X3:Terran Conflict

Publisher: EgoSoft

X3: Terran Conflict is the latest space-based trading and fighting game in the X3 series. It uses masses of polygons and geometry to create the complex ships and so is a very stressful test for a CPU.

Remember that it's the CPU that constructs the wire-frame models of game objects, before a GPU can apply texture and lighting.

To run the benchmark, first download the rolling demo and set the graphics options as below. Then remember to tick the 'Run as benchmark' box and hit go. We average the results of the four component tests to give a single minimum and average frame rate.




The same happens here too: Core i7 has a higher minimum than Lynnfield and both are notably faster than the older Core 2 Quad by some margin. Specifically there are a few frames per second in favour of Core i7 in all cases, but it's not a huge difference by any means.

Performance Conclusions

If you're an Intel loving benchmarketeer then clearly an LGA1366 Core i7 and X58 is still the platform to go for because it ekes out every last frame per second of performance, and benefits from the extra channel of memory bandwidth.

Having the PCI-Express on CPU for Lynnfield doesn't offer a notable benefit. In fact, after further discussion and BIOS dissection it appears the CPU to PCI-Express connectivity via an internal QPI link is almost exactly the same as if it were in two physically difference places - the data still has to travel from CPU to Uncore to QPI to PCI-Express, which is the same path it would have with X58. We think the internal QPI link width is less than X58 (Intel has yet to confirm this), but because only one x16 link rather than 36 plus a southbridge on a Nehalem platform need to be catered for, it's not a bottleneck.

If we're talking about just the gamer though, a few frames per second won't make a huge difference so for the most part an LGA1156 Lynnfield CPU keeps up with the LGA1366 Core i7 and provides a very sound gaming platform. Lynnfield is notably faster than the older P45 platform, even if you own a recent 45nm CPU with larger L2 cache. The front side bus and DDR2 are limiting factors that force frame rate drops on a more regular basis, especially when it comes to multi-GPU where it regularly fell tens of per cent behind.

When we get to multi-GPU scenarios, there is a surprisingly clear difference in favour of the Core i7 system, whose two full x16 PCI-Express links consistently provide a far higher minimum frame rate in almost every case. This is interesting because it actually highlights a subtle difference between PCI-Express 2.0 x8 and x16 that we had previously thought wouldn't be an issue - clearly the available bandwidth in very intense scenarios (where minimums occur) begins to run out on an x8 lane.

It's hardly affecting gameplay, but with PCI-Express 3.0 spec a year or two off at least it could become a system bottleneck. With DirectX 11 bringing more complex game environments and extra calculation data such as GPU accelerated physics and AI, the PCI-Express bus could become a system bottleneck by the end of PCI-Express 2.0's life. This will be strange if it comes to light, as it's an area of the system where we've traditionally always had oodles of bandwidth spare.