Sign in with
Sign up | Sign in
Your question
sticky Closed

GIGABYTE Guide

a c 177 V Motherboard
June 17, 2010 2:54:13 AM

I realize this is kind of a long 'sticky', but look at it this way: it's considerably shorter than the manual you left in the box :non:  when you removed the board!

[:grahamlv] Congratulations! You've got a GIGABYTE board - and they are one of the best manufacturers in the business!

There are problems with virtually every manufactured product, and GIGABYTEs are not immune; the good news is that, for these boards, the 'problem set' is fairly small, mainly repetitive, and the solutions are mostly 'known'! There are also a few techniques that will help make starting up your GB a more pleasant (and successful!) experience...


The Basics[:isamuelson:7]

The very first thing you need to do is find GIGABYTE's 'Home Page' for your board. Start at the left side, under "Product List", select your CPU manufacturer and socket type; you will then get a listing of chipsets applicable to that socket; find yours, select it, and then select your board's part number. From this "Home Page' you will find GB's CPU and memory compatibilty lists, as well as the pages where GB will release new BIOS and drivers for your system - bookmark the last two!

Should you need to contact GB regarding an RMA...

GB tech support...

GB utility programs...

GB 'board numbering' conventions...

GB's 'frequently asked questions'...

GB Intel 1156/1366 and AMD 8xx motherboard iPad charging...

Most GB motherboards use RealTek audio and LAN chips; up-to-date drivers can be obtained here...

GB's X58 guides are here and here...

AMD's tuning guide...

I have posted a tutorial to 'capturing' and posting screen images...

LSD's "X58 Debug Codes" list from TweakTown...

Stasio's excellent (and well-maintained) "Latest Overclocking Programs, System Info, Benchmarking, & Stability Tools" also at TweakTown...

Last, but not least, TweakTown's comprehensive BIOS flashing guides...

BTW - TweakTown is a great resource for GB-related info!


[:fixitbil:6]

Do read through the 'stickies' already here at Tom's - they contain invaluable information:
Motherboards, general...
Memory, general...

and, especially, the 'Homebuilt Systems : New System Build Forum's various 'stickies':
'Boot/No Video'
'Step-by-step Guide to Building'
'Guide to Choosing Parts'
'Buying Guide'
and, 'How to Ask for Advice'


Do be careful with your socket, if you're installing an 1156/1366 (i3/i5/i7) CPU; don't remove the socket cover before the actual installation; don't lose the cover - should you need to RMA, they will not accept a board with no cover; do orient the processor correctly before attempting to 'place' it; do hold the CPU 'level' and parallel to the board as you put it in; don't drop anything on the uncovered socket, or touch the pins prior to the install - these sockets seem infinitely fragile, and 'bent pins' will cause troubles - see i3/i5/i7 'Missing' Memory below!

Don't resurrect 'dead' threads - if there hasn't been a post to a thread for three weeks or a month - let it 'lie in peace'!

Do include a list of all the components (manufacturer & part/model number) in your system, including USB devices plugged in, other than keyboard and rodent (see 'USB Woes' below...); Murphy's Law plainly states that the one component you forget to put in your list, will be the component causing your problem :(  !

Don't post new questions to this thread - would like to 'reserve' it for reporting 'sticky' errors, discussion of content, and suggestions for additional information to be added...

Do, once your problem is fixed, select either 'solved' or a 'best answer' - marks the thread as 'finished', and keeps 'the help' happier!

Definitely, don't post new questions to threads marked 'Solved'; odds are no one will ever look at them again !

Do put your full, accurate motherboard part number (i.e., "GA-EP45T-UD3LR"), as well as the revision number, if applicable, in your thread title... Revision numbers can be found here:



  • Setting Up/Starting Up:[:bilbat:8]


    Load Optimized
    You must 'start up' your board by doing the "Load Optimized Defaults" function from the BIOS - this is not an optional step! Each time your board starts up, the BIOS begins by running a block of discovery code: it sort of 'twiddles its fingers' to see how many fingers it has, as well as which 'hand' each is on...

    The BIOS' "Load" functions (Load optimized/load fail-safe) similarly do 'discovery', but they do a much more thorough job of it, querying every piece of hardware, and then 'setting' the dozens of parameters you will find in your BIOS set to 'auto'; until this is done, your board is pretty much 'brain-dead', and can not be expected to work correctly, if at all. Think of it this way: if you woke up one morning, and not only couldn't remember 'who you were', but also couldn't remember which finger was on which hand - you'd be in a world of hurt![:lectrocrew:7]

    The difference between the two: the 'Optimized' reads the hardware, and makes, well, optimistic assumptions about the settings - tries to set latencies low, and operation fast; the 'Fail-safe' makes pessimistic assumptions, sets the latencies high, and operation slow - in an attempt to get the board to run "no matter what!"


