F5F Stay Refreshed Hardware Desktop Ryzen RAM speeds a myth?

Ryzen RAM speeds a myth?

Ryzen RAM speeds a myth?

Pages (3): 1 2 3 Next
J
JonahLomu27
Junior Member
27
05-01-2016, 12:38 PM
#1
I thought about it for a while too — mainly because I noticed no noticeable gap between the two configurations (only my wallet made a difference). I tried it with R3600 and GTX 1060, and now tested again with Ryzen 2200G, GTX 1060 plus 16GB Corsair Venegance "3200". Superposition scores show 16GB at 2133MHz vs 6334 points at 2933MHz for the stock. Based on these results, it seems buying faster RAM won’t really help for gaming. If you disagree, please clarify. I’m not including GN here since he often adjusts RAM specs and isn’t relevant to this point. This isn’t about overclocking; it’s about what you actually get when you spend more on speed. And yes, I might be mistaken, but the data suggests otherwise unless my motherboard is faulty and displaying incorrect numbers. PS: I’d really appreciate an explanation for that zero difference of around 9 points.
J
JonahLomu27
05-01-2016, 12:38 PM #1

I thought about it for a while too — mainly because I noticed no noticeable gap between the two configurations (only my wallet made a difference). I tried it with R3600 and GTX 1060, and now tested again with Ryzen 2200G, GTX 1060 plus 16GB Corsair Venegance "3200". Superposition scores show 16GB at 2133MHz vs 6334 points at 2933MHz for the stock. Based on these results, it seems buying faster RAM won’t really help for gaming. If you disagree, please clarify. I’m not including GN here since he often adjusts RAM specs and isn’t relevant to this point. This isn’t about overclocking; it’s about what you actually get when you spend more on speed. And yes, I might be mistaken, but the data suggests otherwise unless my motherboard is faulty and displaying incorrect numbers. PS: I’d really appreciate an explanation for that zero difference of around 9 points.

S
Slow_Burn351
Member
78
05-01-2016, 12:38 PM
#2
Wat bedoel je precies? Je vraagt om een versie van deze zin te maken.
S
Slow_Burn351
05-01-2016, 12:38 PM #2

Wat bedoel je precies? Je vraagt om een versie van deze zin te maken.

D
DazDaZz_
Junior Member
41
05-01-2016, 12:38 PM
#3
I believe a more effective method would help clearly observe the differences... though I might be mistaken, this tends to occur mainly in CPU-intensive situations—I’d guess that.
D
DazDaZz_
05-01-2016, 12:38 PM #3

I believe a more effective method would help clearly observe the differences... though I might be mistaken, this tends to occur mainly in CPU-intensive situations—I’d guess that.

T
TheGamerPro25
Member
106
05-01-2016, 12:38 PM
#4
You didn't verify it properly, which caused the issue.
T
TheGamerPro25
05-01-2016, 12:38 PM #4

You didn't verify it properly, which caused the issue.

C
Colefusion
Senior Member
382
05-01-2016, 12:38 PM
#5
Find the outcomes here. Note: Superposition isn't ideal for memory testing. It requires minimal CPU usage.
C
Colefusion
05-01-2016, 12:38 PM #5

Find the outcomes here. Note: Superposition isn't ideal for memory testing. It requires minimal CPU usage.

T
tomasaro
Member
68
05-01-2016, 12:38 PM
#6
That's correct, the 2133mhz CL14 rams are operating at 3200 CL20, which does improve performance, especially in RDR2. The issue is clear: you only ran one workload during testing.
T
tomasaro
05-01-2016, 12:38 PM #6

That's correct, the 2133mhz CL14 rams are operating at 3200 CL20, which does improve performance, especially in RDR2. The issue is clear: you only ran one workload during testing.

D
DinoCrakers
Member
102
05-01-2016, 12:38 PM
#7
Yes, you could be mistaken since that was a key marketing message even then. The claim about needing fast RAM for Ryzen was significant. The iGPU I didn’t use in this test is irrelevant because the same approach was originally designed for Ryzen CPUs without an iGPU. I’m not suggesting it’s entirely incorrect; rather, I’d like a clear explanation of why there might seem to be no difference.
D
DinoCrakers
05-01-2016, 12:38 PM #7

Yes, you could be mistaken since that was a key marketing message even then. The claim about needing fast RAM for Ryzen was significant. The iGPU I didn’t use in this test is irrelevant because the same approach was originally designed for Ryzen CPUs without an iGPU. I’m not suggesting it’s entirely incorrect; rather, I’d like a clear explanation of why there might seem to be no difference.

X
xanderzone317
Posting Freak
957
05-01-2016, 12:38 PM
#8
it’s more about GPU performance testing, but certain CPU tests like cinebench are less influenced by RAM. From what I understand, MHW could work if you run it with low graphics settings to focus on CPU usage. SOTR is another option—just adjust the graphics to a minimal level for a CPU-heavy test.
X
xanderzone317
05-01-2016, 12:38 PM #8

it’s more about GPU performance testing, but certain CPU tests like cinebench are less influenced by RAM. From what I understand, MHW could work if you run it with low graphics settings to focus on CPU usage. SOTR is another option—just adjust the graphics to a minimal level for a CPU-heavy test.

R
RepoRizer
Posting Freak
872
05-01-2016, 12:38 PM
#9
Absolutely, I ran several tests previously and the outcomes remained consistent. The performance didn’t jump even slightly, and the Cinebench scores stayed stable. Given this CPU is already operating near 90% capacity, improvements seem limited. Still, those figures are solid, and factors like RAM speed can play a bigger role than expected.
R
RepoRizer
05-01-2016, 12:38 PM #9

Absolutely, I ran several tests previously and the outcomes remained consistent. The performance didn’t jump even slightly, and the Cinebench scores stayed stable. Given this CPU is already operating near 90% capacity, improvements seem limited. Still, those figures are solid, and factors like RAM speed can play a bigger role than expected.

M
melle_bk
Junior Member
5
05-01-2016, 12:38 PM
#10
When the CPU isn't reaching its full capacity, your GPU becomes the limiting factor, and you're attempting to evaluate your CPU performance. This isn't the correct approach.
M
melle_bk
05-01-2016, 12:38 PM #10

When the CPU isn't reaching its full capacity, your GPU becomes the limiting factor, and you're attempting to evaluate your CPU performance. This isn't the correct approach.

Pages (3): 1 2 3 Next