Motherboard - Lights, fans...no action

[PCMan]

All right,
I just replaced my old fried motherboard with a new Asus P4P800-VM. I have spent the last few hours trying to get the blasted board to post a screen :rolleyes:

There must be something that I am missing. Sylvander, I love your flowcharts and use them frequently; however I have been "through the flow" with them and came up at a dead end.

Let me start off with what I have:
Asus P4P800-VM Motherboard *New* (onboard video)
1GB PC3200 Corsair ValueSelect (2x 512MB sticks)
1.5GHz Pentium 4 478-pin 400Mhz Bus Speed (was on old motherboard...could have fried this???)
Coolmax CM-300 Micro-ATX Power Supply

Ok, I replaced the old motherboard with the new Asus board, connected both of the required power cables (ATX Main & P4 12V), stuck a stick of RAM into the first slot, plugged in the power-on and HDD LED cable (did not plug the front LED cable because the MB requires a 3 pin connector...only a 2pin on the case). The motherboard has onboard video (however I have two AGP graphics card that I have tried in it).

I plugged in the PSU (light on MB came on). Plugged in the monitor and keyboard. I hit the power switch, and the fans started up, but no video on screen.

I have tried reseating the RAM (tried the other stick as well), putting in an AGP video card, resetting the CMOS, removing the battery, attaching a floppy and a HDD.

*Sigh* anyone have any ideas as to why it is not posting a screen?

P.S. This motherboard does not have a system speaker on it, so I can't tell if it has any beep codes. I looked on every major electronics store website I could think of (even newegg!), and I could not find a single motherboard speaker (even the wonderful radio shack does not show one on thier web site)...where does one find a motherboard speaker these days? :confused:

[Sylvander]

"anyone have any ideas as to why it is not posting a screen?"
Perhaps the POST is running [see any signs like blinking LED's?], but something is failing a test [video?] and the POST halts.
Do you think the POST is running?
If you could get an internal speaker connected you could run it barebones without any RAM, and if the POST is running you should hear the POST warn that the RAM test has failed.
That at least tells you the POST IS running and that it's getting that far.
Seen a few lately where that is the case and the POST is halting after the RAM test, but before the display is enabled.

Here is that section of a generic Phoenix POST between the RAM test and the 1st display...

[b]Code_Beeps_POST Routine Description
20h 1-3-1-1 Test DRAM refresh
22h 1-3-1-3 Test 8742 Keyboard Controller
24h Set ES segment register to 4 GB
28h Auto size DRAM
29h Initialize POST Memory Manager
2Ah Clear 512 kB base RAM
2Ch 1-3-4-1 RAM failure on address line xxxx*
2Eh 1-3-4-3 RAM failure on data bits xxxx* of low byte of
memory bus
-------------------------------------------------------yours could be failing after here
2Fh Enable cache before system BIOS shadow
32h Test CPU bus-clock frequency
33h Initialize Phoenix Dispatch Manager
36h Warm start shut down
38h Shadow system BIOS ROM
3Ah Auto size cache
3Ch Advanced configuration of chipset registers
3Dh Load alternate registers with CMOS values
41h Initialize extended memory for RomPilot
42h Initialize interrupt vectors
45h POST device initialization
46h 2-1-2-3 Check ROM copyright notice
47h Initialize I20 support
48h Check video configuration against CMOS
49h Initialize PCI bus and devices
4Ah Initialize all video adapters in system
4Bh QuietBoot start (optional)
4Ch Shadow video BIOS ROM
4Eh Display BIOS copyright notice
----------------------------------------------------------not getting this far?

The usual problem is that most failures don't result in beeps [see the list]; and [even worse] in your case you have no speaker!
A shop could probaly test to see which item of hardware is causing the problem during POST. There are few items left connected, so only a limited list of possible culprits.

[PCMan]

Thank you for your reply Sylvander,

I guess I will have to get a system speaker; however, where does one find such a rare item? :p Like I said before, I have tried all of the stores I can think of...do I have to go to a computer specialties store to get one?

Will a motherboard post a screen if there is no CPU installed? I tried doing this before I installed the CPU to see if the motherboard would run, and it did not post a screen.

[ski]

A fried MB can take other components with it.

If you're doing a barebones setup with just the PS, MB, CPU, RAM, monitor, and KB installed, then place the MB on a piece of cardboard.
If that works, then the MB is shorting out in the case. Remove all metal standoffs that do not line up with a hole in the MB, and check for any loose metal objects under the MB.
If no luck, then place paper washers made from a manila folder between the standoffs/screws and the MB.

