Jump to content

virtuali

Does RAM speed matters ? A quick test

Recommended Posts

Posted (edited)

Since I just upgraded by system RAM from 32GB to 64GB, and went from DDR4-2133 to DDR4-3000 (that's the fastest my MB can use), I took the time to do a small test before switching memory, which might be useful to some.

The system:

CPU i7-6700 at stock clock speed, GPU nVidia 2080 Ti, old ram was 32GB of Kingston HyperX Fury DDR4-2133, new RAM is 64GB of Kingston HyperX Predator DDR4-3000

Situation:

FSDT KORD V2, sitting on Runway 10C, with DD Chicago City X and KMDW airport installed. Ultimate Traffic Live. Settings are were of course unchanged between sessions, and I used a saved flight to be sure nothing changed between the two.

These are the results:

DDR4 2133 32GB RAM

Default 737 External - 100% AI - Avg 26.4
Default 737 VC - 100% AI - Avg 24.0
PMDG 747 External - 100% AI - Avg 22.7
PMDG 747 VC - 100% AI - Avg 14.9

DDR4 3000 64GB RAM

Default 737 External - 100% AI - Avg 33.0
Default 737 VC - 100% AI - Avg 22.7
PMDG 747 External - 100% AI - Avg 25.4
PMDG 747 VC - 100% AI - Avg 16.3

So yes, even on a fairly "weak" CPU, not overclocked, there IS some benefit upgrading with a faster RAM. Of course, I was more interested in having 64GB, to be able to keep many programs opened at the same time ( P3D4, 3DS Max, Substance Painter, Visual Studio, etc. ) but, I thought it could be worth using the fastest RAM I could use without changing CPU and Mainboard.

I'm sure those with a faster CPU might be able to see even more difference.

Edited by virtuali

Share this post


Link to post
Share on other sites

UPDATE

Checking with CPU-Z, I noticed that, by default, my new DDR4-3000 RAM chips were automatically configured to run at default specs of 1.2V and 1200 Mhz speed, as if they were DDR4-2400. This seems to be a default conservative settings when the BIOS is set to Automatic SPD detection.

After setting the "XMP" profile, the clock speed went up to 1500 mhz and CAS Latency went down to 15, from 17. And with the same test, some extra performance was obtained.

DDR4 3000 64GB RAM - XMP 3000 BIOS Profile

KORD 10C - Default 737 External - 100% AI - Avg 34.4 ( from 33.0 )
KORD 10C - Default 737 VC - 100% AI - Avg 23.9 ( from 22.7 )
KORD 10C - PMDG 747 External - 100% AI - Avg 26.2  ( from 25.4 )
KORD 10C - PMDG 747 VC - 100% AI - Avg 17.9 ( from 16.3 )
 

So, if you have fast memory, be sure you check your BIOS setting, and apply the XMP profile, if available.

Share this post


Link to post
Share on other sites

Interesting data Umberto. Thanks for posting these results of your tests. Would you mind posting what the timings were on your 2133 and 3000 RAM. I have found that these also make a difference.

Thanks,

Ted

Share this post


Link to post
Share on other sites

Good test. The only thing left that a reader has to substantiate is if 3-7 FPS gain is worth XXX dollars. In my experience, no not worth it unless you are running sub 30's on the FPS. RAM prices get crazy when you get into the higher bandwidth RAM. You didn't get extreme benchmarking handpicked RAM that often cost hundreds more than stock speeds, those are not worth the cost IMO for an additional 1 or 2 FPS than what you achieved.

Share this post


Link to post
Share on other sites

Hi Umberto,

Thanks for the info ... agree, XMP profiles will improve overall performance as they should.  Some motherboards will offer multiple XMP profiles (one per RAM and one optimized to MB).  However there is a point of diminishing returns and the performance benefits are less.  I use SiSoftware sandra to help me determine raw performance of RAM.

I have no issues with performance at KORD even with extreme tweaked settings ... I enjoy the airport so much I just taxi around pissing off the AI.

Cheers, Rob.

Share this post


Link to post
Share on other sites
11 hours ago, Rob_Ainscough said:

I enjoy the airport so much I just taxi around pissing off the AI.

LMAO.... I thought I was the only one who did that...

Share this post


Link to post
Share on other sites
5 hours ago, TurboKen said:

I thought I was the only one who did that...

Ironically, that's what I do real world when I want to get inside airport facility to take pictures.  I schedule an "exploratory" flight with an instructor and he gets permission to taxi the airport while I frantically take pictures/video for use in my own virtual airport creation.  However, this is for small GA airports as that would not "fly" so to speak with larger airports.

Cheers, Rob.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...