The *Official* MSI K8NGM2-FID GeForce 6150 Motherboard Thread

[avi85]

Originally posted by: rwm
Posting my disappointing overclocking results.

3500+ Venice
Corsair DDR400 2x512 value ram
Mem Limit 166

The max it would go and still pass mprime and memtest was 2.333GHZ (212MHZ). It would boot up to about 2.45GHZ but fail mprime and memtest #5 almost immediately. At 2.5 it wouldn't boot at all. Based on the excellent first entry in this thread, I was expecting better. Still, a nice little computer

rwm

I had almost identical results with a 3200+ I got it prime stable at 2320MHz although now I have it stable at 2420MHz thanks to an added 0.05v from RMclock (and chinna)

 

[chinna]

Originally posted by: avi85

Originally posted by: rwm
Posting my disappointing overclocking results.

3500+ Venice
Corsair DDR400 2x512 value ram
Mem Limit 166

The max it would go and still pass mprime and memtest was 2.333GHZ (212MHZ). It would boot up to about 2.45GHZ but fail mprime and memtest #5 almost immediately. At 2.5 it wouldn't boot at all. Based on the excellent first entry in this thread, I was expecting better. Still, a nice little computer

rwm

I had almost identical results with a 3200+ I got it prime stable at 2320MHz although now I have it stable at 2420MHz thanks to an added 0.05v from RMclock (and chinna)

What is surprising to me is, my X2-3800 is able to do 2480Mhz ( from 2000) with default core 1.35v. Actually it would do even better if the this MB gives stable voltage.

This MB jumps all over for core votage. I was undervolting to see how my X2-3800 will do at 10x (2000Mhz) and it was doing good at 1.1v, but suddenly one instance of Prime95 gave errors. When I ran tests while observing core voltage, voltage is fluctuating somuch, that at 1.1v setting it is giving 1.11v to 1.065v. It is not just MB is undervolting, but jumping all over the place. So, the prime95 errored out when voltage sagged to 1.065v. It is stable as long as it is around 1.1 or even 1.08v.

It is same case at core 1.35v also, it is jumping anywhere from 1.304 to 1.353v.

 

[avi85]

Originally posted by: chinna

Originally posted by: avi85

Originally posted by: rwm
Posting my disappointing overclocking results.

3500+ Venice
Corsair DDR400 2x512 value ram
Mem Limit 166

The max it would go and still pass mprime and memtest was 2.333GHZ (212MHZ). It would boot up to about 2.45GHZ but fail mprime and memtest #5 almost immediately. At 2.5 it wouldn't boot at all. Based on the excellent first entry in this thread, I was expecting better. Still, a nice little computer

rwm

I had almost identical results with a 3200+ I got it prime stable at 2320MHz although now I have it stable at 2420MHz thanks to an added 0.05v from RMclock (and chinna)

What is surprising to me is, my X2-3800 is able to do 2480Mhz ( from 2000) with default core 1.35v. Actually it would do even better if the this MB gives stable voltage.

This MB jumps all over for core votage. I was undervolting to see how my X2-3800 will do at 10x (2000Mhz) and it was doing good at 1.1v, but suddenly one instance of Prime95 gave errors. When I ran tests while observing core voltage, voltage is fluctuating somuch, that at 1.1v setting it is giving 1.11v to 1.065v. It is not just MB is undervolting, but jumping all over the place. So, the prime95 errored out when voltage sagged to 1.065v. It is stable as long as it is around 1.1 or even 1.08v.

It is same case at core 1.35v also, it is jumping anywhere from 1.304 to 1.353v.

could this have to do with your power supply?

cause when mine is supposed to be at 1.45 it does jump around but it jumps more up than down (1.448-1.488 during prime)

 

[RjG]

The CPU voltage is derived from the 5v rail of the power supply, with regulation on the board to step it down to 1.35v. My numbers jump around an insane amount too, with a couple really decent Antec power supplys... so (personally) I don't think it's a PS issue.

It could just be that how they monitor or report the voltage is just crappy, rather than actual regulation problems. It varies too much to be "real" I think.

 

[RjG]

It SURE WOULD BE NICE to have manual voltage control of the cpu though, wouldn't it?

Too bad no "bios hacker" types picked up this board...

 

[grits]

Has anybody been able to use a voice app (yahoo, messenger, etc) with the onboard sound?

I noticed a newegg reviewer said it didn't work, and I also can't get it working. There is feedback even when using a headset, so it seems the mixer isn't working right.

Or maybe I don't have it configured correctly. I am forced to use a soundcard for now.

 

[ad674]

Great thread!