If placing the MB on cardboard does not work, then if possible either swap out all of the barebones components that came over from the old MB, test them on another system, or have them shop tested.

[Sylvander]

"Will a motherboard post a screen if there is no CPU installed?"
I believe it won't.
First the PSU tells the PC that it is capable of supplying power [gives the "Power Good" signal]. Then the PC [with the processor in control] has to tell the PSU to switch on/supply full power. [That's the picture I have in my memory.]

"where does one find such a rare item?"
Phoned my local Maplin store in Edinburgh and he said that modern boards normally have a tiny speaker the size of your thumbnail on them. That case speakers are not normally supplied separately, but Maplin do sell small speakers that will do the job.
See http://www.maplin.co.uk/Search.aspx?criteria=VC84&doy=3m3&source=15
Code : VC84F = 49mm, £1.99

http://www.maplin.co.uk/Search.aspx?criteria=L68AA&doy=3m3&source=15
Code : L68AA = 45mm, £1.99

[PCMan]

I got my hands on a speaker for my motherboard (thank goodness for old computers lying around :D ). I plugged it into the speaker pins and turned on the motherboard...no beep. I then took out the memory...no beeps still. :(

Ski - Yes, I have it on the bench right now (on a piece of cardboard)...still no dice.

I am stuck at this point...could have a DOA motherboard or a dead CPU or a power supply that is not putting out enough voltage. I tested the voltages of the power supply, including the main socket and P4 connector, with my voltmeter (all within range)...could this still be faulty (I tested the power supply while the motherboard was on...some load)?

Sylvander,
Do you have a POST Routine list for an AMI BIOS? It looks like the list you gave me was for a Pheonix BIOS.

[PCMan]

Ok here's something questionable. ;) The case speaker that I have is only a two pin (encased in a four-pin molex connector). One wire is red and the other is black. On the motherboard there are four pins for a case speaker, and they are arranged as follows: pin1-speaker, pin2-ground, pin3-ground, pin4-+5V.

Which two does the speaker get plugged into? I first tried connecting it to the speaker (red) and ground (black) - no beeps. I then tried +5V (red) and ground (black) - no beeps. I then rearranged the molex connector and tried +5V (red) speaker (black) - no beeps, but slight sound.

[Sylvander]

"Do you have a POST Routine list for an AMI BIOS? It looks like the list you gave me was for a Pheonix BIOS."
That POST routine for the generic Poenix BIOS was something I stumbled upon whilst searching at the Hewlett Packard website for information about the PC I'm using here. Unfortunately I don't have the same for an AMI BIOS.

"Which two does the speaker get plugged into"
Sorry, don't know.
I'd gues it should be between "positive" & "negative" terminals. Perhaps between "speaker" [I guess this is positive], but which of the others is "negative". It shouldn't be between positive and ground! Ground is for earthing the speaker body I expect. And the +5v doesn't seem appropriate.
Try it anyway [pin 1 (speaker) and pin 4 (5v)].
-------------------------------------------------------------------------
I always see the supply to a speaker as a push/pull effect. Like a pulley with a rope around it. first you pull on the left, then you pull on the right, and the pulley rotates firs this way then that.
In reality the voltage is positive [current goin in] on the positive wire and the diaphram goes forward. Then the voltage goes negative on the positive wire and positve on the negative wire and the diaphram moves rearward.
As the diaphram goes forward and rearward it sends out compressions and rarefactions. This all happens at high frequency to produce what we hear as sound waves in the air.
--------------------------------------------------------------------------

[PCMan]

Sylvander,

Thank you, I was able to pull the POST routine off of AMI's website (in a PDF).

I think I might have found the problem...and if so, Asus goofed. The BIOS chip on the motherboard does not have a sticker on it to tell what BIOS version is loaded into the chip (the picture on Newegg's website has a sticker on the BIOS).

This leads me to think that Asus never loaded the BIOS into the ROM chip...that would explain perfectly the no beeps & "dead" running state of my motherboard.

I sent a form e-mail to Asus yesterday and asked them about this "no sticker" issue. We will see what they say (if they respond at all :rolleyes: ).

I certainly hope this is the case. If so, all I have to do is get an RMA from Newegg, and get myself another board.

Let me know what you think about this "no sticker on ROM" thought. Most every motherboard I have has a sticker on the ROM chip.

