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

[BernardP]

OK guys, enough of this light stuff. Let's get serious now. Do I have a question for you!

If the "CSM" in ASUS A8N-VM CSM stands for "Corporate Stable Model" as explained in the "Why choose MSI over ASUS section" of the firrst post...

...Then what is the meaning of "FID" in K8NGM2-FID?

 

[renethx]

Originally posted by: BernardP
...Then what is the meaning of "FID" in K8NGM2-FID?

K8NGM2-FID comes with DVI and IEEE1394, Gb LAN
K8NGM2-IL lacks DVI; comes with IEEE1394 and 10/100Mb LAN
K8NGM2-L lacks both DVI and IEEE1394; comes with 10/100Mb LAN

Logical geuss from the above data is:

• F = Fast Ethernet
• I = IEEE1394
• D = DVI
• L = 10/100Mb LAN

 

[mikeaco]

Well, Olive Yew, thanks for the detailed test steps. Unfortunately, I'm running memtest right now and it's failing at 1.75V, 200MHz LDT, 1400MHz CPU and 3-4-4-8 timings. The same Err-Bits are reported (0x00004000 ... I guess it wasn't actually bit 4 like I mentioned above). Looks like I have bad RAM, huh? Bummer...

 

[renethx]

Originally posted by: Olive Yew
The 1st step for any new system build should *always* be to test the memory. I can't stress that enough. It is very bad practice to start stress testing other components in your system without first determining if your memory is functioning without errors. If you jump ahead to test the CPU or the limits of your HTT, you'll never know if any errors that pop up are due to your memory or the limitations of your other components. Even if you've tested the memory you intend to use previously on another system, you still need to test it again. No mobo in existence will work flawlessly or optimally with all available types of memory. The K8NGM2-FID is no different.

Olive Yew, thanks for the opinion. I ocassionally see similar claims like "Checking RAM for errors is a must in todays systems even if you dont have any problems with the system running properly.", but always with the lack of documentations or enough evidence to back up the claim. I systematically neglected memory test options in the main post based on my own experience of dozens of memory modules in the past. It is hard for me to believe that significant percentages of brand new DDR400 modules cause errors at less than 210MHz that forces every user to test memory first. But I may be ignorant. Will you cite documentations or tell your experience backing up your claim *always*? You seem to have enough experience concerning this matter.

 

[renethx]

Originally posted by: cupajava
Even though you're using a divider, more vdimm is required to support the higher FSB.

Do you mind citing documeantions or users' experience that support this claim?

 

[BernardP]

Originally posted by: renethx
F = ?
D = DVI
I = IEEE1394
L = 10/100 LAN

Seems obvious after the fact...

I am not too proud of myself.

There is no L in FID.

Other MSI boards have an -F suffix

The F seems to relate to the Gigabit LAN, but why the letter F? Fast?

 

[mikeaco]

I have just retested each of my DIMM modules individually (not in Dual-Channel), with the same settings listed above, in memtest86+. One of them PASSES and the other one FAILS. The one that fails has the same failure (Err-Bit 0x00004000) as the two module do while running in Dual-Channel.

I have submitted an RMA with Corsair. We'll see how quickly they can process a replacement.

Just FYI..a similar problem was described HERE.

Mike

 

[renethx]

F = "Fast" = Gigabit LAN is a good guess.

 

[jt2417]

Just got this new board and have a strange problem with it. I put a new, unformatted SATA drive on channel SATA1. I put my old IDE drive in the IDE port 1, and my DVD drive in the IDE port 2. For the life of me, I can't get the mother board to recognize that there is anything attached to the IDE ports. But, the SATA drive registers. As it stands now, I can't access anything to format the SATA drive or install windows. Help!!!!

 

[mikeaco]

I just went through a similar thing. The way things were laid out in my case, I needed to have the SYSTEM HDD on the Primary IDE channel and the DVD/RW on the Secondary IDE channel. I also had a SATA on SATA1. When initially booting into the BIOS, the system recognized everything but the IDE HDD. The problem was that I had the jumper on it set to Master, and apparently that's not the right thing to do if you only have one thing on an IDE channel. When I changed it to Cable Select, everything was fine.

Mike

 

[jt2417]

Thanks Mikeaco, I am not next to the system right now, but I do remember that I had the jumpers on the IDE drive and DVD/RW set to master. I will check this out and see if it works!

 

[cupajava]

Originally posted by: renethx

Originally posted by: cupajava
more vdimm is required to support the higher FSB

Do you mind citing documeantions or users' experience that support this claim?

I should have mentioned that my comments are pertaining to overclocking his 3200+ Venice E6 to 2.5ghz with that value ram. All I have to back the claim is my own experience. I found that a 25% overclock with that ram (regardless of which A64 cpu) usually requires a small bump in vdimm for stability at default timings of 2.5-3-3-8 1T. I am assuming he has the dual channel kit like mine. Didn't mean to mislead anyone.

 

[renethx]

Thanks, cupajava. I had a similar experience to you with ABIT AX8. Memory ran at less than 200MHz, but required higher voltage with higher HTT (like 290MHz). Curious but true.

