Speed Matters! Find out why 2400MHz high-speed memory should be your next upgrade with Corsair

Have to agree even the tech idiots should spot the dodgy graph scale. On here it will get ripped apart ;) I did wonder when they started with the in built graphics engine benefits. Not to say there aren't none gaming applications that would benefit but I assume most people here build systems primarily for gaming?
 
I refuse to believe there is 6fps difference between memory. If there genuinely was in all circumstances I would be buying 32gb of platinum for sure. Can some else test this with single monitors on battlefield 4?
 
bemaniac said:
I refuse to believe there is 6fps difference between memory. If there genuinely was in all circumstances I would be buying 32gb of platinum for sure. Can some else test this with single monitors on battlefield 4?
Click to expand...

Well 6 FPS is in my opinion within margin of error, you could probably replicate the results they got by running the tests multiple times until you got the result that fit best.

More importantly though even if this was true for all games, the money could be spent on extra GPU/CPU grunt which is far better.
 
Actually it's been shown before various reviews that faster rams speed DOES increase actual frame rate in games rather than error in margin of error...however the big catch is that it is ONLY if you had socket 1150, due their memory controller can actually benefit from the extra speed of the ram; whereas for lesser platform (socket 1155, AM3+) the chances are 2400MHz won't have any meaningful gain over 1600MHz.

Also we are talking about like may be 3-4fps difference at when frame rate at 80fps+, so at below 40fps, we'd probably be looking at like 1.5-2fps. Is that little extra frame rate really worth spending more on the high speed memory? It's up to the individuals to decide.

I wish OcUK would do a little more homework before unknowingly/carelessly misinforming their customers...
 
Last edited:
lolz at the graph from corsair...at first I didn't even look at the numbers just the graph.....and thought wow double the performance with super corsair ram.....I even started pricing up 32gig lolz....till I seen some comments lol
 
I'm sorry but if someone skews the axes like that on their graph to make the gains look bigger I'm instantly super-sceptical of the claim.

Add to that, when I bought my current GeIL I did research into RAM speeds and found there was little difference in real-world tests and I'm even more suspicious.

My next RAM upgrade will be DDR4 anyway. The set I have at the moment is easily fast enough to last.
 
I think GeiL or some other memory manufacturer came to similar results. Again I think they used BF4 to test. Maybe BF4 does benefit a little?

I was also tempted to make a graph, but do one that shows FPS from 0 - 500 and then mark the Corsair results on it. See how good it looks if you adjust the X-axis the other way...
 
Dicehunter said:
Agreed, That chart is total BS, Why even bother to lie to your loyal customers who are tech savvy ???
Here's a video from a non Corsair sponsored tester i.e Linus

yzrpDmg.jpg.png

B98FBS3.jpg.png
Click to expand...

IIRC he was testing with a relatively low GPU - if you throw in a high end GPU or multi GPU setup you will see bigger gains - albeit not huge differences but a bit more than his results show.

Also shame he missed out on 2133MHz stuff in those graphs - on current intel platforms nicely tuned 2133MHz stuff* actually produces the highest or joint highest fps in many games unless your running like Quad SLI or something that can gain from using silly 3GHz type RAM.

The reason I would buy 2400MHz stuff is due to it generally giving a bit more headroom to play with in regard to finding some decent timing/frequency combination - currently running my HyperX Beast 2400Mhz kit at 2133 9-10-10-21 with sub-timings tightened up to ~6/24 type settings.


* Not tested if 2400+Mhz with the same timings where its possible produces even more performance, but it seems to drop off once your increasing the CAS at higher frequencies.
 
Last edited:
Ken said:
The initial graph Corsair binned...

oMqmEmG.png
Click to expand...

Great to see the un-biased graph. This how it should have been in the beginning.

As expected, the gains are marginal at best considering the cost of 2400mhz kits. So price to performance ratio is negligible at best.
 
Well, this was a good giggle. The other option is save the money and buy a better graphics card. Or maybe a nice case. Or PSU. Or...or just not this. Samsung Green ram proudly doing its job here!
 
cypto said:
Great to see the un-biased graph. This how it should have been in the beginning.

As expected, the gains are marginal at best considering the cost of 2400mhz kits. So price to performance ratio is negligible at best.
Click to expand...
Yep. Can pretty much ignore the 32GB results as I suspect most gamers will be looking to buy 8GB or 16GB kits and just like Deceptor mentioned previously the 3.8fps difference between the 16GB results is probably the best result out of numerous runs.
 
Indeed I was always on the understanding anything over 1600mhz was marginal in terms of performance, so 1600mhz being at the sweet spot for price/performance.

Completely agree with Carrot, investing that money into a better GPU or going SLI is going to get you much better returns than 32gb of 2400mhz ram.
 
You must log in or register to reply here.
Back
Top Bottom