    BIOS Flashing
    [:lorbat:3]First order of business: DON'T use @BIOS! EVER! Inevitably someone will 'crop up', and say "But I've always used @BIOS, and never had a problem!" Usually, "always" translates to: "I got away with it twice!" I often liken this to playing Russian Roulette... You might get away with pulling that trigger once, and only hear a 'click'... You might get away with it a second time, too... But, by the third pull, the odds are starting to catch up with you - you keep on pulling that trigger, you WILL blow your brains out! The underlying problem would appear to be that, unlike the other BIOS flashing methods, an @BIOS flash gone awry can overwrite the BIOS' boot block, which is the piece of the BIOS that, among other things, is responsible for the 'dual BIOS recovery' function; i.e., if your BIOS gets trashed, but the boot block remains intact, the boot block 'checksums' the 'working copy' of the BIOS (in EEROM, that you can flash), finds out it's bad, 'reverts' to the 'backup copy' (in ROM, that you can't 'futz' with, short of a soldering iron!) and loads it, with a couple of 'flags' set, to remind it to tell you about the BIOS' problem, and offer you options to fix it... Once @BIOS trashes that boot block, your board is a brick - useful only to prop open a door (and - a real, actual brick does a better job of that, too...)!! You're pretty much hoping for a 'graceful' RMA :kaola:  This problem is not limited to GB; other manufacturers have 'in the OS' flashing setups, and, so far as I know, are also prone to difficulties that DOS or 'in-BIOS' flashing just don't have...

    There are a couple ways to look at the whole issue of BIOS upgrades (and peripheral firmware upgrades, and, for that matter, driver updates); there is the 'if it ain't broke, don't fix it' view, and there is the 'upgrade as a simple matter of maintenance'... It is always possible to get something bolixed up as a result of the 'latest and greatest' code, but, often, if you don't do the upgrades, you don't know what you're missing (and, a lot of times, after the upgrade, you still don't know what it was about...)! A good example of the possibilty of improvement versus screwing something new up is the recent (like about six months back) series of GB BIOS releases to accomodate the new Intel RAID BIOS code released for the ICH's, from which I found about a 4-6% throughput speedup; followed by the botched 8.9.1023 Intel driver/manager package, which randomly trashed RAIDs - ya' win some, ya' lose some!

    One of the contributing problems is that GB, like nearly everyone in the entire hardware channel, does a somewhat poor job of documenting changelogs; if you get a three word description, consider yourself lucky! You can usually figure out microcode upgrades, to support new processor steppings, by the BIOS revs versus CPU support list on their website; other issues - you're on your own! There is one current situation for which I recommend everyone 'hang on' each GB release, even if the documentation is inscrutable - the 1156/1366 platforms are really 'works in progress' - every BIOS rev is probably worth glomming onto, as they're conducting a 'fix of the week' program (kind of like RealTek's 'driver of the week club'...)

    My advice to most is to consider your experience level - if the idea of flashing firmware gives you cold sweats, and the heebie-jeebiesdon't do it, unless you know you have a problem that you know is addressed in a later GB release - and, DON'T use @BIOS! Also, don't erase a BIOS file itself - you never know if it'll be around to be found later, should you discover the flash wasn't all that great... If flashing is no big deal to you, and you're confident in your procedure, go ahead - if it's a 'dud', you can always 'flash back' to the last thing that worked well for you. Do get in the habit of backing up your CMOS parameters to removable media (see BIOS - Saving CMOS below) - it makes the whole flash & set-up process so much easier, and GB/Award hardly ever (I won't say 'never', but it's very rarely a problem) 'break' a BIOS so that CMOS tables saved in the last release won't load into the latest release...

    If a BIOS flash seems to fail with errors even though the flash appeared to work right, check the following:

  • Download it again and make sure the download was correct (how to do this is unknown without a checksum from Gigabyte - jusy 'recycle' and pray!)
  • --> Alternate source: StationDrivers (...polish up your high school French, s'il vous plaĆ®t!)
  • Check that the floppy is good
  • Ensure your floppy drive is in good working order (confirm the cable is connected correctly and the power cable is firmly attached)
  • Try again and check the checksum value reported. Confirm it's the same as the first time you tried and, if different, download it again and try a different floppy.


