HD2400Pro and 3870 on the same core?

I have a folder (4800+ @ 2.7, 3GB RAM, 16+8X PCI-E, Win XP), in which I'm currently running an overclocked 3870 GDDR4 and the SMP client on the free core. Today, I got my hands on a 2400Pro in a dumpster today, and I wonder if I'll be able to run both cards on one CPU core. Will there be any performance issues related to it?

I'm asking, as I have a free PCI-E slot in it, and I'd like to get that 300PPD the 2400Pro can deliver. If it won't run on the same core, the CPU produces almost twice that, making it stupid to even run fah on it.

Are you sure you're making the preferred and final SMP deadlines on one core with the SMP client? AMD x2s were close on it with both cores. If you're not making the preferred deadlines, you're generating points, but not doing any science. If you don't make the final, you don't even get any points.

I'm not sure if ATI cpu utilization has been reduced to the point you can run two clients on one core.

It makes around 450-550PPD on that core. I've seen CPUs make the deadlines with less.

To change the subject a bit, I have a problem even getting both the cards running at the same time. When the 2400 kicks in, the 3870 goes down to 15-30%. Switching to the console client won't do squat, nor will reinstalling drivers, running with force gpu_r600 flag or setting affinity to different cores all over. What should I do?

EDIT: Yep, the 4800+ on a single core falls way within deadlines, even on the stock 2.5GHz:

p2665 is the only SMP project with a 6 day deadline. All the other projects have 3 day preferred deadlines or less. On p2605 you must make at least 587 ppd to make the preferred deadline. You're right that you're well within the deadlines on p2665, but I wouldn't bet on it with other WUs. If you can't make the preferred deadline, you're hurting the project rather than helping it and should switch to the unicpu client.

Did you use the -gpu flag to assign gpu numbers to each client. THe first gpu must have -gpu 0 in the startup string and the second must have -gpu 1. Without that both clients run on the same gpu. I think you'll need a third party affinity program to force both GPUs on to one core.

I am running one SMP and one GPU client on my X2. @ 2.7 Ghz he should be making the preferred deadlines even on the three day ones. I am making them with about 12 hours to spare on my system @ 1.9GHz (clocked back down to stock due to new RAM). I would not however run two GPU clients with SMP. It's worth a shot to see what happens but I doubt SMP will make the first deadline.

From time to time you will pick up a p305x or p306x with less than a two day deadline and a production rate less than that of p2605. Neither of you will make those deadlines. Fortunately they don't get assigned to dual cores unless the DC server is down.
You are using both cpu cores to run the SMP client aren't you and even then it makes a large difference if the GPU client is an ATI due to its much higher cpu utilization? THe OP is going to be limited to one effective core when running two ATI GPU clients and won't make the deadlines, as you point out.