 

[Olive Yew]

Originally posted by: renethx
Olive Yew, thanks for the opinion. I ocassionally see similar claims like "Checking RAM for errors is a must in todays systems even if you dont have any problems with the system running properly.", but always with the lack of documentations or enough evidence to back up the claim. I systematically neglected memory test options in the main post based on my own experience of dozens of memory modules in the past. It is hard for me to believe that significant percentages of brand new DDR400 modules cause errors at less than 210MHz that forces every user to test memory first. But I may be ignorant. Will you cite documentations or tell your experience backing up your claim *always*? You seem to have enough experience concerning this matter.

That is an interesting question you ask. But how do you know your system is operating properly without testing it? Renethx do you have some special method of determining if your computer is operating without any errors without testing it? If so, please enlighten me. If not, then you're either avoiding the issue or just hoping/praying/assuming it doesn't. Given that this and every other computer related forum is littered with troubleshooting issues, I'd say that's a foolish assumption to make.

Documentation? There's plenty of documentation. Where have you been the past year since the NF4 chipset was introduced? Check on the various online forums and you will see hundreds upon hundreds of problems with memory compatibility and NF4 chipsets. DFI Lanparty NF4 users in particular got the short end of the stick with the earlier revision of that mobo. Cold boot problems, problems with compatibility, failing to POST, RAM sticks failing after 2-3 months, etc. It's no secret. Representatives from DFI and OCZ have come out and publically admitted there were issues and moved to resolve them. Some of the forum threads regarding these problem run into the thousands of replies. MSI's own Neo4 still suffers from compatibility issues with late Winchesters and early Venices that prevents it from surpassing 219-229FSB with some memory. Something that MSI still refuses to fix (if they're even able to) with a BIOS revision.

While your own experiences may be positive, it's hardly statistically relevant compared to the masses of user reports you can find by just scouring the web. Ram still arrives DOA or not up to spec. Various mobos will have compatibility issues with various types of RAM. Not all RAM can overclock even 10mhz. Of the ones that can, not all are guaranteed to be able to do it default settings. So user error becomes a factor as well.

My question is how can you *not* test your RAM? Do you have some way of knowing if your memory is operating as intended with no conflicts or errors without testing it? If not, then you're just rolling the dice and praying nothing is wrong. It's up to you if you choose to live with that. But please don't try and convince people that it's not necessary. It does a disservice to anyone who's ever experienced memory related problems. And to be blunt about it, anyone who can't be bothered to leave their PC running once overnight to check for memory errors really should reexamine their situation and consider going to premade systems like Dell who take care of their own QC.

 

[Olive Yew]

Originally posted by: mikeaco
I have just retested each of my DIMM modules individually (not in Dual-Channel), with the same settings listed above, in memtest86+. One of them PASSES and the other one FAILS. The one that fails has the same failure (Err-Bit 0x00004000) as the two module do while running in Dual-Channel.

I have submitted an RMA with Corsair. We'll see how quickly they can process a replacement.

Just FYI..a similar problem was described HERE.

Mike

Well it sucks that your memory is bad but I'm glad you found your problem. Hopefully you won't have to wait too long for a replacement set.

 

[Bozono]

I plan on purchasing this and an X23800 on Monday, but I have been getting conflicting info. on whether or not the TV-out bracket is included or optional. I checked MSI's site and it lists it as optional. So is it included?

 

[ibnarabi]

I just got the board in the mail today from zipzoomfly (it was $90 3 days ago now its $110, they must be running out of them).
Anyways, it DOES have the tv out connector on a bracket included with the board, RCA composite and S-Video. As is a 1394 bracket, a serial port bracket, and an audio out bracket.
It ships with the crappy 3.0 bios though.

I hope that helps

 

[renethx]

I checked at PriceGrabber and found that right now only 4 merchants have the MSI K8NGM2-FID in stock. This motherboard may be totally out of stock in the US market shortly.

 

[renethx]

Olive Yew,

Thanks for the reply. Unfortunately you confused the essence of the problem with other non-relevant problems and cited irrelevant "documents". You need to clarify your logical thinking seriously.

What you are confusing is the nature of memory errors:

1. memory errors caused by the incompatibility with the motherboard
2. memory errors caused by the defective memory chip
3. memory errors caused by other factors (soft error)

You are discussing the first two cases simultaneously and confusing them (deliberately or non-deliberately). I had a DFI NF4 LAN Party board that caused lots of memory errors while the memory modules themselves were nondefective (they passed Memtest86 in another motherboard). Poor crappy NF4 chipset and/or poor implementation of the chipset in the motherboard has caused numerous memory-related problems that are the subjects of the "documents" you cited. However these documents have nothing to do with the K8NGM2-FID or memory modules per se. You also wrote

The K8NGM2-FID is no different.

Right now there are three major threads on the K8NGM2-FID motherboard with more than 1200 posts in total:

• NV GeForce 6150 Board Shootout in AnandTech Forums
• MSI geforce 6150 instock @ the egg in Hard|Forum
• Nvidia 6150 motherboards and htpc build questions/help in AVS Forum

However zero compatibility issue with memory has been reported. This stands in marked contrast to NF4 boards. Of course nobody can assure that the motherboard is comaptible with any kind of memory. But as the author of the main post of this thread, I don?t feel any reason for warning users of de facto non-existent compatibility issue with memory of this motherboard.