Big thanks to Renethx for the exhaustive review and information in the first post.

Have had one of these boards now for about 6months and its great, however, I have recently upgraded from an 80GB PATA drive to a 250GB SATA II drive and it seems than standby S3 and hibernate will now no longer work. After re-installing the OS and installing all my drivers and apps etc, I realised that the PC would not enter standby or hibernate correctly.

The symptoms are:

1) when selecting Standby on the start menu, windows claims to be
entering standby (blue screen, says entering standby) then the LED on
the case switches from green to orange but only for a split second
before it goes out completely. When I try to wake the machine I have to
use the power button and the system reboots as if it was shutdown.

2) when selecting Hibernate on the start menu, windows claims to be
hibernating (black screen and progress bar, says hibernating). When I
try to wake the machine I have to use the power button and the system
reboots as if it was shutdown.

Before I replaced the HDD the system woke from standby via the MCE
remote, keyboard and mouse without any problems, its only since
upgrading to the 250GB SATAII drive that the problem has occured.

I've double checked the BIOS settings and even re-installed windows as
I read that installing without S3 set in the motherboard BIOS can
prevent windows from ever going to S3 correctly.
Ive made sure the OS is upto date with all the microsoft updates. Ive
added the registry fix for resuming from S3 via USB and Ive even played
around with dumppo. Ive flashed the BIOS with all the versions from 3.10 to 3.50 but to no avail.

Can anybody suggest anything?

Hardware Spec:
Athlon 64 3000+
MSI K8NGM2-FID
512MB DDR 400
250GB Samsung SP2504C.
MCE 2005.

 

[grits]

I can't help with your not going into S3 now, but I can confirm that S3 works fine even when installing the OS with bios=S1. As you noted, the registry patch is only needed for MCE resume by remote. BTW, I use a WD sata1 drive.

Try unchecking 'enable hibernation' in power management.

 

[chinna]

could this have to do with your power supply?

cause when mine is supposed to be at 1.45 it does jump around but it jumps more up than down (1.448-1.488 during prime)

Nope, not at all. CPU power is taken from 12v 4 pin plug, and 12volts never fluctuated and never dropped anything less than 12.10v. For that matter none of the PSU rails fluctuated. 3.3 is at 3.38v, 5.0v at 5.05v. Ofcourse just to make sure, I measured voltage with DMM, as well tried two PSU ( one Antec Smart Power 350W, another ThermalTake PurePower 420w). CPU load is hardly 65watts at default vcore, so it only takes about 6 AMPs on 12v rail. It is rated for 21 AMPS.( as per many calculations online even with OC my rig only takes about max 160-180 watts).

It is quite commong among many motherboards with fluctuating vcore( it is based on VRM design and howmany phase VRM it is etc etc).

But some fluctuate more than other. My previous ASUS board used to fluctuate very little at 1.525v, it used to give anywhere from 1.51v to 1.53v, but my previous ABIT and AOPEN boards used to fluctuate even more. Most of those we see some sag in vcore at full load. But this MB it is nothing to do with load, it fluctuates as much at idle also.

 

[avi85]

new onboard sound drivers are available here.

 

[vsonerud]

 

[vsonerud]

Does anyone know what PCI revision the MSI K8NGM2-FID and K9NGM2-FID cards support? PCI 2.1 / 2.2 / 2.3 ?

In the manuals for the 2 cards (downloaded from MSIs website) it only says:

* Two 32-bit Master 3.3v/5v PCI Bus slots

Reading up on the PCI standards, Wikipedia for instance says the following:

# PCI 2.3 permitted use of 3.3 volt and universal keying, but did not support 5 volt keyed add in cards.
# PCI 3.0 is the final official standard of the bus, completely removing 5 volt support.

Does this mean that there still is some kind of 5 volt support in PCI 2.3, just not for so-called '5 volt keyed add in cards'?

The reason for me wondering is that I am about to upgrade my old AMD Athlon XP 2200+/Asus A7n8X Deluxe (nforce2 ) system, and the MSI K8NGM2-FID and K9NGM2-FIG cards have caught my attention as potential motherboards to buy to achieve the cheapest possible upgrade path to a S939/AM2-system.

In my current old system I have a Hauppauge WinTV PVR-350 card which I definitely want be able to reuse when upgrading. This card however is a PCI 2.1 card. I have e-mailed Hauppauge support about the PCI revision support, and specifically about PCI 2.3 and this was the reply I got:

>PVR350 is a 5Volt card and Keyed for 5 Volts. , so PCI 2.3 would not be
>compatible.

