MSI K8N Neo4 boards not posting

[GearCat]

*** UPDATE: Please see my subsequent post on 8/6/05. The problem I descibed in this first post was the result of not having connected the 4-pin 12V CPU power connector to the motherboard. The MSI K8N Neo4 boards have been performing as they should and there is no reason I can see not to purchase them! ***

In the last month I've bought 3 different MSI K8N Neo4 boards (and returned 2). These are the K8N Neo4-K, K8N Neo4 Platinum, and K8N Neo4 SLI. All had 4 red LEDs on the D-Bracket stay lit on power up and nothing comes up on my monitor. You would think the processore was bad or installed wrong according to the manual, but this processor worked fine when tested in my working system (see bottom of this post). I diagnosed with MSI tech support every way imaginable with bare minimum set up, which is as follows:

- AMD Athlon 64 3200+ Winchester
- OCZ Gold Series 1GB (2 x 512MB) Unbuffered DDR 400 Dual Channel Kit (OCZ4001024ELDCGE-K)
- Kalman ZM400B-APS ATX ver. 2.03 power supply (12V, 20 pin)

Not only have I tested the processor in my working system, I've tested all the components listed above on my working system and they all worked fine. I even tried all the components listed below from my working system on one of these other bords and again I get no post (all red LEDs lit).

I've read about the capacitor plague, but I can't find any evidence of faulty capacitors by physical appearance or use of capacitors from suspect manufacturers.

So anyone who can shead some light on this problem, please reply ASAP!!!!!!!!

The only thing I can do if no one knows the problem is to either send the board to MSI for diagnostics and repair or return to ZipZoomFly as defective and repeat this process until I get a working board. I just hope it's nothing I'm doing wrong.

BTW, MSI has a free uprade to their K8N Neo4 SLI boards for compatibility with the Athlon 64 X2 and the GeForce 7 PCIe video cards.

Thanks,

Scott

 

[FastEddie]

Could very well be a power problem as your psu is only providing 18a MAX on the 12v rail. The odds of three boards in a row being bad? It's your psu.

 

[GearCat]

I considered that, but since I have essentially the same Kalman PS (same specs) on my working system and I'm now trying essentially the same motherboard except for SI's SATA RAID), I discounted that possibility.

But, considering your point, I have a slightly higher power (20 amp) Antec True430 PS I'll try when I get a chance Monday. How did you figure 18 amps on the Kalman PS. I thought is was 15 amps at +12V?

 

[FastEddie]

The newest (most current) version of the Zalman ZM400B-APS lists 18a on the 12v rail. You are pressing it with 20amps in the Antec. The one common demoninator with all NF4 boards is power---they all hammer the 12v rail, especially the Neo4 Plat (and Sli).
I would be looking for a psu with 26a on the 12v (minimum).

 

[GearCat]

I guess this is a case where the PSU manufacturers haven't kept up with new motherboards. I looked at many PSUs over 500W and only found two from Anandtech with 26A at +12V (True550) and 24A at +12V (TrueControl 550). The new yet to be released CoolMaster Real Power 550 for example is only 18A.

You have any PSUs with high current output at 12V you'd recommend?

Alternatively, you have any Athlon 64 motherboards (e.g., with VIA chip) you'd recommend with lower current requirements (don't need SLI, but it's OK if it's there)?

 

[asearchforreason]

I'm not a PSU guru, but it's not just about wattage. The amperage on the +12V rail is more important these days. The wattage is just the max it can put out on all the rails, but if all of that power is going to the +5V, it's not going to help your situation much. There are a bunch of PSUs with more amperage than you listed on the +12V, unfortunately I don't know what they are. I only know mine which is the Antec NeoPower 480 W, it has dual 12V rails with 18A on one and 15A on the other for a total of 33 A.