[Sylvander]

See www.pcguide.com/vb/showthread.php?t=44070&page=1&pp=25
Budfred cannot get his new Asus board to complete the POST and found that it was because they fitted the wrong BIOS ROM chip and the BIOS cannot cope with his dual processor even though they say on the box that it will.
He is most unhappy with their response to his calls to them.
He was trying to get them to send out a new chip, but that was full of delivery problems, so he has given up and intends to buy a new board.

In the UK, if something is "not fit for the purpose for which it is intended", a full refund or replacement must be supplied by the seller if demanded by the buyer.
A new law says that if a fault is found within 6 months of purchase, the seller must either prove that the item is not faulty [the buyer doesn't have to prove anything], or else refund or replace.

Don't know anything about stickers on ROM chips I'm afraid. :(

[PCMan]

Sylvander,
Thank you for the link. I have been following Budfred's thread since I first had this issue, and I have some of the similar issues with Asus's support.

I was able to get in-touch with a tech support guy from Asus this morning (after they did not return my call from yesterday :rolleyes: ).

The tech guy had me go through the routine of removing the CPU and memory in-order to force an error code or beep (LOL...this board does not have an onboard speaker!). He said to connect my main speakers (I guess some of their boards have a BIOS that is able to diagnose errors through the main speakers). After all of this (and at the expense of my phone call...no toll-free number! :eek: ), he said to get in-touch with the RMA department.

*Sigh* So I will call them tomorrow morning, and get a RMA for this board.

I have always heard that Asus makes a great board, and I am willing to make allowance for a few DOA boards. However, to not back up their product with competent support (namely not returning my call when they said they would) has really dropped my confidence in them...not good for a first time buyer from them.

Hopefully they will make it up with my RMA...I wonder if they would throw in an extra CPU with the board :D (a guy can dream can't he?).

I asked the tech guy about not having a sticker on the ROM chip, and he said that this means that it has the original BIOS release in it (that means this board has been on the shelf since 2003!). *Sigh* I still stand that they never loaded the ROM with a BIOS!

Anyway, I will keep you posted with the outcome. Hope to get my computer back up and running soon!

[Sylvander]

If this were my wife doing this she'd buy 3 boards, find the one she liked best [the one that worked], and return the other two! :)

Of course for that to work, the shop must accept returns with no fuss.

[PCMan]

That's just the problem. I bought this board from Newegg, and Newegg leaves the RMA of this board up to Asus.

So, I am left with Asus's RMA.

[Sylvander]

Here in the UK, the law says that there is a contract between the SELLER and the BUYER.
That if the goods are faulty it is the responsibility of the seller to honour the contract by supplying goods that are "fit for the purpose for which they are intended".
THE BUYER HAS NO CONTRACT WITH THE MANUFACTURER, although the seller does, and the manufacturer may have responsibilities to the seller.
Where are you and what are the local laws governing your dealings [as buyer] with the seller?
You may, if you feel so inclined, co-operate in this arrangement of by-passing the seller, but I'd guess you are not required by law to do so.

[PCMan]

I am in the USA, and I am embarrassed to say that I do not know exactly what the rules are that govern the responsibilities between the buyer, seller and manufacturer.

But as far as Newegg is concerned, they will not RMA this motherboard (their online RMA does not allow me to RMA the board). However, if I were in UK and knew these rules of sales, I would defiantly challenge Newegg and hold them to the laws governing sales.

But as long as Asus holds up to their end of the deal, and it gets my computer running, I will be happy.

[PCMan]

I am starting back up on this thread because my computer is still in a "dead" state.

I was able to get the new motherboard RMA'd through Newegg (wonderful support...unlike Asus :mad: ), and it arrived today.

I put everything back in place: CPU, 1 stick of RAM, fan & heatsink, ATX power connector & 12V connector, front panel cables, and hard drive w/ power cable.

The motherboard has onboard video so I hooked up my monitor to it. I plugged in the power supply and pressed the power button: fans and front power light came on, no activity on the HDD LED, and again no post screen.

This is a true stumper, and I appreciate anyone's input at this point.

Assuming that the motherboard is good, this leaves the CPU, RAM, and power supply to troubleshoot.

1) The CPU could have been burnt when my old motherboard died (I first thought that the motherboard fried because when I replaced the power supply [thinking the original died] the old motherboard turned on for 20 seconds and then turned off...I could feel some extreme heat coming from the capacitors and regulators next to the CPU...could have been CPU?)

2) The RAM could be incompatible with the motherboard or CPU (see my thread "Matching CPU and Memory Speed")

3) The power supply may have been burnt when I hooked it up to the old, possibly faulty, motherboard. (I have checked the voltages for this power supply and they are all within range)

Would getting a POST diagnostic Card help? The motherboard does not have an onboard speaker...so no POST beeps.