I am therefore very interested in making absolutely sure that the K8NGM2-FID or K9NGM2-FID cards are compatible with my PVR-350 card.

Maybe someone else here uses the PVR-350 card with any of these motherboards?
(note: My PVR-350 is an old hardware revision board - from Sept 2003 )

 

[avi85]

Originally posted by: vsonerud
Does anyone know what PCI revision the MSI K8NGM2-FID and K9NGM2-FID cards support? PCI 2.1 / 2.2 / 2.3 ?

In the manuals for the 2 cards (downloaded from MSIs website) it only says:

* Two 32-bit Master 3.3v/5v PCI Bus slots

Reading up on the PCI standards, Wikipedia for instance says the following:

# PCI 2.3 permitted use of 3.3 volt and universal keying, but did not support 5 volt keyed add in cards.
# PCI 3.0 is the final official standard of the bus, completely removing 5 volt support.

Does this mean that there still is some kind of 5 volt support in PCI 2.3, just not for so-called '5 volt keyed add in cards'?

The reason for me wondering is that I am about to upgrade my old AMD Athlon XP 2200+/Asus A7n8X Deluxe (nforce2 ) system, and the MSI K8NGM2-FID and K9NGM2-FIG cards have caught my attention as potential motherboards to buy to achieve the cheapest possible upgrade path to a S939/AM2-system.

In my current old system I have a Hauppauge WinTV PVR-350 card which I definitely want be able to reuse when upgrading. This card however is a PCI 2.1 card. I have e-mailed Hauppauge support about the PCI revision support, and specifically about PCI 2.3 and this was the reply I got:

>PVR350 is a 5Volt card and Keyed for 5 Volts. , so PCI 2.3 would not be
>compatible.

I am therefore very interested in making absolutely sure that the K8NGM2-FID or K9NGM2-FID cards are compatible with my PVR-350 card.

Maybe someone else here uses the PVR-350 card with any of these motherboards?
(note: My PVR-350 is an old hardware revision board - from Sept 2003 )

I'm not using a Hauppauge TV card but my TV card is a PCI 2.1 from like 7-8 years ago (believe it or not, it still works great) and I use it in this motherboard, although I may have to finaly upgrade when vista comes out cause there probably won't be any vista drivers for my card.

 

[mshan]

I just got a Divo Fusion HDTV 5 Lite HDTV tuner card and am wondering which PCI slot will minize IRQ conflicts (I am using integrated graphics)?

PCI slot #1 or slot #2?

 

[renethx]

K9NGM2-FID is in stock at Buy.com (the lowest price) and several merchants.

 

[jaeaetee]

I have a playback quality problem; I'm using this k8ngm2-FID board with integrated vga..

When playing MPEG-2 hdtv content through Nvidia Purevideo decoder, I get stuttering and shaky image, which is most noticeable on subtitles (subtitles "burnt" into the picture). Also the playback performance seems to be disappointing With the MPC integrated mpeg-2 decoder the picture is sharp and stable, no twitching at all.. I've tried various combinations of output settings in Media Player Classic, the overlay has best performance (but still twitches a little) but poorest pic quality

Here's what I'm trying to use:

-k8ngm2-fid
-athlon xp 3500+
-2 gigabytes of ddr400 (installed correctly, double and triple checked )
-integrated vga connected to viewsonic vx2025vm 20" monitor through dvi
-xp pro sp2
-all latest drivers for nvidia vga and others
-newest, updated bios
-newest version of Nvidia PureVideo decoder
-media player classic / media player 9

If I've understood right, the performance of the setup should be at least sufficient for playing mpeg-2 hd content.. I can play 1920x1080 x264 with FFDSHOW or CoreAVC decoders just fine, especially with coreavc no problem.

Any ideas? Or is the 6150 + nvidia purevideo just not good enough for HD material with hardware acceleration?

 

[mrnuxi]

Originally posted by: renethx
K9NGM2-FID is in stock at Buy.com (the lowest price) and several merchants.

newegg now has it for 79.99 + shipping....

 

[renethx]

I looked at images of K9NGM2-FID at NewEgg and noticed that a TV-Out bracket is not included. If this is true (usually product images at the NewEgg site are very accurate), ASUS M2NPV-VM may be better for some.

• ASUS M2NPV-VM: TV-Out bracket: YES, SPDIF bracket: NO, Audio: 5.1 channels, Overclocking options: FSB up to 400MHz (corrected), but actual overclocking is poor.
  
• MSI K9NGM2-FID: TV-Out bracket: NO, SPDIF bracket (RCA only): YES, Audio: 7.1 channels, Overclocking options: None (use ClockGen or nTune to overclock)

If you don't care about reusability of DDR2 memory modules in future upgrade, K8NGM2-FID is still the best GeForce 6150 board.