    BIOS - General Settings
    Some BIOS have 'hidden' or 'locked' settings; some even have whole menus 'locked'; to see these, do a <CTRL><F1> at your BIOS' main page; your screen will 'flicker', and the previously hidden goodies will appear!

    To make entering your BIOS easier, and to see all the information the POST (power-on self-test) displays, "Disable" the "Full Screen LOGO Show" item on the "Advanced BIOS Features" page...

    "Disable" the "Legacy USB storage detect" (later BIOS say "USB Storage Function") on the "Integrated Peripherals" page, unless, on your next boot, you intend to either access a USB from the BIOS, or boot to a USB - all other times - off! (see USB Woes below)

    BIOS and processors support 'green' features. These 'down-clock' and/or 'down-volt' parts of your system. On a notebook, these features are a godsend - they cut battery useage to a minimum; if you're trying to overclock, or having memory stability issues - not so good! You don't really wan't something 'wobbling around' your clocking :heink:  To kill these, on Intel boards, disable:

    On the "Advanced BIOS Features" page:
  • "CPU Enhanced Halt (C1E)"
  • "C2/C2E State Support"
  • "C4/C4E State Support"
  • "CPU EIST Function"

    On the "MB Intelligent Tweaker(M.I.T.)" page:
  • "C.I.A.2"

    For AMDs, these functions are 'lumped under' "Cool & Quiet"...

    For ease of startup, and to 'localize' problems, I recommend setting "ACPI Suspend Type" to "S1(POS)", on the "Power Management Setup" page, to begin with; you can shift to S3 later, but some boards will have difficulty 'returning' from S3 when overclocked; you want to get your OC working first, then deal with S3 issues, should you feel you need S3...


    BIOS - Saving CMOS
    A new skill involving the BIOS: do the <DEL> at the boot to enter the BIOS; notice, at the bottom, the <F11> "Save CMOS to BIOS" - hit this, and you should get a menu that will show a number (the count varies by BIOS) of empty 'slots', each of which will store an entire set of BIOS parameters, to be re-loaded from the corresponding <F12> "Load CMOS from BIOS"; this is a wonderful overclocker's feature. What I do with it, is to save my 'baseline' working parameters, so if I change something that 'irritates' the board, and forces a reset of all the parameters to defaults, or, even worse, get so screwed up I need to do a 'clear CMOS', I can get back to my starting point with no effort, and without having to remember 85 separate settings! Another thing it prevents is two hours' troubleshooting, having forgotten a change to a crucial parameter - like, "wait a minute - didn't I have the Trd at seven?!" It's pretty self-explanatory, and I alway urge people to start right away by taking the time to give the 'slots' names that mean something: in two hours, "Try2" and "Try3" will not be very helpful, but "450@+10MCH" and "450@+15MCH" will! Another use is for 'green' settings; overclocks, as a rule, do not 'play well' with green features, such as 'down-clocking' and 'down-volting'; with the storage slots, you can set up one profile, say "Green", with all the settings at 'stock' values, and all the 'green' features enabled; another, say "Balls2Wall" with a full overclock, and all the 'green' stuff turned off... Another neat feature of this 'slot' system is, for most BIOS, the mechanism itself will keep track of which ones have booted successfully, and how many times (up to, I believe, a max of five)...

    If you look at it carefully, you will also see this storage feature has a 'mechanism' for saving these parameters elsewhere - either to a floppy or a flash drive. This is a smart thing to do before flashing a new BIOS (which will 'erase' your stored profiles), or for general 'disaster recovery'!


    Sorting Out SATA
    Unless you have one of the 'cheapies' less-expensive boards, your GB board likely has several pairs of SATA ports, in two colors: purple and yellow, white and blue. The more numerous set of these is connected to your board's southbridge; the others are connected to either jMicron two-port controller chips (which GB calls 'GSATA' ports in their BIOS and documentation), or, in the case of the new 'A' suffix boards, a Marvell two-port SATA3 controller... I recommend 'using up' your southbridge controller ports first, for your hard drives - until you 'run out' of 'em! Four reasons: first, stability - the drivers for these are older (i.e., have had more revisions) and more 'robust'; second, 'coupling' - being in the southbridge, they can't help but be more 'intimately' connected to the processor; all other controller chips must be 'hung off' the southbridge's PCIe bus, thereby introducing another 'layer' of 'translation' for your data streams; third, speed; and fourth, SPEED! [:lorbat:9] Even the Marvell SATA3 is no match for the venerable Intel ICH series: some benchmarks...


