Overclock.net › Forums › Components › Memory › An Independent Study: Does The Speed Of RAM Directly Affect FPS During High CPU Overhead Scenarios?
New Posts  All Forums:Forum Nav:

An Independent Study: Does The Speed Of RAM Directly Affect FPS During High CPU Overhead Scenarios? - Page 24

post #231 of 277
Quote:
Originally Posted by XLifted View Post



In my case the max and average both boosted. As far as I could tell. I dont think it impacted minimum negatively, because perception of frame dips decreased altogether
Hmmmm. I was actually hoping that minimum fps would have increased. Thanks for confirming.
post #232 of 277
Quote:
Originally Posted by Scotty99 View Post

I am a complete noob with it comes to memory overclocking. I watched this video on if i should upgrade my 2500k that shows memory speed can make a difference (this wasnt the case when i built my rig in 2011):
https://www.youtube.com/watch?v=frNjT5R5XI4

I have a very good kit of ddr3 here:
http://www.newegg.com/Product/Product.aspx?Item=N82E16820231445

I would like to try and get 2133 out of this ram do you think its possible? What would be the timings and voltage i would need to enter into my bios, remember ive literally never OC'd ram lol.

Thanks.


http://www.newegg.com/Product/Product.aspx?Item=N82E16820231538&cm_re=8gb_ddr3-_-20-231-538-_-Product

Might as well get this set here rather than the one you linked. The one you linked is G.Skill Ripjaws 1600mhz CL8, whereas this is the exact same except 1866mhz CL8 with pretty similar timings etc.. so a bit better, and it's still only $46. It's also still only 1.5v so you could probably easily get to 2133mhz CL8-9 on it by raising it to 1.65v.

I imagine something like 2133mhz 8-10-10-26-1T would be possible at 1.65v on that set. (and changing to 2T would be almost guaranteed to work from my experience. Although nothing is 100% guaranteed when overclocking of course; so keep that in mind)


Or there's always this set: http://www.newegg.com/Product/Product.aspx?Item=N82E16820231476&cm_re=8gb_ddr3-_-20-231-476-_-Product

It's 2133mhz CL9 (although the timings aren't quite as good at 9-11-10-28 compared to 8-9-9-24 on the 1866mhz set i linked) so on this set you are guaranteed to hit 2133mhz, and could probably get the timings down a bit too, or maybe shoot for 2400mhz C9. (this set is also $45)
Edited by DarkIdeals - 7/24/16 at 10:16am
 
CPUMotherboardGraphicsRAM
i7 5960X ASUS Rampage V Extreme MSI GTX 1080 Gaming X 16GB Corsair Dominator Platinum 2666mhz DDR4 C15 
Hard DriveHard DriveOptical DriveCooling
PNY CS2211 MLC SATA III SSD WD Blue 500GB 7200rpm Writemaster DVD/CD +/- RW EK Supremacy EVO 
CoolingCoolingCoolingCooling
Mayhem's Pastel Ice White Coolant Coollaboratory Liquid Copper TIM XSPC EX480mm Radiator Black Ice GTX 360mm Radiator 
CoolingCoolingCoolingCooling
EK XTC 420mm Radiator Swiftech MCP655-B 12v Pump EK RES X3 250 Reservoir Bitspower/EK 3/8" x 1/2" Compression fitting 
OSMonitorMonitorKeyboard
Windows 8.1 ASUS ROG Swift PG278Q 27" 1440p 144hz G-Sync ASUS VG23AH 23.5" Passive 3d 1080p 60hz  Razer Blackwidow Chroma Tournament Edition 
PowerCaseMouseMouse Pad
EVGA Supernova G2 1000 Caselabs SMA8 -XXL Window, Ventilated sides/t... Razer Ouroboros Norman Rockwell collection series print Mousepad 
AudioAudio
Sennheiser HD700 300ohm Open Back Headphones SupremeFX Hi-Fi 5.25" Bay AMP/DAC  
  hide details  
Reply
 
CPUMotherboardGraphicsRAM
i7 5960X ASUS Rampage V Extreme MSI GTX 1080 Gaming X 16GB Corsair Dominator Platinum 2666mhz DDR4 C15 
Hard DriveHard DriveOptical DriveCooling
PNY CS2211 MLC SATA III SSD WD Blue 500GB 7200rpm Writemaster DVD/CD +/- RW EK Supremacy EVO 
CoolingCoolingCoolingCooling
Mayhem's Pastel Ice White Coolant Coollaboratory Liquid Copper TIM XSPC EX480mm Radiator Black Ice GTX 360mm Radiator 
CoolingCoolingCoolingCooling
EK XTC 420mm Radiator Swiftech MCP655-B 12v Pump EK RES X3 250 Reservoir Bitspower/EK 3/8" x 1/2" Compression fitting 
OSMonitorMonitorKeyboard
Windows 8.1 ASUS ROG Swift PG278Q 27" 1440p 144hz G-Sync ASUS VG23AH 23.5" Passive 3d 1080p 60hz  Razer Blackwidow Chroma Tournament Edition 
PowerCaseMouseMouse Pad
EVGA Supernova G2 1000 Caselabs SMA8 -XXL Window, Ventilated sides/t... Razer Ouroboros Norman Rockwell collection series print Mousepad 
AudioAudio
Sennheiser HD700 300ohm Open Back Headphones SupremeFX Hi-Fi 5.25" Bay AMP/DAC  
  hide details  
Reply
post #233 of 277
Quote:
Originally Posted by PowerK View Post

