a cluster, set you cluter size to 4k and if you make a 1k size file, it'll still end up taking 4k of space. Stripe size is the amount of data stored in each stripped portion of a RAID array. but it's all down to what you use your computer for, does your pc like it with a cup or a whole bloody troth? file server would like the troth since files are mostly big and chunky or a database and E-commerce serve which only manipuates small propotions of data at a time, preferring a small tea cup with a hint of sugar. but be warned that after you pass the limit of 4kb per cluster, you can't compress or defrag your file, nothing big if you're running a server though, since defraggin is quite futile as data movement is constant/

All that spins has bearing! it's all about those bearings

As most things spinning at high speeds need bearing, so do hard disks. out there are 2 kinds of bearings in general circulation, first the tried and true ball bearings and the oil based FDB or Fluid Dynamic Bearings. Ball bearings are good and all, but as we all know they can be abit on the noisy and friction prone side, since the the balls themselves are relatively unlubricated and try to present a minimum profile to the spinning object, making them quite the little heat makers. up the scale is the FDB, which is bearing based on a thin coating of oil, and since the liquid conducts heat far better and has far lesser friction than ball bearing, they often lead in most departments like performance, sound and reliability, where they lose out is cost, they cost a pretty penny to make.

Queue jumping was never this sweet! all about queue's and how to jump em!

Just imagine of you were in some lift that prioritized the floor is went to in the order they were presses, sounds like serious bull and filled with the pain of waiting right? yep, it's that bad in hard drives. to to speed things up based on efficiency and urgency of data, some eggheads cooked up two great new was of sorting commands, TCQ or Tagged Command Queueing and NCQ or Native Command Queueing. Remember boys and girls that amount of crap you have stacked up is called a queue stack. NCQ capable hard disks are what most people should be looking at since it's incorporated with most motherboards shipping today, it's not really evident at first but try loading your hard disk with queues of 16 and deeper and you'll see the benefit. TCQ is a SCSI or Scuhzee thing, but if you use Western Digital's top of the line Raptor drives, it'll come in handy, but it's only supported by a handful of addon cards, so check you specs.

If your brain had the density of gas you wouldn't be reading this shite. Density baby! and how it affects the hard disk!

Just imagine if you had marbles...no dammit...not those. Anyways back to the marble part, imagine you had a whole bunch of them in your hands, the marbles are all in your hand, and you count them, you have 40, now put em in a basket, now get another forty marbles go to Carrefour and put 1 in a parking space until all 40 marbles are used up. which one is easier? the damned basket ot trudging through the whole carpark, if you answered the latter, call me so i can kill you. the point is the if the amount of data can be compressed into an even smaller space the distance the mechanical parts would need to move would be less! so in turn faster access time and higher overall performance, in tech terms we normally term it as, gigs per platter, take the total size and divide by the amount of platers, easy isn't it.....i should be given a nobel prize for comming up with this dammit. Unlike in humans the denser the better!

Buffer, it aint have nothing to do with buffet.....
now, let's get with the imagine (i'm a real barney the dinosaure whore, so sue me), imagine yourself shitting, and as you shat, the water kept flowing to flush your turds down...damn all that piss weak flowing water can't flush shit! literally, so that's where the tank AKA tangki simbah comes in, it gives water as you need it (imagine got many variable flush options), so when you need the water it's there and waiting, the water is the data, the turds are the processor, so the turds only need a certain amount of water to flush them down, small shittings only require a little water, same with data, complex instructions with little stored information requires little data. so the hard disk buffer fills itself up anticipating the next request with prediction algoritms, ready to deliver data at top speed instead of going back to the disk and asking those damn motors to spin and those read heads to go woppy. as always, bigger is better, the bigger the buffer, the better the HDD can store anticipated data and less time spent on starting back up to refill the buffer. 8 megabytes is the standard at this time, although higher end drive like the MAXLINE are sporting 16mb buffers. Buffers can also be called caches, both mutually interchangeble.

PS; i again must assert ownership of you and your children by reading this, an feel free to correct me, coz i'm currently half awake and naked, plus i damn well don't know what the heck i'm writing about.

MOTHER BOARDS A-Z BY EMPIRE23! WHEE

AMD or Intel? Make the leap of faith right here!

As for the processors themselves i'll leave it up to you and is beyong the aspect of this article, but make your choice wisely. And remember that certain version of processors with different pin configurations imparts different features, due to more pins being required to transfer more information at the same time. For example, Athlon 64 boards with 754 pins do not support Dual Channel memory transfer, unlike the 939 pinned based bigger brothers. So choose your processor wisely young padawan.