    I usually tell people to put SATA DVDs on the jMicron (GSATA) ports to keep them from 'using up' a southbridge port; Esata is also a good use for the GSATA ports; they do AHCI pretty well, it's RAID where they stumble a bit... If you want to be able to 'hot-plug' (i.e., plug the drive in while the system is already booted, and have them recognized), you'll need to enable AHCI mode on the GSATA - for most boards:
    On the "Integrated Peripherals" page of the BIOS
    "Onboard SATA/IDE Ctrl Mode" to "AHCI"
    and then you need to put in the AHCI 'pre-load' drivers - this is easiest done during the OS install - it's a nasty, more complicated job to add 'em to an already installed system... If you are loading Seven, the southbridge AHCI drivers are now 'native' - you won't need to put in the 'pre-load' drivers. As for AHCI itself, I did some testing, and could not see any realworld difference; but, all my testing was on a single drive (Seagate 1.5), was done on the jMicron ports, was done before they 'fixed' the firmware, and was nowhere near comprehensive (didn't know about ATTO back then); mostly did it out of simple curiosity! AHCI enables the drive's NCQ:

    and some more thorough AHCI performance figures...

    Again, for my money, what I get out of AHCI is 'hot-plug' for eSATA; results may also be better for the ICH driver, as Intel appears to have highly optimized it... The 'remove' option is to make sure the in-memory disk cache is 'flushed' to the disk, before the connection is physically 'broken'.

    When the MOBO starts POST, the BIOS does a chunk of code called 'discovery' (see Load Optimized above); in this code, it sort of 'wiggles its fingers', and counts how many there are... It sees how much RAM it has, and how it's organized (the screen says 'testing', but it's not really - it's just 'ennumerating'!); if it has anything on any of its PCI or PCIe busses, and how they're hooked up; whether it has any drives, and how many, and what type and organization...

    If AHCI and RAID are disabled, the drive discovery is done early, in IDE mode (for compatibilty, even if they're SATA drives), and even though there is no 'master/slave' mode in SATA, the 'found' drives will be shown on the standard POST screen as "channel#/Master" and "channel#/Slave"; ignore the master/slave designations, they mean nothing on actual SATA controllers, and are some kind of 'compatibilty vestige'! If the controller is configured in either AHCI or RAID, this 'discovery' process is not done in the standard (IDE) POST - because you can't have it both ways - if the drives were discovered in both POST (IDE) mode, and AHCI or RAID mode - the board would 'think' it has two 'hook-ups' to each drive! So, after the standard POST runs, the 'discovery' done by the 'Option ROM' is run... This, also, will 'ennumerate' the drives. It usually occurs after a 'screen clear', and will display, first, the Option ROM version information, then the 'discovered' drives, along with their 'type', and then a "Press <some key> to configure..." message. For RAIDs, this discovery is kind of slow, as the RAID BIOS must verify the physical status of the actual stripe(s)/mirror(s)...

    You may not be seeing some of this if you have the boot 'splash' screen enabled (see BIOS - General Settings above); disabling this will allow you to 'see' more of what's going on...

    You want your boot drive on SATA2_0 (depending on the board type, these may be labeled SATAII_0, etc.), any remaining drives on consecutively ordered SATA2_x channels (i.e., the drive you want to be 'D:' on SATA2_1, the drive you want to be 'E:' on SATA2_2, etc.); but, the whole drive organization thing doesn't matter much these days - you can go into your OS and 'designate' any drives you want to be any drive letter, I just install them 'physically' in the order that I want them to be 'logically' to keep life simpler; two caveats: I don't think you can change the letter of your system drive, once your system is installed; and, with Xp, if you install to a second physical partition/drive, the damned thing will wind up as 'D:' forever [:bilbat:7] more...

    Another often confused issue: the "Native Mode" setting for the SATA controller; unless you are one of the six people who ever actually bought Windoze ME, you always want this setting at "Enabled" - it turns on interrupt sharing for the controller, and ME was the last version that did not support this...

    One more pointer: if you are using MOLEX to SATA power adapters, beware! They are prone to problems - flex 'em one way, work fine - flex 'em a different way (or, more infuriatingly, let 'em 'relax' into a new position for an hour), poof - drive gone! If you plan to use 'em, never a bad idea to order a few extra!


    RAID & 'FakeRAID': Speed vs Data Security
    The basics:

    RAID0 'stripes' your data across the drives (and, with the Intel ICHs, two drives is optimal - more than two degrades performance - I've tried it...), allowing it to read/write two or more ports at a time... While this is fast, there is no redundancy - any single drive failure will take out your data!