Ok. I've done some digging around and found an informative article here.
http://www.crucial.com/usa/en/memory-performance-speed-latency

True latency (ns) = clock cycle time (ns) x number of clock cycles (CL)

So, for DDR4-2400 CL10-12-12-28
1,200 MHz = 1,200,000,000 Hz
Clock cycle time (ns) = 1 / 1,200,000,000 * 10^9 = 0.83 ns
True latency (ns) = 0.833 (ns) * 10 (CL) = 8.33 ns

For DDR4-3200 CL16-18-18-36
1,600 MHz = 1,600,000,000 Hz
Clock cycle time (ns) = 1 / 1,600,000,000 * 10^9 = 0.625 ns
True latency (ns) = 0.625 (ns) * 16 (CL) = 10 ns

So... I think I'll opt for DDR4-2400 CL10-12-12-28 because I think it can clock to 3200 MHz with 16-18-18-36 with ease while DDR4-3200 CL16-18-18-36 may not clock at 2400 MHz with CL10-12-12-28 timing. Besides, CPU + motherboard may not reach DDR4-3200 MHz.
CPU and motherboard I'll be using are 6950X and ASUS Rampage V Edition 10.

DDR4-2400 CL10 = 8.333 ns
DDR4-3600 CL16 = 10 ns
DDR4-3600 CL15 = 9.375 ns
DDR4-3200 CL14 = 8.75 ns
To have a better latency than DDR4-2400 CL10, DDR4-3200 needs to be CL13 (8.125 ns)
.
Did it work out to get the DDR4-2400 CL10-12-12-28 to clock to 3200 MHz with 16-18-18-36 with your CPU+MoBo?

BTW, wicked CPU+MoBo you got yourself there.

EDIT : You might be interested in reading this MoBo review where they've discovered fast degradation issues (related to overclocking) with i7-6950X CPUs : http://www.hardocp.com/article/2016/07/27/msi_x99a_xpower_gaming_titanium_lga2011v3_review/7#.V6CgIOt95hE

In the "Discussion" link at the bottom, the reviewers discuss it further at the forum.
Edited by Dokter Bibber - 8/2/16 at 6:34am
post #234 of 277
I have ddr4 3200 cas 15

I'm wondering how much true latency affects game FPS in ddr4
post #235 of 277
It's almost always better with higher speed.
post #236 of 277
My 6700k@4.7 was better at handling physics than my 980ti but now with a 1080 they are about the same where I notice no difference in 3D Mark so RAM matters even less and I have gone to 3600 T1 to test this.
Edited by MaFi0s0 - 9/30/16 at 9:22am
post #237 of 277
Well the pagefile on the HDD is slower than RAM so having less than 2GB and low frequency is going to kill FPS.

That's why builders recommend 8GB if you plan on gaming. To simply boil it down RAM lets the CPU calculate more instructions and values. Making tasks faster and gaming better to the person using it. But also motherboards are a culprit. If the RAM isn't listed with the motherboards compatibility you may run into problems down the line if it isn't supported.

Ah but I believe newer operating systems contribute in FPS drop as Windows 10 requires more RAM to achieve average performance. That's why I still prefer playing on Windows 7 it uses less RAM and runs smoothly. Mind me I still use Windows 10 on a different PC, because I test beta builds! Vista is a memory hog!

SDDs are designed to make gaming easier, because of the fast loading times/boot times playing games are going to be a breeze. It is faster to read and write than a mechanical hard drive. Higher RPM in mechanical hard drives do have a performance boost, but that will never outperform an SSD...

But there's another theory to talk about. Does video memory affect FPS? Those shiny looking GTX 1080's with 8GB of VRAM. I wonder if SLI affects FPS too?

Also higher frequency RAM takes less time than lower frequencies. The higher she goes the faster I go!

Only time and testing will tell. I'm still learning, but if i'm wrong I might facepalm myself for all eternity... teaching.gif
Edited by FlyinEagle - 10/1/16 at 5:21am
post #238 of 277
Great and informative test you've made there, thanks!
post #239 of 277
Quote:
Originally Posted by MaFi0s0 View Post

My 6700k@4.7 was better at handling physics than my 980ti but now with a 1080 they are about the same where I notice no difference in 3D Mark so RAM matters even less and I have gone to 3600 T1 to test this.

Have you tried different ram timings?

Lower true latency might help
post #240 of 277
firestrike i see around 0.3% increase in physics from 3200c13 to 3866c16 but timespy sees 3.2% while games like arma\fallout see a fair bit more still
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Memory
  • An Independent Study: Does The Speed Of RAM Directly Affect FPS During High CPU Overhead Scenarios?
Overclock.net › Forums › Components › Memory › An Independent Study: Does The Speed Of RAM Directly Affect FPS During High CPU Overhead Scenarios?