bf4 ram speed benchmarks?

[lhikary]

anyone knows whats up with that? corsair said it improves fps by around 15%
like did aynoe made a benchamrk beside corsair

At 1920x1200 with Ultra Settings, Corsair recorded minimum and average framerates using both DDR3-1600 and DDR3-2400 RAM. Here's how it broke down:

• DDR3-1600 (Minimum Framerates): 72
• DDR3-2400 (Minimum Framerates): 79
• DDR3-1600 (Average Framerates): 91.8
• DDR3-2400 (Average Framerates): 112.6

Corsair then repeated the tests at 5760x1200. Let's have a look:

 

• DDR3-1600 (Minimum Framerates): 35
• DDR3-2400 (Minimum Framerates): 43
• DDR3-1600 (Average Framerates): 46.8
• DDR3-2400 (Average Framerates): 53.9

[jscho]

Battlefield 3 had a fair bit of improvement with higher speed memory too.

[Hecknologist]

Do we know what the cas latency on the ram tested is, as I would assume if the 2400 was CL11 and 1600 was CL9, that there wouldn't be a massive difference

[lhikary]

Do we know what the cas latency on the ram tested is, as I would assume if the 2400 was CL11 and 1600 was CL9, that there wouldn't be a massive difference

explain more please

[SteepGnomeKing]

explain more please

 

As you increase the speed (mhz) of ram you increase the cas latency (Ex. cl9 on 1800mhz) In some situations these 2 things can have no effect. Because while you are increasing the response you are increasing the latency and they cancel each other out in some situations.

No idea if it is this way in games because ram speed hasn't ever made a difference in ddr3.

 

But still, you don't believe AMD for there benchmarks anymore than you would Nvidia. So why do we trust Corsair. Sure George said he seen it himself and can't believe it but he is a spokesman for the company and needs to promote it. He wouldn't say "It barley ever happened but in this one situation we took the lowest number from the lower ram and put it up against the highest number of the higher ram" 'Please buy our more expensive ram'

[Hecknologist]

explain more please

I'm not a expert but someone described it as "Imagine RAM as a tunnel, the clock is the speed that things can go through the tunnel and the timing(CL) is how wide the tunnel is." or close enough, meaning that having really fast ram with really bad timing has similar performance as slow ram with great timings. Of course take this with a grain of salt as it probably differs in different situations/circumstances, but I want to know what the latency of the RAM in the test was to test this in BF4 

[Kilerbomb]

Do we know what the cas latency on the ram tested is, as I would assume if the 2400 was CL11 and 1600 was CL9, that there wouldn't be a massive difference

It would probably be an accurate assumption to say that they are using Corsair RAM. I could be wrong, though.