    RAID1 'replicates'[:lectrocrew:2]your data on a pair of drives; while writes occur at about the same rate as using a single drive, reads are about as fast as a zero, as you can, once again, read off two channels...


    RAID10 (sometimes called '0+1') is a combination of both; while you get the speed enhancement of zero, you get the data protection of one, at the cost of half your drive space... RAID10 can use more than four drives, but the number of drives must always be even.

    The rest of the RAID levels are best implemented on an actual RAID card, as they involve parity calculations for every byte of data written; RAID5 uses a single parity stripe for data protection; RAID6 writes two parity stripes, which gives you protection (typically in larger arrays) from two drives failing at once; in addition, RAID6 allows you to designate one or more 'hot spares', which will automatically be 'transitioned' into the array upon a failure...

    The linux guys have the most apt name for any RAID you can do with the hardware on your motherboard - they lump it all under 'FakeRAID': what's happening is that the CPU has to do all the work of both running the SAS protocol stack, and the parity calculations (remember - for every byte of data written), in the case of RAID5 - that's why the ICHs are notorious for being really slooowww[:fixitbil:5] doing RAID5. Cheap (say, less than $400?) RAID cards do the same thing, and that's why drives on them can generally be 'discovered' by windoze without a driver, unless they are RAIDed - then they, too, require the 'F6' driver procedure. On pricier RAID cards, what's happening is a dedicated processor (an SiS or Intel IOP341) is running the protocol stack and the parity calcs, which 'frees' the main CPU from the burden. When you start getting way up there in price, you might even get an Intel IOP342, or 348, with dual cores - one running the SAS stack, and the other dedicated to the parity calcs - and the throughput goes through the roof!

    When RAIDing, there is a 'chunk' of RAID chip/chipset BIOS (called an 'Option ROM') added to the system BIOS, which is only activated when you have enabled RAID for the ports in the main BIOS, and the chip/chipset detects at least two drives; this RAID BIOS is where you 'build' the RAIDs. Once this RAID BIOS 'takes over' the drives, they are no longer 'discovered' by the main BIOS - this is why the drives 'disappear' off the standard POST screen. After you have created the RAID volumes, during the OS install when you are selecting the drive and partition, there is a 'load drivers' item (<F6> prompt for Xp) - you need to do that, with the 'pre-load' drivers for the selected chip/chipset - while the 'run-time manager' stuff 'watches' the RAIDs, detects any inconsistencies or problems, and repairs them (if possible) 'on-the-fly'... For the procedure to 'generate' a 'pre-load' driver disk for your selected RAID install, look in the index to your manual, toward the end, for an item called "Making a SATA RAID/AHCI Driver Diskette"...


    IDE Cabling/Jumpering
    Here's an IDE cable:

    The 'M' end is the motherboard connector; the 'D' end goes to the drives.

    There are two kinds of cables: 'standard' (on which the drives are jumpered to identify them), and 'cable select' (on which the cable itself sorts out the drive IDs).

    If there are no labels (often, a large plastic or fabric 'pull-tab') saying 'master' and 'slave' on connectors 2 and 3, you have a 'standard' cable - jumper as follows:
    it doesnt matter what connector goes to what; your primary (boot) HDD will need to be jumpered as 'master' (MSTR) on the drive; your secondary (or ODD) gets jumpered as 'slave' (SLV) on the drive.

    If there are labels saying 'master' and 'slave' on connectors 2 and 3, you have a 'cable select' cable - jumper as follows:
    Both drives get jumpered as CSEL on the drive; your primary (boot), or only, in the case of just one, goes on connector 3, which should be labeled 'master'; your secondary (or ODD) goes on connector 2, which should be labeled 'slave'; connector 1 goes to your MOBO IDE port...


    Floppy Drives
    Floppy drives are like case speakers - get one!! Seven dollars! Ask yourself "what do I value my time at?" You won't often need it, but when you do need it, it's either all you have to work with at that point, or makes life infinitely simpler! Some of these functions can be performed with/by a USB flash drive, but getting the flash drive to work is 'iffy'; some flash drives just won't work with GB boards (see USB Woes below); and nearly all of these functions have a 'no effort' default to floppy...