 

[t3hPwnag3]

Thank renethx for the awesome guide. It's helped me alot. I built first then found this and used the quick recipe for extra value in this budget build.

thermaltake full size atx case v7000c
thermaltake 430w psu
geil golden dragon ddr433 (2x256)
maxtor 80gb ata133
maxtor 200gb ata133
acer 1916w 19"lcd
athlon 64 3200+ currently 2.44ghz

From what i've seen this is the board to for people like me on an extremely tight budget but still has expandability like x2 compatible (even though not listed on newegg as such)
and gigabyte lan

I've used two different ram: original 2x512 = 1 gig of geil pc3200 2.5-8-4-4 (bought for this build) Using stock voltage only able to get to 232 with on board video running 64mb.

the ddr433 geil mentioned above was pulled from my other system just to see if was able
higher. which it was. 244 cpu-ldt using stock voltage and 64mb video.

I would like to mention, if you are using a full size atx case with this board you feel kind
of stupid because the board is so small and if you to install a video card later (like I do) you will run into problems with your jumpers not being long enough for your tv-out connector (jtv1).

love this mb! thanks again for the time and effort put it

 

[winguy]

Originally posted by: renethx
ASUS M2NPV-VM: FSB up to 240MHz

Both the specifications page and product manual shows up to 400MHz.

But the lack of other settings renders it useless.
http://vip.asus.com/forum/view.aspx?id=...&model=M2NPV-VM&page=1&SLanguage=en-us

 

[renethx]

I have added three small subsubsections as part of motherboard selection guide:

• Section 2. Choosing a GeForce 6150 Board > Available MicroATX Motherboards with GeForce 6150 and Other Equivalent Chipsets > 3. Socket 939 Radeon Xpress 200 Boards
  
• Section 2. Choosing a GeForce 6150 Board > Available MicroATX Motherboards with GeForce 6150 and Other Equivalent Chipsets > 4. Socket AM2 Radeon Xpress 1150 Boards
  
• Section 2. Choosing a GeForce 6150 Board > Available MicroATX Motherboards with GeForce 6150 and Other Equivalent Chipsets > 5. Core 2 Duo Boards

 

[juninjesus]

I accidently posted this. How can I delete this posting?

 

[juninjesus]

Same as me..

What ever I did, clock was not changed either...
So, I stucked at 2.5GHz... My CPU could go 2.6GHz @ stock 1.35v

Are there any way to get higher clock speed?

Originally posted by: benp
Hmm, I have: BIOS 3.50, A64 3000+ e6 stepping. With Clockgen 1.0.4.7 the chipset is detected as Nvidia 6100, and the clock speeds are read correctly, but any changes I make aren't applied. It looks like it works, but the clocks don't change and nothing melts when I set it to 4000 .

Are there any BIOS settings that could cause Clockgen to fail? Does anyone have it working with the 3.50 BIOS?

 

[renethx]

Originally posted by: juninjesus
Same as me..

Originally posted by: benp
Hmm, I have: BIOS 3.50, A64 3000+ e6 stepping. With Clockgen 1.0.4.7 the chipset is detected as Nvidia 6100, and the clock speeds are read correctly, but any changes I make aren't applied. It looks like it works, but the clocks don't change and nothing melts when I set it to 4000 .

Are there any BIOS settings that could cause Clockgen to fail? Does anyone have it working with the 3.50 BIOS?

In my system the problem occasionally occurs when both ClockGen 1.0.4.7 and CPU-Z are running at the same time. ClockGen 1.0.4.2 does not suffer from the problem.

Using nTune?

 

[juninjesus]

I tested both versions with/without CPU-Z. Still Same. And nTune is not helpful. nTune allowed upto 250MHz HTT. Well... are there another way?

Originally posted by: renethx

Originally posted by: juninjesus
Same as me..

Originally posted by: benp
Hmm, I have: BIOS 3.50, A64 3000+ e6 stepping. With Clockgen 1.0.4.7 the chipset is detected as Nvidia 6100, and the clock speeds are read correctly, but any changes I make aren't applied. It looks like it works, but the clocks don't change and nothing melts when I set it to 4000 .

Are there any BIOS settings that could cause Clockgen to fail? Does anyone have it working with the 3.50 BIOS?

In my system the problem occasionally occurs when both ClockGen 1.0.4.7 and CPU-Z are running at the same time. ClockGen 1.0.4.2 does not suffer from the problem.

Using nTune?