Why?Pentium Extreme Edition Redux? Broadwell or bust at this point, spending that much on a Haswell at this point is folly.
That's just silly. How long do you think you'll have to wait until you can get an Enthusiast-class Broadwell chip?Pentium Extreme Edition Redux? Broadwell or bust at this point, spending that much on a Haswell at this point is folly.
Pentium Extreme Edition Redux? Broadwell or bust at this point, spending that much on a Haswell at this point is folly.
That's been the price of their Extreme Edition proc on the enthusiast platform since this division between sockets started.$999?
Lol. Intel is never getting my money. Corrupt company.
And this is Crysis 3, which can use 4 or more threads on a CPU efficiently.
![]()
I can't wait to buy one in 2-3 years![]()
Well, take a look at the 8350 vs 2500K. The 8350 gets absolutely trounced in anything that uses 3 or fewer threads by the 2500K. Heck, the 8350 gets trounced by the 920/750 in anything that uses 3 or fewer threads.Looks like two threads to me. The G3258 is a dual-core CPU.
That's not why they're corrupt. Check the ruling of the European Union competition authority.That's been the price of their Extreme Edition proc on the enthusiast platform since this division between sockets started.
Bummer that you'd avoid all purchases from a company based on that one thing![]()
Unless there is a demo mode (Starcraft), it's really really really hard to benchmark RTS games due to the fact that there's a lot of factors outside the control of the person running the benchmark.Most benchmarks and recommendations are based on FPS/Action games, but I am much more interested in next-gen RTS games a la the StarSwarm demo.
It seems like there is almost no data on these types of games. I would assume complex simulations and RTS games would be more CPU-bound though, no?
I would happily spend 1K on a CPU if I thought it would help me play RTS with 5K+ distinct units on the screen like in the StarSwarm demo.
Like what?Unless there is a demo mode (Starcraft), it's really really really hard to benchmark RTS games due to the fact that there's a lot of factors outside the control of the person running the benchmark.
Like what?
It shouldnt be too hard to draw up two similar situations. Benchmarks should never be taken with 100% certainty anyways, so any minor differences in variables shouldnt detract from an overall 'trend' in performance between different configurations.
That's not why they're corrupt. Check the ruling of the European Union competition authority.
But the point is, with Haswell-E you get the best currently available IPC, high clock rates and a lot of cores.
If there is one thing i have learned of Pc gaming is that the word "overkill "does not exist in its vocaluarby.
Well that's what I'm talking about here - having both more cores, plus more powerful individual cores.
I'm a big user of vsync and I like a nice, smooth and capped 60fps. To achieve this, I sometimes have to actually achieve an average of 70-75fps, as some games can vary and dip considerably. So I might actually need to be able to hit 80-90fps just to lock it to 60fps.
I'm also considering a 1440p/120hz monitor in the next year or two.
And I'm buying the Oculus Rift CV1 as soon as its available, which will be 90hz at the least, and 1440p resolution.
If a 6 or 8 core CPU is going to help me out in these situations, then I wouldn't call it overkill and I wouldn't say its 'not for gamers' either.
Well resolution matters because it still ultimately eats into my performance potential.The resolution won't make a difference. But the 90Hz requirement may be something for which you could use a beefy CPU like this.
Bottlenecks don't work quite like that. They can exist on multiple parts at the same time, with different things in the game eating into the bottom line for various reasons. It's really engine dependent for the most part.Well resolution matters because it still ultimately eats into my performance potential.
If I become GPU limited in a certain game, its better to just be GPU limited than be both GPU and CPU limited.
I probably worded that badly. I meant like in a hypothetical situation where I know I can get 90fps with my fancy CPU at 1080p, but going to 1440p might drop me to 60-70fps or whatever. But at least I know that 90fps is possible and I'm not CPU limited.Bottlenecks don't work quite like that. They can exist on multiple parts at the same time, with different things in the game eating into the bottom line for various reasons. It's really engine dependent for the most part.
Yea, the price is what's making it really tempting.All that being said, if the 5820K kind of setup isn't that much more out of reach than a 4790K, then I will be the last one to disagree with that kind of purchase. Especially since I'll be doing it myself.![]()
Skylake and Broadwell in the same year?
lol best paradox ever.
Shouldn't be an issue with SC2 arcade modding I dontt think, any bloke can whip up a low-startup consistent high-actor map.Most RTS games require a build up to get to a certain point.
Play the same map 100 times, and due to AI, there will be a huge number of different outcomes.
I'd imagine you'd also want to test it when there's some fairly intense action going on in the game to get a real idea of performance, and again, to make sure you get to a fairly similar spot in the game 15+ minutes in means the typical 3-6 repeat per "run" could take 2+ hours. By the time that you run the different systems, you're looking at a ridiculous time investment with no guarantee of any consistency.