[TBG] Does RAM Speed Matter? DDR3-1600 vs. 1866, 2133, and 2400 in Games

[wand3r3r]

The Tech Buyers Guru has a new RAM article, focusing on the speed this time.

The impact of system memory on gaming performance is a topic lots of gamers are interested in, but finding in-depth analysis can be hard. Back in mid-2013, we published our analysis of how RAM quantity affects gaming performance in our article Does Memory Matter? 4GB, 8GB, and 16GB in Gaming. That article has gone on to be one of the most popular articles on The Tech Buyer's Guru, to this day! So it’s with great pleasure that we bring you the next chapter in our RAM benchmarking analysis, a deep dive into the impact of RAM speed on gaming performance.

http://techbuyersguru.com/ramspeedgaming.php

Take a look, it's an interesting read. It's always interesting to see the real world effect that memory has on (primarily gaming) performance. The conclusion offers solid advice and alternatives for potential buyers and it's great to see the value proposition for memory! It can make the purchase easier based on price/performance, or simply all out performance.

 

[denis280]

it depend what cpu is in use.

 

[evident]

meh, i find 16 gb of ram nice when i have chrome open with 10000 tabs, along with a VM running. I'm not sure if speed makes any difference.

 

[evilspoons]

Guys, read the post. The OP is not asking a question, they're posting a link to an article for you to read!

The article itself has interesting information, although the person who wrote the article could use some education on how to present data in graphs. All the similar results should be posted together (e.g. every physics score for 3Dmark should be grouped together and then sorted by winning RAM type, rather than being grouped by RAM type and then you having to squint at which one actually won. As the graphs are presented, they're not actually any more useful than just a list of numbers!)

The article itself really boils down to "it doesn't matter for most people" though, and at most, you're gonna want to go one or two levels above the regular stuff instead of going for the top-of-the-line.

 

[KingFatty]

I thought it was disappointing to see how they did the graph on "total system cost" and determined that it made no difference for total system cost which RAM you purchased. That seemed silly to me because the lowest price choice results in money in your pocket, but the highest priced choice costs extra money, but their graph treats them all the same. Seems like a useless graph for this comparison.

 

[Cerb]

meh, i find 16 gb of ram nice when i have chrome open with 10000 tabs, along with a VM running. I'm not sure if speed makes any difference.

Makes tons of difference. That's also what makes 4C8T worth the cost over 4C4T. A perfectly clean system, with nothing in the background largely doesn't exist, outside of performance review website systems. Whether either are worth adding another $70+, especially if that is taking away f$70+ from some other components, well...

 

[Essence_of_War]

I was REALLLLY hoping there'd be further pages exploring igpu with memory speeds! If you're reading this TBG, and you're not already planning that as a follow-up, I think that'd be a pretty good idea.

 

[Termie]

I was REALLLLY hoping there'd be further pages exploring igpu with memory speeds! If you're reading this TBG, and you're not already planning that as a follow-up, I think that'd be a pretty good idea.

This is mostly of interest to AMD APU users. Is that what you're referring to, or are you interested in Intel HD Graphics in particular?

 

[Essence_of_War]

I would be interested in both! I know that's twice as much work, but I think that could be a neat comprehensive piece to point people to on what is and what is not affected by memory speed.

The figures on the cost-per-fps made me immediately think of an APU/igpu analysis could be useful. We occasionally see people here considering the relative cost of an APU vs. a cpu + entry level dgpu, and although it usually breaks to the latter, knowing what the limits (or at least the price/perf inflection points) of faster RAM for APUs would be useful.

 

[master_shake_]

i've read that AMD chips love fast ram, intel...

marginal at best.

 

[Dasa2]

in my testing of ram speed i have found that the higher the cpu oc the more starved for memory bandwidth it becomes but as the cpu gets faster the number of cpu limited games lessens so gains from faster ram also drops

in a cpu limited game like arma though you can see 15% incrase in fps going from 1600c11 to 1600 c8 and from 1600 c9 to 2133 c9
which is equivalent to a fair oc on the cpu

http://forums.bistudio.com/showthre...rmance-comparison-1600-2133-up-to-15-FPS-gain

http://forums.atomicmpc.com.au/index.php/topic/55771-cpuram-performance-in-thief/

i think reviewers need to compare ram speed vs cpu speed so people can see that faster ram wont help if a faster cpu isnt making any difference
why spend $400 on a 4790k and skimp to save $10 on 1600c11 ram when a 2500k with faster ram will perform about the same in games

 

[Cerb]

i've read that AMD chips love fast ram, intel...

marginal at best.

It's more the AMD users loving to benchmark file decompression and video encoding, and review sites loving to use non-zeroed graphs in memory comparisons, more than huge benefits from faster RAM. AMDs can use a bit more than Intels can, but outside of such tasks, the differences are very small, and among those last little things to spend on, when you have a $1.5k+ budget (basically, when you cannot reasonably get a faster CPU speed bin, better CPU cooler, or a faster GPU). Unless, of course, there's a good deal going on, like the G.Skill right this second, that's on quite a few Haswell mobo QVLs, despite being 1.65V rated, for <=$70 .

P.S. Ninja'd with data.