http://setiathome.berkeley.edu/show_host_detail.php?hostid=2302665

It may be a single die, but four cores equals four cpus.

...in fact, I kinda wish [Apple] had gotten together with IBM or some other PC/Clone company years ago and shared ideas. Who knows where computers would be now if that had happened?

Umm … have you ever heard of a RISC processor called the PowerPC, the later versions of which are known as the G3, G4, and G5? It was developed as a joint venture between Apple, Motorola, and IBM. Only when it became apparent that IBM wasn't able to make G5(+)s low enough in power consumption at the clock-speeds Apple wanted was the switch to Intel made, Motorola having been out of the picture since the ‘underclocked G4’ fiasco a few years back.

The two active machines on your account show they are doing work.

Are you running more than one project? If so, sometimes the debt of a project gets to high, and will not request work until the debt is paid off. Then it will get new work, again.

Anybody have any experience with ASUS K8N-DL's?

Pros, Con's, Problems?

Alinator

Let me jump in here and either explain this or add more to the confusion! Haha!!!
Currently we have a problem with the credits issued for various angle range work units. It is hopefully going to be corrected in the future. However right now there will be instances like you have seen, where some very low angle range units take an extra long time to crunch but don't give extra credits. I think the point that is being made is there is a difference in the times needed for various floating point operations but some people are under the impression that all work units require the exact same kind of operations. This is not true. There are different kinds of operations carried out on a work unit depending on it's angle range. These calculations look for pulses, spikes, triplets and gausssians. Each type of search requires it's own combination of mathematic operations. Up to a certain point as the angle range decreases the amount of calculations increases. Also the complexity of calculations increases. Until a new application is created that takes this into consideration we will still see the variations in credits not matching the time spent for all work units.

If the same work unit is sent to a P4 operating at 3.0 Ghz and a K6II operating at 550 Mhz shouldn't the overall credits be different? The P4 will have much more on chip cashe memory, possible hyperthreading, and SSE2 and SSE3 instruction sets; where the K6II will only have a fraction of on chip cashe memory and only SSENOW instruction sets.

Nope.
As the stock client is designed to be used by all suitable x86 CPUs, it doesn't take advantage of things such as SSE2 etc as many x86 CPUs don't have it. A cache does speed things up on Netburst & Core based CPUs, but it doesn't affect the number of operations that need to be done.