    The built-in-the-BIOS 'flasher' defaults to looking for its BIOS file on floppy. There is a procedure for (sometimes) recovering a board that has not been killed by @BIOS, but has somehow corrupted, or is unable to load the backup BIOS, called a 'blind flash'; most info is here, and, should anyone need it, I can do a walk-through. To be able to do this requires that you have a floppy, and that the floppy be first in your boot order - before you 'dump' the board - as, after you 'dump' the board, you won't be able to enter the BIOS to change this! This makes the system waste a second, and make a little bit of noise as it 'hits' the floppy, every time it boots - but, it's a small price to pay for disaster preparedness! Note that there are a few boards which can't 'blind flash', or Qflash from floppy - some i5/i7 boards (not many) have 2 meg BIOS - to check, simply download and extract a BIOS for your board; the largest file there will be the actual BIOS...

    Another useful thing for disaster recovery is the fact that your BIOS will not only store CMOS sets of parameters to the set of 'storage slots' built into the BIOS, but will save them to an external device, a floppy by default; also, flashing a new BIOS erases your stored parameter sets.

    And, last, but not least, the default location for all recent OS to 'find' the 'preload' (<F6> for Xp) drivers for AHCI/RAID is a floppy!


    Memory Installation
    For the last entire generation of boards, we have gotten 'used to' the DIMM installation rule: "use the slots closest to the processor first." With the 'new gen' boards, this has changed! Rather than try to 're-hash' the changes here, simply do this: READ [:mohsentux:6] your manual; close to the front, all GBs have a 'pictorial guide' to installing memory, showing slot IDs and nomenclature, as well as a table showing "where to put 'em", like this:


    If your board appears to be having problems with "LoadOpt" not 'recognizing' your RAM, here's a trick:
    Remove all but one DIMM, being careful to ensure it is in the slot recommended for 'single module' use; power up and do the "LoadOpt"; <F10> to save & exit; power down and add the remaining modules...

    If you have a socket 775 board, and are trying to use more than two DIMMs, you will need to 'bump' the MCH voltage by a tenth, or a tenth-and-a-half, to accommodate them. Realize that, if you are doing this, and running memory faster than 800, which is already 'overvolted' (likely, 2.1V instead of 1.8V JEDEC), you are both over-volting and over-clocking your MCH - stability may require MCH cooling:



    Power Supply - Connectors
    Almost all modern CPUs, on recent MOBOs, require an ATX-12V connector, but the whole 2x2 vs 2x4 thing is 'iffy' - it's not set in stone... My rule of thumb is: 90 Watt TDP - if it's above, 2x4; below, 2x2 will likely do. If you're buying new - may as well overkill - might want to swap in a 'monster' later. A CPU around ninety may work fine with a 2x2, but may become unstable in OC; a 2x2 to 2x4 adapter may work, or may not, and may depend on 'where the current is coming from' - the PSU rail; or, may depend upon the adapter's quality itself!

    And, do not forget the vidcard PCIe connector(s) - without 'em - no video! Also, be aware that there are a few VidCards that are 'fussy' about where their power comes from (I think, but am not positive, all nVidia) - if the six pin and eight pin don't come from the same rail - won't work...


  • Troubleshooting: :o 


    'Boot Loops'
    'Boot loops' (i.e., situations where the machine never boots successfully, but 'cycles through' trying to boot over an over again) come in several varieties - and here is one place where the case speaker (mentioned in Speaker - Beep Codes above) will be an invaluable aid to 'sorting things out'!

    First, the normal 'double boot' - after changing BIOS' parameters, getting two reboots in a row is perfectly normal behavior; it seems that, when you change certain settings (and I don't exactly know which ones - the only sure two I know are Trd and Command Rate - if you change them, I think you get the 'twin' reboot) it boots once to 'see where it's at', recalculates its remaining 'auto' settings, saves them, shuts down for a second, and then boots again. Three reboots in a row, however, usually indicates that the board was 'given indigestion' by your settings, and is going back to defaults. This sometimes goes astray, and it doesn't get back into proper operation - for example, at this point, mine will sometimes 'lock' itself into 111MHz x a six multiplier - and take a week to do a whole boot - that's time to do a CMOS reset...

    'Thermally induced' boot loops - this syndrome is seen when the board appears to boot OK, completes POST (usually) and gives the single "I'm good" beep, but before it can load the OS, and without a 'powered down' interval between, resets and POSTs again. These 'loops' may get closer (shorter) each time, until the system is never actually finishing POST. This usually indicates a problem with the CPU's ability to 'dump' heat to the HSF (heat-sink/fan assembly) - the processor is repeatedly going into 'thermal shutdown'! For this set of symptoms, you need to first examine your HSF installation; an 'unseated' locking pin, or a cracked one, will cause this, as will a non-functional fan (defective, or not plugged in correctly), or poor application of thermal paste. You will likely need to remove your board from the case to examine the locking pins carefully...

