So what's your excuse for being a ramlet? Time to get dat

File: 1.jpg (268KB, 1150x645px) Image search: [Google]

268KB, 1150x645px

>ram speed is just a meme!

is this a thread where we argue to death about margin of error 2 fps?

I'm not going to see a difference with my i3 so who cares? Not everyone is a retard with more money than brains.

>>60018868
>DDR4

Meme.

Guys! Ryzen NEEDS the fastest ram possible guys, I swear! Only to find out there is a 2-3 fps difference. Yea, nah. I'll stick to my 32GB of low latency 2400mhz thank you very much. If you have fast ram, great. But it's not the end all be all everyone swears it to be. This was only made worse when Ryzen hit the scene. Now people don't shut the fuck up about it.

>>60019018
>niggah doesn't know about true latency

>>60018966
Yeah and the pattern of the margin of error just happens to be an upward trend among all tested cases

>>60019018
Isn't 3200mhz memory like 10-20 dollars more expensive? Like why wouldn't you pay 20 bucks for 2-3 fps anyways? Also it's not 2-3 fps, it's 1% to 15% in some games. That can mean going from 54 fps to 60 fps just by paying 20 extra bucks which is less than 2 Pizzas, like seriously kys.

>tfw bought 3200 mhz ram because the price difference was marginal over slower ram
I made the mistake of buying 1333 mhz in 2012, I wasn't going to make that mistake again.

at 2933 with my 1700. fuck u lisa

>>60019072
you may have a point if thsoe tests were done properly, several runs, different parts of games, longer than 30sec runs
then it would have true AVERAGE performance this is just showcase of one run

>>60019343
I think you're retarded

>>60019343
Ofcourse it isn't a showcase of one run.

Any respectable reviewer will take an average of at least 3-5 runs

MUH SINGLE DIGIT SUPERIOR AVERAGES / MINS

>>60019403
Increasing minimum framerates is actualy one of the best and most noticable things you can do for your gaming experience.

You won't notice a difference in 100 and 110 but you will definately notice a difference between 30 and 40 fps.

>>60019396
>Any respectable reviewer will take an average of at least 3-5 runs
haha.

>>60019438
Even if it was one run (it isn't), what do you think the chances are of this upward trent on 7 x 5 = 35 configurations showing up just by accident?

You have to be an absolute tinfoilhat tier mongoloid to believe this is just margin of error stuff.

>>60019429
>1080p
>30 and 40 fps
STOP TALKING ABOUT 4K

>>60019509
Since 85% of gamers are running shit tier low end gpu's this also applies to 1080p

>>60019538
STOP TALKING ABOUT SHIT TIER LOW END GPUS

>>60019553
No

>>60019583
YES

>>60018868
>tfw bought 3200mhz ram that only runs at 2133mhz

>>60019597
shit tier low end gpu's

>>60019626
Check your bios XMP profile

I bought EVGA DDR4 that says 3000Mhz, I used xmp profile in the bios and now it's 2933Mhz

>>60018868
What motherboards support this anyways? Everytime I read reviews about what motherboards to buy, it seems like none of them can actually reach these speeds. I'm pretty sure Asus is right out. Is it Gigabyte? MSI? Asrock?

>>60019494
framerate tools make errors, it's common knowledge because they are freaking hard to make, when difference is under 5% on short runs you need to do longer runs to confirm consistency
difference maybe even bigger for all I care

I do not trust a single reviewer to do that, specially linus.

>>60018868
>>60018896

Looks like a gpu bottleneck
Link?

>>60019706
ASRock, Gigabyte, Asus.
all get stable 2933 at least, IMC apparently can do 3500+ or at least that's what AMD says, if you have time hold of on RAM
avoid MSI they barely got their shit together only recently

>>60019706
asus: prime/C6H; gigabyte: gaming 5/k7, asrock taichi/fatality should all hit 3200MHz with supported 2x8GB kits (samsung B die, single rank). As long as you use a newer AGESA 1004a based BIOS. However at this point higher frequencies than that will only work if you change the BCLK.

>>60018868
>every other benchmarking website: "Don't worry guys, your old Intel quad-core still runs this game just as well as Kaby Lake or Ryzen, and RAM doesn't make much difference anywhere."
>Linus Shill Tips: "1800X outperforms 7700K by 20%! Faster RAM makes all the difference! Buy buy buy!
Really makes you think...

>>60019745
>results consistently show an upward trend in performance across several results
>the lowest is ~10fps below the highest in avgs, ~15 in 97th percentiles
>m-margin of error!
kek

>>60019018
32gb of 2400 is way more expensive than 3200 of 16gb RAM. So you overpaid for an actual loss of performance, since 32gb is not needed for anything.

Cuck.

>>60018868
If you're getting almost 200 fps in game I don't see the point for a 10-20 fps boost

File: 1488165794180.jpg (49KB, 560x577px) Image search: [Google]

49KB, 560x577px

>>60018896

Ryshit need 3200Mhz RAM to reach intel performance with 2133Mhz RAM

File: Spec.jpg (55KB, 538x393px) Image search: [Google]

55KB, 538x393px

>>60018868
Because I can't afford modern hardware.

>>60019745
You're avoiding the argument

>>60020035
>all other benchmarking website
Really? Lets see those numbers

>>60020711
>2133Mhz Mins on For Honor are still higher than 3600Mhz on 7700K
KEK

>>60018868
>Ram speed matters when you run games @ 1080p with a 1080 ti
so, for the overwhelming majority of people, ram speed does not matter.