 

[FastEddie]

OCZ Modstream 450W (26a on the 12v rail)
OCZ Modstream 520W (28a on the 12v rail)
OCZ PowerStream 420W (30a on the 12v rail)
OCZ PowerStream 520W (33a on the 12v rail)
A+GPB Athena Power500W AP-P4ATX50F12 (28a on the 12v rail)
A+GPB Athena Power520W AP-P4ATX52F12 (28a on the 12v rail)
A+GPB Athena Power 500W AP-P4ATX50FE (28a on the 12v rail)
Tagan TG480-U01 (28a on the 12v rail)
PC Power & Cooling 510 Sli (34a on the 12v rail)
PC Power & Cooling 510 Express (34a on the 12v rail)

These are all single rail psu's, Which I, myself, prefer. These will also work with any of the other MSI nForce 4 boards, (and any other manufacturers nForce 4 offerings as well).

 

[GearCat]

Thanks for the PSU info. I should have checked OCZ as I knew they made nice PSUs. I'll pick up the 520 Watt PowerStream at CompUSA near my house and give it a try Monday. I'll post the results.

I actually bought this OCZ PSU for my working system belwo with the MSI K8N Neo4 Platinum/SLI board, but it would not fit in the case I'm using with this board, so I returned it. The case is a desktop type I use in a rack and unfortunately the PSU mounts right behind the 5 1/2" bays where I must have a rack for a removable hard drive (a really tight space). That's why I went with the Kalman and the fact that MSI said it would work.

While the system below in the desktop case is working, I may be posting a question on high amp PSUs with smaller dimensions should I choose to replace the Kalman.

 

[GearCat]

BTW, is there any special reason why high amps on a single rail is better than lower amps on two rails? I would think motherboards with the 24-pin power connector have been optimized to use the amps on the two 12V rails when a 24-pin BTX/EPS 12V PSUs is used just as well as the motherboard would use the amps on a single 12V rail when a 20-pin ATX 12V PSU is used.

 

[mudboy]

Try replacing the CMOS battery. If that was it, please PM me with the BIOS version you are running. I have a working theory on current Neo4 Platinum boards, and it has to do with them eating CMOS batteries at the rate of 1 or so per month

 

[FastEddie]

Originally posted by: GearCat
BTW, is there any special reason why high amps on a single rail is better than lower amps on two rails? I would think motherboards with the 24-pin power connector have been optimized to use the amps on the two 12V rails when a 24-pin BTX/EPS 12V PSUs is used just as well as the motherboard would use the amps on a single 12V rail when a 20-pin ATX 12V PSU is used.

The newest ATX standard calls for dual 12v rails, one suplying the processor, the other supplying all else requiring 12v. NF4 was designed and released just as the atx standard was changing from a single rail to multi-rail. NF4 could benifit from dual rail, but not as it is implemented with most of the power supply manufacturers today. Most, if not all, of the dual rail psu's are of the 15a/18a varuety. It's good that the processor has it's own power source to draw from, but 15a is quite alot for the processor, while 18a is no where near enough for everything-else, especially Sli. A single rail psu, with sufficient amperage on that single 12v rail will have no problem powering even the most demanding perifs.

 

[GearCat]

Well guys, I just have to sit back a laugh. I must have been out for a cup of coffee when they where giving out the last brain upgrade (I need these badly as I'm getting older). Let me explain. I picked up a OCZ PowerStream 420 (520 was not in stock) and hooked it up (I thought correctly) to the MSI K8N Neo4 SLI board and on power up, the same problem, no post. After a breif moment it dawned on me. I never hooked up the 4-pin 12V power connector for the CPU. I mean how brain dead is that! Well it posts now! In fact I hooked my Kalman PSU back up, this time with the CPU power connected, and it also worked, just like my other working system (at least I got that one right). Of course I've got the new board loaded at bare minimum (one non-SLI GeForce 6600 video card and memory). So you're right, if I load the board up down the road (including two SLI GeForce 7 video cards), I'll need the power capacity of the OCZ PowerStream.