Memory errors caused by a bad chip is a totally different subject. This includes mikeaco?s case. This could happen to everybody irrespective of the kind of motherboard, but the probability of picking up a bad module is perhaps very small (though no statistics exists). The thread?s subject is MSI K8NGM2-FID and clearly I don?t need to include this issue particularly in the main post. (If I had to include all of these generic kinds of issues, perhaps I would have to write a book of building PC.)

In my impression, your opinion is strongly influenced by your sad experience with NF4 chipset. However you cannot simply extrapolate GeForce 6150 from your experince with NF4. In fact your worry does not apply to K8NGM2-FID. Please tell me if I missed a critical issue that affects seriously many users' computing experience (if you want to discuss memory issue, please separate the two cases clearly and discuss logically; please always include proper documentation as I cannot trust rumors or mere speculation; and don't mention your experience with NF4 any longer, I am sorry that your experience with NF4 is totally useless in GeForce 6150; nVIDIA should have never released such a crappy chipset).

 

[jt2417]

I always order from Newegg, but I have been waiting since Feb 2 for them to get this back in stock. Went with ZipZoomFly, and they shipped it for $85 in two days. Yes, it is getting very hard to find this board- people must like it.

 

[aeroguy]

Originally posted by: ibnarabi
I just got the board in the mail today from zipzoomfly (it was $90 3 days ago now its $110, they must be running out of them).
Anyways, it DOES have the tv out connector on a bracket included with the board, RCA composite and S-Video. As is a 1394 bracket, a serial port bracket, and an audio out bracket.
It ships with the crappy 3.0 bios though.

I hope that helps

That's funny. I ordered Tuesday night from zipzoomfly @ $99 (free shipping). Also got an OEM Microsoft Remote Keyboard for MCE for $57. Everywhere else is selling these for $80, and up to $105 at Worst Buy.

 

[aeroguy]

Originally posted by: renethx
I checked at PriceGrabber and found that right now only 4 merchants have the MSI K8NGM2-FID in stock. This motherboard may be totally out of stock in the US market shortly.

Actually, only 2 of those places have stock when you click on the link, and Super Warehouse lists it as 'discontinued.' What's up with that?

 

[Olive Yew]

Originally posted by: renethxThanks for the reply. Unfortunately you confused the essence of the problem with other non-relevant problems and cited irrelevant "documents". You need to clarify your logical thinking seriously.

It seems almost as if you want to ignore any issues that pop up so as not to tarnish the image of the K8NGM2-FID. How are they non relevant? Memory errors caused by any reason are non desirable. Testing by Memtest86+ does not distinguish why memory might not be performing properly so much as if. And to the end user that's what matters. Whether memory errors are due to revised a BIOS that changes memory management strategies (already reported) or otherwise. What you're essentially arguing is that it's *ok* to have memory issues if they're not due to physical problems with the RAM. That's some seriously twisted logic as I can see it.

Also, I have to chuckle a bit at your assumptions. Because you have not personally read about it you assume that every single comp built with this mobo has been 100% trouble free? Though having read through several forums I've seen more than a few reports of users having some memory related trouble with this mobo. Not shocking by any means for any mobo. Most people who overclock are rather competent and can actually troubleshoot themselves. And if they can identify errors with Memtest and correct it with a few tweaks, it's hardly worth posting about. Though that's not's quite so obvious for someone new to system troubleshooting. Renethx I assumed you had ample experience with proper proceudres for overclocking. But I do have to question this assumption now. Especially seeing as how you need proper "documentation" (what exactly constitues that on an internet forum?) for rather common issues that deal with any computer build. Any computer related forum will recommend that you test your memory with Memtest with overclocking. I don't know of ANY userbase that recommends not testing your memory when overclocking.

I'm still rather puzzled at your logic in this case. You recommend testing with Prime95 but you cannot recommend testing with Memtest86+? So CPU's can have issues when overclocking but memory can't? You've ducked the original question I posed so please do me a favor and answer it now. How do you know your memory is operating properly without testing it?

 

[imported_derekn]

That's funny. I ordered Tuesday night from zipzoomfly @ $99 (free shipping). Also got an OEM Microsoft Remote Keyboard for MCE for $57. Everywhere else is selling these for $80, and up to $105 at Worst Buy.

I wish I didn't wait too long to order this board. I saw the $89. price from zipzoomfly earlier this week on Tuesday. It was $99. on Wednesday and it was $109. yesterday. I ordered it yesterday. I thought I got a good price $90. from Amazon.com on Wednesday but it ended up being backordered.

Now, I am wondering if I should get another board since I will probably like this board so much I will want to build another one.......

 

[area123]

How about overclocking the Opteron 146, would the performance still be better on the venice 3200+? I can't decide whether to get the 146 or the Venice 3200+, the 146 is only about $20 more on sale at Monarch.