Weird advertisement post... I had my overclocking ram in for that one. It was pretty stable, too. Managed a Firestrike run. http://www.3dmark.com/3dm/1472198? Massive physics score is massive.... Sadly after about twenty minutes in the CPU got too hot and became unstable. Bah ! need a dedicated water loop There is a guy on OCUK called Alex who has his at 5.2 all day stable on a water loop. It's not even a massively expensive one either. Edit. Wow, actually that run was at a mere 5ghz with my 1600 mhz ram. Meh. I want access to my PC dammit !!
Andy what fans are you using on the h100 you use? I have a custom loop with a 120 rad and at 5.2 and temps were at 55c after a run of 3dmark and then a run of cinebench.
I'm using Cougar Vortex mate. The best, basically. Huge SP and very quiet, plus rather nice looking too. That's it. No pull, just an intake push. They're really expensive in the UK which is a shame. I found those two on Ebay for £20 inc, half of what you pay through importers
Those are arguably the best fans for watercooling. I am not using only for the fact they are orange but performance they are sweet as anything.
Mate if I could save money by going AMD and not seeing a performance drop then I would. But when I bought my system, that's what gaming benchmarks suggested.
Your physics score is hardly massive considering you have 4 more threads than me... http://www.3dmark.com/3dm/153487 And that was at a mere 4.7ghz
Benji having a benchmark system and a gaming system are 2 different things. In the real world you would not see any difference
Ignore what the benchmarks say. Seriously, it's all bloody pish. It's what happens in the real world that matters. The yanks seem to be huge fans of the AMD FX CPUs and over at Linus' loads of people are using them and are thrilled with them. The 6300 is just epic for the price. It will do 5ghz usually all day, worst I have seen was 4.6ghz which is still a hefty overclock. I very, very rarely monitor my FPS in a game unless it's exhibiting problems. Take Metro : Last Light as an example. It was running like a slug on certain levels. Turned on the FPS monitor "That ain't right !" then found a patch that sorted it out. I usually just fire up a game and play it.. I never looked at my BF3 FPS before. All I knew was on my GTX 470 it was bad at max settings. Turned out to be not enough vram causing input lag via texture streaming. Replaced with 6970 problem gone. On the subject of BF3... Here I am playing it at 1440x900 absolutely maxed out on my rig. It seems it has a 200 FPS limit. According to 'law' though I should never have found that limit on AMD. Funny thing was when I posted that the Intel lot were saying "But that's not multiplayer !" No it isn't, but then the video I posted earlier covers that one too.
It's still higher. I'm also running SLI which reduces the physics score. See if you can match nearly 10k. I've already been up against a 2550K running 4.8ghz and won. Thing is, what will be forgotten here is that the 2500K was never £110 brand new, was it? it was more like £170. 4670K delidded @ 4.8ghz Close but no cigar. For a laugh I will run one GPU, I know I can break 10k. So we're back to price..
A benchmark from a game is designed to simulate playing a game is it not? If it isn't, you lot all need to read a dictionary and do your bloody jobs properly. You know, seeing as these benchmarks are taken in-game.
Gaming benchmarks usually always do not depict actual gameplay. What I mean is they're usually really hard nosed and torturous. However, you're dead right that they can be used to compare hardware. At the end of the day a benchmark is always the same, absolutely every time. There is a tiny margin of error, but this can be easily sorted by simply running 3 passes. By that time the system has loaded everything it needs into memory and video memory.
What is the purpose of this spanner? I'm very confused. Their flagship product is an i7 powered SLI Titan machine...
Benji A bench mark even from a game is designed to test the system, where as the games is deigned to work as evenly as possible across different spec systems. So a games does not work like a benchmark and never will. the 2 pieces of software have very different functions that is why benchmarks do not necessarily show you everyday performance. Hence you can build a benchmark machine or a real world gaming machine
You yourself have said that very few games have built in benchmarks, so how do they make it torturous?
But the point is it shouldn't be close now should it? If your fx cpu is so mighty? I can't be arsed to revisit this test but it doesn't look like a massive physics score to me for a cpu with 4 more threads... We're on price not performance? We all agree AMD is good value. But it isn't fastest.
They usually run everything all at once. Metro's benchmark is really hard on a rig. Play the game though? you usually always net much higher FPS when playing the game itself.
It isn't the fastest no. The 3770k and 4770k will absolutely wipe the floor with it. Yesterday a guy delidded his 4770k and got 1000 points in Cinebench. The best I've seen from a 8320 is 830 or so. But then let's be real. The 3770k and 4770k were never £110 were they? they were £240. Which when you look at it (the 4770k is around 20% faster than the 8320 in Cinebench) it's pure crap value for money eh? 20% faster, 100% more expensive. As I keep saying there is no quad core Intel that can beat the 8 core 8320. You can look at it however you like, using certain analogies.. 1. The AMD is barely faster in benchmarks (but it is faster in every benchmark that supports it) 2. It has 8 cores so should do a lot better. ETC, ETC. Negative nay sayer mode on. Or, you can look at it this way. 1. The 8320 is £60 cheaper than the 4670k and it's faster. 2. The 8320 has Virtualization and can be overclocked with it. And so on. The bottom line though is price. AMD are winning hands down, by a long way.
