CPU Question

I know I'm probably the king of the stupid question, but....

Will having more cores on the CPU help running 4 instances of EQ2 on a single computer? Can ISBoxer manage 6 cores?

[lax]

More cores is always better for running more instances, until a point. Ideally, you want a CPU core per game instance, so if you're running 4 copies of a game you ideally want a 4-core. A 6-core is likewise better than a 4-core, particularly if you are running more than 4 game instances.

Someone recently asked about comparing an Intel i7 (4-core with hyperthreading to trick Windows into thinking it's 8-core) to a 6-core AMD Phenom. I haven't seen relevant benchmarking and don't have a Phenom to compare with (while I do have i7 boxes). However, I would expect the 6-core Phenom to outperform the 4-core i7 in a multboxing scenario, especially if playing more than 4 instances.

As far as where "Until a point" comes in, there will be diminishing returns as the number of cores outpaces the number of game instances. Most games don't heavily multithread -- they almost all "have multiple threads" but other than the main (game logic, rendering, etc) thread they will typically be idling. So that means that additional cores beyond 1 per instance will still somewhat help, but nowhere near as significantly.

And finally, yes ISBoxer will help you configure all of your CPU cores, until you pass 32 or so of them but that's a few years out

I'd like to run 6 instances of EQ2 on the same machine. I currently run 4 using an E8400 dual core OC'd to 3.96ghz. I have to run all instances at maximum performance and still suffer fps issues.

So I'm thinking a new Gulftown 6 core processor is the way to go? Or will a Quad core do it? Does the program see 8 cores or just 4?

Thanks so much Lax!

[lax]

A 6-core processor would be ideal for running 6, but a quad core will work fine. You'd be doubling up on some cores, but not all of them -- and therefore 6 on a 4-core would outperform 4 on a 2-core. The program will see whatever number of cores is seen by Windows; for a normal quad core that's 4, for an i7 that's 8 (with half of them being fake).

I run 6 Lotro clients on a quad core machine (8 virtual cpus). The CPU is almost never the bottleneck for me and the active virtual cpu's (the ones with a Lotro instance) hover around 50% most of the time when I'm playing. The virtual threads from hyper threading are certainly helping. Most cores with only 1 thread can not completely utilize the execution units, especially when a thread is stalled when fetching something from memory.

That said, I'm getting to test a 6 core (12 virtual cpus) chip soon, so I'll post back if I notice a difference. I am not expecting much as my gfx card and hard drive are usually the bottlenecks for me.

I run 6 Lotro clients on a quad core machine (8 virtual cpus). The CPU is almost never the bottleneck for me and the active virtual cpu's (the ones with a Lotro instance) hover around 50% most of the time when I'm playing. The virtual threads from hyper threading are certainly helping. Most cores with only 1 thread can not completely utilize the execution units, especially when a thread is stalled when fetching something from memory.

That said, I'm getting to test a 6 core (12 virtual cpus) chip soon, so I'll post back if I notice a difference. I am not expecting much as my gfx card and hard drive are usually the bottlenecks for me.

Thanks for the info. There may be quite a difference between how LoTR and EQ2 act though. EQ2 performance relies almost entirely on CPU speed. The graphics card really doesn't matter. I'm not sure how LoTR handles it. But I would be very interested in hearing back from you after the 6 core is tested!

Thanks for the info. There may be quite a difference between how LoTR and EQ2 act though. EQ2 performance relies almost entirely on CPU speed. The graphics card really doesn't matter. I'm not sure how LoTR handles it. But I would be very interested in hearing back from you after the 6 core is tested!

Things seem smoother, but not in the sense that there is a night and day difference. It's more that I haven't noticed any big slow downs or stalls recently, other than when all the characters load in for the first time. Before, I'd notice an occasional stall or slow down. It may just be coincidental, and I haven't played on the new chip more than a few days. It could easily be that I've had better than average lag since I've had the new chip.

If I had purchased this chip with my own cash for the sole purpose of 6 boxing Lotro, I would not consider this a good value upgrade over my 4 core. I think this speaks more to the fact that 8 virtual cpu's is plenty to handle 6 instances of Lotro.