    'Corrupted CMOS' boot loops - this will usually give you a 'twin boot' symptom, much like the normal 'double boot' when changing parameters; your system will boot once, power down for a second, attempt to boot again, reset without the power down, attempt to boot yet again, power down - rinse, lather, repeat! These are most often caused by an 'errant' USB device (see USB Woes, below), but can be caused by an 'unacceptable' attempt at overclocking parameters, a buffer overrun (you do have the "No-Execute Memory Protect" BIOS item on the "Advanced BIOS Features" page enabled, don't you?), or a bad DIMM...

    'Power related' boot loops -
    If your power supply fails 'gracefully', you will get the telltale repetitive short BIOS beeps - however, often this is not the case! Typically, the symptom here will be quickly recycling attempts to boot, while never activating the display, or beeping at all. Often the 'giveaway' here will be either the fans momentarilly 'revving', then shutting down, or the 'phase LEDs' flashing once or twice between shutdowns. A bad, 'folding back', or unacceptably 'noisy' power rail will do this, as will failure of the PWR_OK signal to 'come true' - see Power Supply - Basic below...

    ...to break a 'boot-loop' (assuming it's not hardware induced - bad RAM, PSU, or 'thermal cycling'), try the following:
  • hold the front panel power switch depressed until it finally quits trying to restart;
  • switch off the power supply - or, if no switch, unplug it;
  • depress the front panel power switch for twenty seconds more;
  • jumper, or (carefully) short the RST_CMOS pins momentarily;
  • unplug all USB devices, except keyboard and rodent;
  • PSU switch back on;
  • power up - first thing, enter the BIOS and do the "Load Optimized";
  • <F10> to save and exit;
    hopefully - good to go!


    USB Woes
    GB's have a well-known 'pickiness' about some USB stuff; I've never heard of either a keyboard or mouse causing this problem, but I've seen it with: external hard-drives, external CD/DVD writers, various brands of pen drives, and, I think I recall, someone's brand of MP3 player... 'Boot loops' are the most common symptom. On the BIOS' "Integrated Peripherals" page, set "Legacy USB storage detect" (later BIOS say "USB Storage Function" ) - either way, set it to "Disabled"... If you still get looping with all USBs out, and that setting made, see 'Boot Loops' above, for the 'break a loop' procedure!

    If you've been reading here for any amount of time, you've seen me fix a couple machines just by having people unplug USB devices, and you've probably seen (likely more than once) my lecture about the "Legacy USB storage detect" (AKA "USB storage function" in later BIOS) wanting to be off for most purposes. These two items account for most of the 'dreaded GB reboot loops'... Had an interesting experience a while back that is a perfect illustration:

    I have had a number of USB devices in my X48-DS5 for a long time - HID (Logi G11) keyboard, mouse, ESA fan controller board, IR remote eyeball, webcam, and most important to this narrative, an 8G HyperX thumbdrive, used for ReadyBoost - no problems. Did have a peculiarity several months ago, when I tried to install a USB connected X-10 house control repeater - software refused to install, as it appeared never to 'find' the device - never thought much about it - put it on another system without as much crap loaded onto it, can remote it from anywhere on the house net - didn't care... Couple days ago, was browsing at HackMediaCenter7, where they mentioned better performance for your 'live TV' pause buffer by moving it to a thumbdrive - had a couple laying around (8G OZC 'Diesels' - free from NewEgg with some OCZ DDR2-1150 for testing) so thought I'd try it - wasn't happy, put it back, and rebooted with the 'Diesel' plugged in - WHOOPS! - two reboots, with a brief power-down between them - VOILA! - overclock gone! Loaded saved OC from CMOS storage, saved & rebooted - WHOOPS! - two reboots, with a brief power-down between them - VOILA! - overclock gone again! Hmmm... Pulled USB stick, loaded saved OC from CMOS storage, saved & rebooted - back to normal! Hmmm... Put stick back in - same problem repeated! Thought, just for the sake of testing, let's enable "USB storage function" - saved, rebooted - and rebooted, and rebooted, and rebooted - AH-HA! The dreaded 'GB reboot loop'! Replaced the first Diesel with the second one, just for thoroughness' sake: same performance repeats - "USB storage function" disabled, loses OC every boot; "USB storage function" enabled, reboot loop! Mind you, HyperX and Diesel both 8G, both formatted NTFS, neither containing boot files - one good, one not so good...

    So, that's exactly why I recommend unplugging USBs when trying to escape from the 'reboot blues'...