Could some one recommend a good POST card at a reasonable price, and where to get one?

[SimonBarker]

directron.com sells a front panel speaker for 4.99.

I have a similar problem that I haven't resolved yet. If I take out the ram, I get the expected 3 clear beeps repeated every couple of seconds. I put the ram back in and the thing runs for a while, emits a burst of 2-3 very closely spaced beeps just once, they're almost merged into one, then turns over the harddrive for a while (for a couple of minutes) .

However nothing appears on the screen, I'm using the onboard video. The AMI beep codes are not really helpful in my opinion, I would get a board with a pheonix bios in future.

FWIW: p2.8 800 fsb 512k + pc3200 ram samsung

Good luck

[PCMan]

Thank you, SimonBarker.

I agree with you...the beep codes do not do much for me...especialy since I don't get any :rolleyes:

Rather than purchasing a speaker for this problem; I have decided to take the whole computer into a shop and have them troubleshoot it.

Does anyone have any suggestions/experience with taking it into a shop? They say they test everything, and are able to tell exactly what is the problem :eek: (makes me wonder what they use ;) ) Does anyone have any good suggested questions I should ask at the repair shop?

I plan to take it in this weekend, but would appriciate any input! I have my fingers crossed that this will work.

[Sylvander]

"I have decided to take the whole computer into a shop and have them troubleshoot it."
Good move. :)

"Does anyone have any good suggested questions I should ask at the repair shop?"
Ask them to give you a [written] "Quotation" for the cost [what they WILL charge and what they will do] rather than an "Estimate" of the cost.
Then they MUST do what they agreed at the cost agreed.
If they refuse to give a quotation and insist on only giving an estimate, then BEWARE; the cost could escalate.

Ask them to give a written report on what they find.

[PCMan]

Thank you, Sylvander, for the tips; I made sure they gave me a written price!

I took in my computer to the shop this afternoon. We will see what will happen. They said they should have it diagnosed in three to four days (they have quite a few computers they are working on).

I will keep you all posted on the outcome.

At this point, I would like to thank everyone who has helped with this issue!

As far as I am concerned, this is THE best PC forum on the internet, from the amount of activity, to the knowledge, to the wonderful atmosphere...Thank you!

Much thanks to Charles (a.k.a, ixl) for creating and maintaining this forum! Excellent job!

I will post back as soon a I get the word on the patient... :D

[saphalline]

The BIOS support isn't a problem since that particular P4 has been fully supported by all BIOS versions of that mobo. I was quite surprised, actually, to see Willamette support on an i865G-based mobo.

As for the problem, I think it may be the CPU. I don't have the details as to what happened before, but from what you've said in this thread, I tend to think the CPU is dead. Not a bad thing, really, since a new CPU for that mobo would be dirt-cheap. A lowly Celeron D would easily out-pace that old P4 1.5 you have now, although the BIOS version is still of concern at this point. :rolleyes:

Hmmm... Asus has been having problems lately... They're on my probation list now.

[PCMan]

saphalline,

Thank you for your input! I have a feeling that it is the processor, too. But I did not want to shell out more money without knowing for sure. :)

Along the same lines, if it turns out to be the CPU, I would like to know what CPU are supported by this new motherboard.

I have a printout on all the CPUs that Asus says will work with this motherboard: Here's the link (http://support.asus.com/cpusupport/cpusupport.aspx?SLanguage=en-us&model=P4P800-VM)

However, I do have one main question about Pentium 4 CPUs. What does the "CPU Core Stepping" mean?

I am looking at a possible CPU. However, it has a stepping of E0, and the listing from Asus does not come right out and say which CPU Core Stepping it supports at this CPU speed.

I am looking at a P4 3E GHz Spec # SL7PM. Here is the Specs (http://processorfinder.intel.com/scripts/details.asp?sSpec=SL7PM&ProcFam=483&PkgType=7022&SysBusSpd=ALL&CorSpd=ALL). However, the Asus listings show most of the supported CPUs with C0/C1/D1 stepping, only a supported Celeron D has a E0 stepping.

So what exactly does the CPU Core Stepping mean and does this have a bearing on what my motherboard will support?

[saphalline]

The CPU you linked to is the P4 530, at 3GHz. Often called the P4 "E" versions. Likewise, the Celeron D's are "E" versions, too, because they also have a model number. Normally speaking, the "E" versions denote the beginning of the Prescott core revision, but there are a few Intel CPU's that don't follow this loose rule. If you want to use that CPU you linked to in your new mobo, you'll need the latest BIOS version. Just FYI.

