Just bought a new GTX 1080 - Is my RAM holding me back?

[matthewjacko]

Hey guys,

So I have upgraded my GPU recently to an ASUS ROG Strix 1080 (not the OC one).

I have kept my old CPU - an i7 4770k @ 4.2ghz which does not seem to be bottlenecking the card whatsoever.

I am also running my old ram.. which is 16gb of Corsair Low Profile RAM overclocked to 1600mhz.. Is it possible that my RAM is holding my new 1080 back?

Thanks

 

[RobCrezz]

No I doubt it very much.

 

[matthewjacko]

No I doubt it very much.

Okay thank you Rob

 

[JackNaylorPE]

Is it holding you back ? I wouldn't say so.

Would the machine be faster with faster RAM ? Definitely, tho that would depend on which component is the bottleneck. In gaming for example, THG showed that certain games, faster RAM had no impact on performance ... in that game (Metro 2033) the GPU was the bottleneck. In other games, (i.e. F1), the increase in going from 1600 to 2400 was 11%.

In other w=uses, like CAD, video editing, rendering, etc ... RAM has a significant affect on performance.

 

[lakimens]

In gaming, it's a very small probability, if you're doing some work, Premiere, CAD, 3DSMax... Then yes, those applications benefit from faster RAM.

 

[CRO5513Y]

In gaming, it's a very small probability, if you're doing some work, Premiere, CAD, 3DSMax... Then yes, those applications benefit from faster RAM.

RAM Speeds do actually make more of a difference in some games then you would expect. There is multiple Digital Foundry videos on it. Provided 2 screenshots for you below for example, note not just the FPS at the current time but the average is higher as well. However i wouldn't suggest someone to buy faster RAM specifically if they already have say a 1600 MHz kit unless they are buying a whole new kit for other reasons.

 

[lakimens]

Wow, man that is a seriously bottlenecked build, i3/FX paired with a Titan X? Don't you think that might be the cause? I mean sure, it probably does play a part in games, but I wouldn't trust that bottlenecked system as a benchmark.

 

[CRO5513Y]

Wow, man that is a seriously bottlenecked build, i3/FX paired with a Titan X? Don't you think that might be the cause? I mean sure, it probably does play a part in games, but I wouldn't trust that bottlenecked system as a benchmark.

That's the idea... Putting a Titan X means the bottleneck would be the CPU so the test can push the CPU to the limits. No point using something like a GTX 1060 for example if it will be doing as much work as the CPU and potentially give invalid results. This is probably going off topic anyway... Just wanted to point out RAM speeds can indeed make some difference in games not necessarily as big as a better GPU or CPU of course but it is something. OP doesn't need a new kit tho 1600 MHz DDR3 would suffice.

 

[matthewjacko]

Thanks for all the input - seems like I am O.K with my 1600mhz kit. Thanks all

 

[JackNaylorPE]

DDR3-1600 = 159
DDR3-2400 = 177

177 / 159 makes 2400 11.3% faster than 1600

Again, I wouldn't say it's holding you back but the notion that RAM speed has no impact on gaming is simply false. Some games are GPU bound, some are CPU bound and some are RAM bound.

Another problem with most reviews is that the only test for average frame rates with a single card

a) When RAM gas an impact, it shows up more in minimum frame rates than average.
b) Adding a 2nd card will often eliminate the GPU bottleneck in which case CPU and RAM performance will assume that role.

Again, not suggestion you toss ya 1600 set and run out and buy 2400 ... but next build, I would it is something that should be taken into consideration