Elephants aren't the only ones with huge memories

Simply said, buy the memory that fits your board, because different boards use different memory. But it can't be that hard can it, there are only 2 standards out there DDR1 and DDR2, i'm sure you won't get confused because DDR will only fit in DDR1 slots and DDR2 will only fit in DDR2 slots, of course the latency of the memory will affect performance, and different chipsets like different settings, go do a benchie do find out man!

Form factor, personally i like em cute and small, but the smaller they are the less well endowed they come

When it comes to mother boards Size does matter, with extra size you can jam pack more features into it. But this is more about form factors than features so let's get to it.

ATX- Bog standard the great balance between the giant server form factors and the small ATX-MINI form factor, generally ATX is a standard that specifies the size of a motherboard and various clearings. And that means if a board or casing is ATX certified, it'll fit. ATX-MINI is for smaller and less poverful pcs and the boards based on this standard usualy have lower specifications and performance.

BTX- Next generation form factor proposed my intel to manage various heat problems and optimize air flow, but this time the casing opens from the opposite side, and the processor is located at the cold air intake, and turbulance zones are eliminated to air flow is clean and hot air isn't recycled. Might solve your preshot problems, but then again mother boards for this form factor aren't even here yet, but a few manufacturers are starting to sell ATX and BTX hybrid cases.

Chipsets man, personally i prefer mister potato and nvidia.

There are many chipsets out there with different pros and cons, so weigh each one with objectivity, not just the one with the flashing lights and biggest bewbies. Each chipset can be boiled down to different points which i will tackle later. Let's just say that some chipsets are superior that others, and to me the superior one is the one that can contain all it's crap in the smallest single package, because, no only does it save space, but latency, power consumption and such. And the only single chip solutions out there built by Nvidia, as in the Nforce Series. The rest just use the tried and true North and South bridge method, at a small imparting of performance due to fact that any device connected to the south bridge must take a 2 fold journey to the land of EAX.

PCI-EXPRESS or AGP, personally i prefer PCI-EXPRESS, the expess makes it sound faster...heck IT IS!

AGP is standard as old as the chewing gum stuck under the seat at the NRD and utilizes the good ol' parrallel method transfer to get it's way. It come in flavours of 2x, 4x and 8x in which 4x and 8x cards aren't backward compatible with 2 x due to voltage difference.
Taking over the helm of High speed graphics bus is PCI-express, that beats AGP hands down with twice the amount of bandwidth via it's serial and bi-directiona method of transfer. But remember the bandwidth used by graphics cards cant even saturate an AGP 4x bus, so saturation of a 16x bus or even an 8x of the PCI-express variety is unlikely at best and does no impart any extra performance. But for the sake of future proofing i recommend getting a PCIE based motherboard, spend now, less regret later.

SLI, why have one wife when you can have 2 or even 4!

Currently Nvidia is extolting the virtues of SLI, what the heck is SLI you ask? boleh makan ah? No silly, simply put it, what happens if you think that you're graphics card ain't up to the challenge? Simply with SLI you add in another one! But both the cards must be SLI compatible and of the same brand, make and yadda yadda yadda. Currently only the NFORCE4 chipset supports SLI. But it seems that ATI has something up it's sleeve, called Crossfire, yes it sounds pretty, but is it worth the money? if you ask me. Yes, it seems that Crossfire will work with ALL programs unlike SLI, ditches that ugly connector that SLI uses and you only need to have ONE crossfire card, the other can be any other ATI card on the PCIE bus. Hows about 4 grahics cards? Gigabyte is looking into it, so keep your undies on.

IT'S A RAID!! surrender your women and none will be hurt!

RAID is a relatively old technology to the computer word but has recently been given some atteantion by motherboard makers. Since most motheboard makers already support spanning RAID over mutiple interfaces, we'll leave that to the TS guys. But as i always say the more RAID levels the better but i'll give you the skinny on RAID

-RAID 0, fake RAID because it isn't redundant, data is split between the 2 hard disk to make reading faster, but if one drive fails, your goose is cooked
-RAID 1, total RAID, both disks have an exact copy of each file right down to the last bit, so if one fails, your goose isn't so cooked, but then again diskspace is wasted, you only get 50 percent of your true capacity
-RAID 5, this level of raid always wastes around 10-25 percent of your capacity due to the fact it creates a parity to restore data, but this parity calculation is at the expense of CPU processor cycles, a real resouce muncher, but combines all the powers of lower level raids
-RAID 10, a combination of RAID levels 1 and 0, where half of each hard disk is partitioned to a different level, one half is stripped and the other is mirrod and space waste is around 25 percent only, but then again you get no parity protection

Everybody loves extras!