Ok, so what is a "CPU Core Stepping" then? Put simply, it's an incremental enhancement. The design of any particular CPU isn't static, the engineers are always working on it, always tweaking little parts of it. A core stepping represents the smallest change to a CPU that can be made. The hierarchy for CPU changes is this:

Microarchitecture - An entirely new CPU design that involves a family of processors. The older Intel P6 microarchitecture for instance was the basis for the 1995 Pentium Pro, the 1998 Pentium II, and the 1999 Pentium III. In 2001, Intel came out with the NetBurst microarchitecture that was the basis for all Pentium 4's and the Pentium D. And of course Celeron variations on all of these were made. Recently, we saw the release of Intel's latest Core microarchitecture that is the basis for the new Core Solo and Core Duo mobile processors, as well as the desktop variation called Conroe being released later this year.

Core Revision - A major set of changes within a microarchitecture that indicates a new version of a CPU. A core revision may usher in a new processor name, such as PII vs PIII, or it may be a major change for a processor of the same name, such as Northwood P4 vs Prescott P4. Process size reduction, changes in the amounts of cache, FSB speed bumps, extra instruction sets, and other significant modifications are usually the trademark of a core revision. These represent a change big enough to be a buying decision.

Core Stepping - A minor tweak within a Core Revision that makes small changes to transistor layout and circuit-level efficiency modifications. Usually these only concern heat dissipation, energy usage, and resiliency within the CPU core. Engineers are constantly working on existing processors to see if more headroom can't be made, and occasionally these experiments result in a higher speed version of a CPU. Once a successful core stepping enhancement has been verified in the testing labs, all subsequent fabbed CPU wafers are made with the new core stepping. These are of great concern to hard-core OC'ers, but everyone else can gloss over them. Core steppings are not of a great concern to anyone running at stock speeds or with moderate OC'ing in mind. Core stepping can be identified by the CPUID string (available for viewing in utilities such as WCPUID) or by the S-spec number on an Intel CPU, or the OPN on an AMD CPU.

[PCMan]

Thank you saphalline! I will keep all this in mind when (and if) I need to get a new CPU.

[PCMan]

I received the diagnosis from the shop and picked up my computer last week.

It turns out that it was the CPU that was dead (it died a hot death...really felt the heat when it died :rolleyes: )

So I looked at my options and got a slightly used Pentium 4 3.0E GHz (800mhz FSB & 1MB L2 cache) processor on ebay. Works wonderful! Plugged the CPU into my new motherboard and applied thermal paste and attached the included Intel Fan.

One push of the power button and voila...a POST screen (such a beautiful sight after weeks of pulling, buying, and sweating over parts :D )!

The processor runs at a toasty 44 degrees Celcius (112 F) idling in the BIOS (I will need to keep an eye on the temperature). I ran Memtest 86 and the CPU/Mobo/Mem combo really showed their stuff! I can't wait to see how it will perform in the Windows environment with a few games ;) !

Unfortunately, the OEM copy of Windows XP from the manufacturer did not work (was only a Ghost image...I didn't expect it to work).

So now all I am waiting for is my copy of Windows XP that I ordered last night from Newegg. Should be up and running sometime next week.

Thank you all for your time spent helping me with this issue: saphalline & Sylvander, and the others that helped in my other post!

This forum is a wonderful resource, and you guys are what make it so!

[PCMan]

I am connecting everything right now to the motherboard but...

Here is a cable connection question. My original motherboard had two USB 1.0 front panel connectors which plugged into the pins on the motherboard via the IDC connector show below.

This new motherboard has USB 2.0 supported pin slots for front USB connectors.

However, the pins on the motherboard do not match the original front USB panel IDC connector.

First off, are all USB connectors the same? What I mean is, are these front panel USB connectors compatible with both USB 1.0 & 2.0?

My reason for asking is because all the cables are on these connectors (I assume in the correct order)...except for the no pin cap (white cap...pin #9) and the No Connect wire (black cable...pin #10)...these are swapped for some reason (the cap is on pin #10 and the cable goes into pin #9).

Should I chance it by swapping the cap and the wire to conform to the new motherboard, or is it possible that this front panel connector is not compatible with my motherboard?

http://www.intel.com/design/motherbd/av/pix/av_fpan_usb.gif
http://pinouts.ru/connectors/idc10m.gif