    ...my all-time favorite one of these is here, where someone had a backup drive plugged into his USB for a long time - but had never used it. Unbeknownst to him, his daughter 'flipped' the drive's 'on' switch, and, POWEE - instant 'boot-loops'[:fixitbil]!


    Speaker - Beep Codes
    There is only one thing to say about case speakers - get one!! You won't often need it, but when you do need it, it's usually all you have to work with at that point... A lot of people operate under the misapprehension that the 'diagnostic beeps' should come through the speakers attached to their sound-card/chip - not so! Your three hundred dollar Altec-Lansings won't do you any good here - you have to have a case speaker attached to the front panel header!

    Award BIOS beep code descriptions:
  • 1 short: System boots successfully
  • 2 short: CMOS setting error
  • 1 long, 1 short: Memory or motherboard error
  • 1 long, 2 short: Monitor or graphics card error
  • 1 long, 3 short: Keyboard error
  • 1 long, 9 short: BIOS ROM error
  • Continuous long beeps: Graphics card not inserted properly
  • Continuous short beeps: Power error


    i3/i5/i7 'Missing' Memory
    If you have 'X' amount of memory installed, and either the BIOS, or Windoze is only showing 'Y' amount available, the very first thing to do is: pull your CPU, get a flashlight & magnifying glass, and check your socket for bent or distorted pins:

    Nineteen times out of twenty, you will find 'em! Excellent pictorial guide to installation/examination from Intel.
    ...if the pins are not too badly mangled, they can often be 'teased' back into place, using:
    a jeweler's tweezer (most large hobby shops)
    a jeweler's screwdriver (any dollar store)
    a dental pick/tool (often found at dollar stores or hardware store counters)
    or
    a 'pin-vise' with a sewing needle mounted in it (a 'pin-vise' is a little thing the size and shape of a kid's crayon, that has a tiny chuck at one end, much like a drill chuck - they are used to drill ultra-fine holes in small work, and can be found, again, at any large hobby-shop or hardware store...)
    ...just takes a steady hand, and a bunch of patience [:bilbat]


    Audio - 5.1 Missing 'Rear' Sound
    So, you've gotten a nifty new set of 5.1 speakers, have plugged 'em in, set 'em up, and are getting no 'rear' sound. Well, to you, they're 'rear' speakers; to me, they're 'rear' speakers, too... In fact, apparently to everyone but RealTek, they're 'rear' speakers - to them, they're 'side' speakers, and if you don't set 'em up that way - no workee! Go into the 'Audio Manager', and do this:

    Clicking on the 'arrow' circled in green will produce a (strange) 'test sound' which will 'circulate' around the room. If you find you've plugged 'em into the wrong connectors, right-clicking on the picture of the jack will allow you to 're-task' it to match how you've plugged 'em in. Clicking on the 'Advanced Settings' circled in red will allow you to set whether plugging your headphones into the front panel will 'mute' your main speakers or not...


    Audio - No Sound
    If you have recently installed a new video card, and find your sound has 'gone away', right click on the speaker icon in the right-hand corner of your task-bar, and select 'Playback Devices' - your drivers have likely set your default audio to HDMI, and you, like as not, don't have an HDMI plugged in...


    Audio - Pops, 'Stuttering'
    If you are experiencing drop-outs, pops, clicks, or 'stuttering' in your audio playback, download the 'DPC Latency Checker'; if it shows grief, sometimes the latest BIOS will clear up the problem; if not, you need to check your start-up programs, and drivers for the culprit! These days, it seems like every %$#@ thing you install has to put some POC (or two, or three) in startup - lessee, we gotta have an upgrade watcher, and a quick startup loader, and, um, an extra drive indexer (as if MS's wasn't bad enough), and, oh yeah, a monitor to tell us where your browser's been, so we can find more suckers to sell this ^&%$ to...


    Power Supply - Basic
    You can at least do a cursory check of your power supply, if you have, or can get access to, a VOM (otherwise known as a 'multimeter')

    to see if it appears to be working...

    The "Paper-clip trick" to turn it on...
    Connector pin-out...

    Note pin 8 (PWR_OK - gray) - if it does not come 'high' (over 3.3V - should really be right around 5V), your system can never 'start-up', even if all the 'rail' supplies are at nominal voltage... Also, be aware - while you can check 'no-load' voltage, if there's a short somewhere, or the rail itself is bad, it may be 'folding back' under load (kind of like blowing a fuse or a circuit-breaker), and not actually delivering the power you 'see' at 'no load'...
  • More about : gigabyte guide

    !