Normally motherboard manufacturers throw in extras to make us buy their shit, and that's not a bad thing at all. I'll just give you the quickie for extras around here.
-Extra Hard Disk Controllers, does what is says, allows you to attach more hard disks!
-Firewire, Ouch, HOT! This connection should be the choice of ipod owners and all video editing geeks around.....comes in flavours of A and B, the latter being twice as fast
-WIFI, the standard for WIfe InFIdelity...uhh....really! Some manufacturers thrown in WIFI cards tied into the mainboard's bus, makes your pc wireless
-BlueTooth, kalau this also tak tahu, pergi lompat KLCC.

And always to research on what bus the extra is tied into, since the PCI bus has a combined limit of 133 MBps and PCIE a much higher 266 for each 1x interconnection, so if you feel the your extra HDD controller is abit slow the PCI bus might be the culprit

Sometimes noisy IS a good thing!

Not all motherboard soundcards are created equal my friend! From dingbatted brain's analysis the current winner has to be the Nforce SoundStorm, but has now sadly been cancled by Nvidia due to Dolby licencing costs. the next best thing should be Intel's HD audio standard or known to me as dear AZALIA! But some mobo makers intent on smacking others with my-epenis is bigger than yours have implemented certain solutions into their mother boards, DFI and ABIT have implemented audio via daughter cards, and MSI has their audio produced by a Creative Labs chip. And ofcourse you'll have to look at the quality of the DAC, because not matter how good the DSP (main chip), if you have a crappy DAC (converter) you'll get crappy sound. live loud my friends and may the music kill you!

The words "Computer" and "Hot" do not mix well my friend...

Cooling is normally only needed on the Northbridge (Giant chip besides your Processor) of most motherboard. Cooling quality can be generally judged by 3 factors, Size of the heat sink, Fan Speed and thermal interface, the latter of the 3 you can help with general ease by replacing the original thermal paste with high quality silver addative based ones. Cooling can also take place on other area on the motherboard which are hot spot, which are mostly the MOSFETS and VRMs near the processor socket, some better versions of normal mother boards sport extra coolers for these part, and althought their use for non overclockers is questionable, for rabid ocers like me, they're a godsent.

the BIOS, where there is basic input there is output

I'll keep this short and sweet, bioses that come in removable socket packagings are far better than soldered in ones, because incase the pc freaks when you flash the BIOS, you can hot flash it and your mobo will be saved. Some mobos even come with 2 BIOSes so incase the first dies, the next will take place and reflash the save the original's ass by allowing you the chance to reflash the BIOS. Note also if given the chance, always format your BIOS to the latest version, updates often cure many headaches. and some bioses come with the magic port 88 display, making your life so much simple when it comes to diagnosis, it look like an LCD with 2 eights on it.

As in real life the cleaner the better!

when power is filtered and regulated each step it goes in a cycle the more stable and peak free the power gets, so in theory the more stages the better, only a qualified computer geek can tell you how many power stages a mother board has at any current moment. But i can sure as hell tell you that gigabyte with their Dual Power System card have 8 staged power, and to supply even cleaner power, newer mother board come equipped with the brand spanking new ATX 2.0 power connector with 24 pins instead of 20 to supply more stable juice, if given the chance, always use the 24 pin connector, and try so without the converter, because the converter kind of defeats the purpose of having dedicated power and not shared power with other lines.

I feel we should have a latest update on this . Anyway THANKS for this guide. This is very informative. And also we should have a section on value for money buy from time to time. Thanks again.

More readings for everyone

Graphic AA and AF explained:

http://www.bit-tech.net/hardware/2005/07/0...iltering/1.html

Don't say that your problem is urgent. It might be urgent to you, but that's irrelevant to the question itself. You aren't paying for the service so you don't get to set the priorities.

is the port for hdd, pata=ide?
can have sata as master and ide as slave?

Hello all,

Arrr,... want to ask a question. I am just computer idiot here! Today, one of my Harddisk malfunction. Can it be repair? When I power on the computer. I can heard a sound like "ti dat, ti dat, ti dat"? So, what is wrong with the harddisk?!

By the way, is the the right place to ask question for my case? sorry if me make mistaken here!

Thanks

want to ask ,external hdd is plug to usb hub ?and ...hdd oso got speed 1 are ?????
my friend say the hdd got duno what speed when he use external hdd files from cc hdd ..........i thought its was the usb speed ,right ??

I know its apple and orange,but can I know for normal appllcation like
running MS Office,watching movies,burning CD,and some games(not graphic intensive ), which is better.

1 . Intel Celeron 1GHz
2 . Intel Pentium 3 600MHz

Why is Intel comes out with different socket just for P3 system ?
I know the design is different but Intel could have design ALL P3 and P4
to use the same socket so as not to confuse non-technical ppl.

Is it marketting strategy and\or pricing strategy ?