Do you think going to the Power Stream 520 is worth the extra 3 amps on the +12V and the +10amps on the +5V?

On the issue of the CMOS battery. I tried replacing the battery before I discovered my mistake with the CPU power. This had an interesting effect. Instead of getting all red LEDs on the D-Bracket, I got LEDs 1 and 4 green and LEDs 2 and 3 red, but w/o the sequencing through any initialization or testing steps and no posting. So not sure why that occurred, but clearly the battery was not my problem and I have put the original battery back in.

THANKS FOR ALL THE HELP ! ! !

 

[FastEddie]

:beer::thumbsup:

 

[ethiopianbuffman]

I am having the same deal, I thought it was the CPU but not, I have a Thermaltake Purepower 460w with 15A on the 12v rail. It was running just fine and then all of a sudden it died and that was it. It would be interesting to see if it was the CMOS battery, b/c the guy said if he left it on long enough it told him to replace it.

I have amd64 3500 venice, ocz gold ram, same as above.

 

[BeigeOtaku]

I'm having problems with my K8N Neo4 Platinum eating CMOS batteries ever since I upgraded my PSU.

old PSU = RAIDMAX 420W rated for a whopping 13A on the +12V rail
new PSU = XClio 450BL rated for 15A/17A

System ran flawlessly on the old 420W 13A single rail PSU - I only upgraded to try to solve a video glitch that turned out to be the nForce drivers.

After about a week with the new PSU, it didn't come back from a reboot. Fans came on the instant the PSU was switched on, but no POST and system did not respond to power button.
Naturally, the first thing I tried was swapping the old PSU back in. I fiddled around for a while and finally got it working again with either PSU. I noticed that the system clock had cleared back to default date/time which was preventing it from saving any BIOS configuration settings. This was my first indication of a CMOS battery problem.

When it died again the next week, I immediately swapped out the CMOS battery for a spare from an old mobo and after resetting the system clock it seemed to do the trick. Now I'm on my third battery and it's starting to really annoy me.

Symptoms include:
- fans come on when PSU is switched on, but no boot
- BIOS system clock cleared, won't save settings, won't reboot on BIOS exit
- clock running fast in WinXP
- games running too fast (specifically FFXI)

To do:
- try completely swapping old PSU back in and see if it continues to eat CMOS batteries

So is this a known issue on the MSI nForce4 boards?

 

[Mykl]

I just installed an MSI K8N to replace my buggy DFI NF4 Ultra-D. I also upgraded my PSU to a 480w Antec NeoPower.

It was quite possibly the easiest, smoothest install ever. The only problem I had was that the board didn't like my USB mouse/keyboard, even when plugged into PS/2 with adaptors. The problems cleared up when I struggled through the XP install and got the system drivers loaded so it would recognize them over USB.

So far it's been stable.

 

[GearCat]

Well I can now say that my MSI K8N rigs (I 've buit two now, one of which runs all the time) have not had any problems with eating CMOS batteries. Not yet anyway. It's only been a month on the second "always on" rig using the Athlon 64 3200+, K8N Neo4-K mobo, and a Kalman 400W power supply. The first system (see my 1st rig setup at the bottom of my posts) has been running since March with the K8N Neo4 SLI Platinum. I'm about to build a third system on with a K8N Neo4 SLI mobo using X2 and GeForce 7800 GT video card and OCZ Powerstrean 520 PSU, so we'll see how that goes. I sent this mobo back to MSi for the free upgrade for compatibility with X2 and Geforece 7. I will post any problems here or in a new thread.

 

[CraigRT]

Man you should really upgrade your original post

This is going to scare many builders away from the K8N Neo4 boards!.. hehe

I am myself considering getting one!

 

[GearCat]

Thanks for that observation Craig. I have updated original post.

I'll be configuring a K8N Neo4 SLI in a few weeks as I noted above, so I'll post on the results of that hoping it will encourage (I hope) others.