Disclaimer: Not linux related and slightly spurred on by windoze necessity.

So I ended up with a couple of old Dell computers from the lab last week. They are GX 9000 something or others, made in 2005. Point is Win7 will install on them which for me is important because the newest versions of Solid Edge and Solid Works will not work on XP. My state of the art (I built it in 2005) gaming computer that I have been using for CAD for the past 5 years will only run Vista, which I refuse to use.

The problem lies in the thin profile cases of the Dells. My video and sound cards will not fit with the cases closed, so I want to swap the mobos and put the Dell chipset into my old tower. The question I have is this: the dell mobo has grounded clips instead of screws to attach to the case. There are mounting holes on the mobo that will match the bolt pattern on my case but is it necessary to have more than one ground point to the case? The clips are metal and therefore ground to the case but there is only one of the mounting holes that is grounded and the rest are not.

In My experience desktops have NO ground point via the Motherboard.
Only via the power supply.
This best practice is not to have earth/ground loops, (may cause Radio frequency radiation and other problems)
Dell are unusual in their earthing arrangements, probably get away with it by heavy case shields.
http://en.wikipedia.org/wiki/Ground_...electricity%29

I hope this helps
Trev

On my ATX desktop motherboard the holes for attaching the motherboard to the case have contacts around them suggesting that the screws used to attach the motherboard to the case are also contacts fro ground. The case I use (Fractal R4) is also screwed and riveted together suggesting that all the metal is electrically connected. The case doesn't have a dedicated ground connection though so there is no ground loop -- in fact I think in actual fact it's the motherboard which grounds the case and not the other way around.
The above said I have read of people using wooden cases with no grounding so it seems if my case is providing a ground that isn't a must.

Have you tested continuity from the pads to ground 0V.
It may look like ground pads but unlikely.
Read the Wiki it will explain the necessity to use only one point as Ground / 0V.
please do not confuse Mains Ground/Earth with 0V ground on most computers they are not connected DC but AC only via a coupling capacitor.
The final test is does the motherboard work on a insulated top or stack of unprinted paper, ( print may contain carbon a conductor )if it does then leave the earthing to the power supply.
Trev

I haven't tested the pads, no, so I suppose the metal may just be there for mechanical reasons (less resistance to the bolt heads than the motherboard material). However, if they are linked to the motherboard it will, of course, be the motherboard grounding the case and not the other way around. I'm not suggesting either that the case needs to be grounded in order for the PC to operate just that from what I can gather it may be grounded for reasons like RF shielding. On trying to google this it seems that everybody out there is confused about the topic.

The problem is the word Ground and 0 volt mean differing things depending on how they are connected.
Absolute ground is direct connection to mother earth often via a copper plate or on a water going vessel a anode ( possibly Zink)
Mains ground or Earth is connected via a true earth connection on the nearest transformer or in the case of multiple earth the nearest pole or distribution.
Mains Neutral should be the center point of a 3 phase mains network and should be connected to true mother Earth at the distribution point.
Now to the Computer case it should be connected to a true mother earth via the mains plug or in the case of double insulated power supply via a house Earth point.
0 Volt earth of the mains power pack need not be connected DC wise via the Mains in fact it is often not, however often is AC wise via a capacitor.
So if it works out the case do not worry.
After all mobile PCs work without an Earth.
Now we can see how confusion comes about and how the designers of the mother board has chosen how it will work. it is not wise to circumvent this intention.
I would like to say I hope this explanation helps but know it only makes it more difficult.
I Have tracked down many many Earth/Ground loop problems in my Engineering career, and never found a universal solution for all events, nor to my knowledge is one documented.

I spent two years in electrical engineering before I switched to mechanical engineering so I do understand the difference between "GND" and earth Ground. I always assumed that the case groundings (as noted by the silkscreen on the pcb) were just there to prevent static buildup on the board. In conjunction with everyones input, I assume that using just the one ground point will suffice.

I had this problem last semester in the lab. I designed 2 piezoelectric driver circuits for fabric impact stress analysis. The Jaguar (mechanical vibrations daq hardware, runs on Open Suse yay!) had no problems with the circuits but the other computer we were using to take temperature/pressure etc. data had massive ground loop problems that took me a month to figure out. Both circuits were exactly the same but the acquisition hardware was not, it was a nightmare.