Thanks

good info..thx a lot

just came across this and thought i amy as well reply..

sata and ide use different cables and plugs. They are entirely different interface. You can't plug a sata into ide and vice versa.

all sata are master. there are no master and slave in sata.

and since the ide has nothing to do with your sata harddisks, you may choose to set it to master or slave. or course choose master laaa

if you are planning to put 4 or 5 hard disk into ur comp, make sure u got enough power and resources

hi guys wanna ask........

when searching for a new mobo.......which wan to be priority?........the clock or the FSB?

0. COST <--- most important b4 u decide to buy anything
1. Chipset
2. PWM (Pulse Width Modulator)
3. MOSFET
4. Capacitor
5. PCB layers
6. BIOS Options
7. Warranty Issue

dude....thx for the guide....

but when i saw the price list .......it showed the value of PCB n clock, beside the price n product........

so that s y im asking......

i think my ques still not answered........

but charge.....thx

No probs

actually the clock shows the FSB. Faster means better, however nowadays most of the board has a base clock of 200MHz, so u cant realy differentiate them by using FSB only.

For PWM, MOSFET & Capacitors, you can actually see them on the spot and check if they are good or not. BIOS options can be found in online review, or even online manual from the motherboard websites.

Hope i can clear yr doubts

ya talking about caps, how do you differentiate taiwanese and japanese caps from the ciplak caps?

can anyone confirm this

i read somewhere b4 for AGP.. example AGP8X, the write to VGA memory is full 2GB but write to system memory(RAM) 1/10th of that or around 250MB..

they did a benchmark for that.. but i duno where izzit oledi..

what i know is PCIE is full bidirectional

How do you differentiate PWM, MOSFET & Capacitors of good and bad qualities?

AFAIK, HIP6301 is a very high quality PWM, not sure if any good one available nowadays.

MOSFET I'm not sure about the brand, general guideline is to choose a higher Ampere rating MOSFET (80A and above). Higher rating one allows can handle higher current and has less internal resistance, thus wont heat up that easily.

Good capacitors are those japanese one, such as Sanyo, Rubycon, KGZ and etc.. Not very sure about the brand though, hahaha. Higher capacitance value (usually 3000-3500nF) is better. Besides, there's also a voltage rating, higher means the capacitor able to operate with higher voltage.

There's another important factor : The power phase. Usually a motherboard has 3 phase power. Board with 4 phase power like the DFI Expert series can output smoother and more stable voltage. Well, this doesnt mean 4 phase is better than 3 phase too, bcoz we must see if the components in each phase is good enough. FYI, each phase consist of MOSFET and some inductor in a group. A 4 phase power system with crappy MOSFET perform worse than a 3 phase power with good MOSFET.

Actually you guys can js go google and search for some good info, I'm js providing some guideline about how to determine a good motherboard.



Jinaun, AGP 8X should be able to transmit up to 2.133GB/s theoretically --> able to write to system RAM at 2.13GB/s. Local VRAM write speed depends on the VRAM speed and the GPU memory interface. It has nothing to do with AGP8x or PCI-e 16x

well.. i read abt it somewhere... its regarding writing to system memory through vga card is 1/10 slower... but system to vga is full speed... anyway.. i'll try to find the article..

by the way.. here is a VAR benchmark for my AGP card

=====================================================
C:\VARMemBench.exe sysmem
Allocating system memory array...OK
Running system memory benchmark...Done
System memory set speed: 1968 MB/s


C:\VARMemBench.exe agpmem
Getting wglAllocateMemoryNV proc address...OK
Getting wglFreeMemoryNV proc address...OK
Allocating AGP memory array...OK
Running AGP memory benchmark...Done
AGP Memory set speed: 3745 MB/s


C:\VARMemBench.exe vidmem
Getting wglAllocateMemoryNV proc address...OK
Getting wglFreeMemoryNV proc address...OK
Allocating video memory array...OK
Running video memory benchmark...Done
Video memory set speed: 3748 MB/s

============================================

here is the link to the site... http://cwdohnal.home.mindspring.com/var_me..._mem_bench.html

hmm.. wierd values... how come when go across the AGP bus.. its over 3.5GB?

ah.. i see.. thanks lots. so i guess its worth to pay rm 200 more to go for DFI infinity ultra-D instead of modifying DFI infinity ultra into a 'fake' DFI infinity ultra-D
the capacitors are different.... and there is also an added fan on the chipset if i have not mistaken. NF4 chipsets are hot.

The values seemed ok. AGP memory = Video memory in this case. They are the VRAM of your graphic card. System Memory refers to yr RAM.

As I mentioned earlier, theoretically the max System Memory access by AGP is 2.13GB/s, and we would never get the perfect value in real life. 1968MB/s is close to the theoretical value already.

Both 3748MB/s and 3745MB/s are very identical, hence they usually come from the same source connected with the same bus.


Joylay83, I didnt know the Ultra-D and Ultra has components difference LOL. Thanx for your info