F5F Stay Refreshed Hardware Desktop RAM speed and timings influence performance in a positive way.

RAM speed and timings influence performance in a positive way.

RAM speed and timings influence performance in a positive way.

S
Super_AapjexD
Posting Freak
766
07-18-2019, 06:15 AM
#1
I noticed someone suggesting ram speed isn't crucial on Ryzen. I thought it'd be helpful before pushing timings too hard. Over the next week or two, once secondary timings for 3733 are sorted, I might adjust other speeds, try a few games (only a handful), and gauge actual performance. Current setup: 3.7ghz @ 1.0125v. Geekbench multithreaded scores show single-thread gains, lowest at 1110, now 1125—no monitoring yet until timings are tight. Key ballistix 3000mhz cl15 specs: 1.38v, 2400, with timings around 3600, 3600, and a few others. Minimum possible times listed range from 3600 to 6803. I'll review these once the settings are locked in.
S
Super_AapjexD
07-18-2019, 06:15 AM #1

I noticed someone suggesting ram speed isn't crucial on Ryzen. I thought it'd be helpful before pushing timings too hard. Over the next week or two, once secondary timings for 3733 are sorted, I might adjust other speeds, try a few games (only a handful), and gauge actual performance. Current setup: 3.7ghz @ 1.0125v. Geekbench multithreaded scores show single-thread gains, lowest at 1110, now 1125—no monitoring yet until timings are tight. Key ballistix 3000mhz cl15 specs: 1.38v, 2400, with timings around 3600, 3600, and a few others. Minimum possible times listed range from 3600 to 6803. I'll review these once the settings are locked in.

K
karlerik_1999
Member
205
07-18-2019, 03:51 PM
#2
This appears to be the Crucial Ballistix, though it's unclear which specific model you're referring to. The timing seems a bit off for high-end RAM.
K
karlerik_1999
07-18-2019, 03:51 PM #2

This appears to be the Crucial Ballistix, though it's unclear which specific model you're referring to. The timing seems a bit off for high-end RAM.

L
lemonlimer43
Junior Member
18
07-18-2019, 04:58 PM
#3
Yeah, sorry, I assumed it was in Ballistix with XMP 3000 CL15
L
lemonlimer43
07-18-2019, 04:58 PM #3

Yeah, sorry, I assumed it was in Ballistix with XMP 3000 CL15

O
Oxidian_LP
Member
205
07-22-2019, 03:33 PM
#4
That's the Micron E-die. It's not particularly high quality, but it's not terrible either.
O
Oxidian_LP
07-22-2019, 03:33 PM #4

That's the Micron E-die. It's not particularly high quality, but it's not terrible either.

S
Sly202001
Member
169
07-22-2019, 11:05 PM
#5
It seems like you're recalling a frustrating experience with a Samsung device that won't start up.
S
Sly202001
07-22-2019, 11:05 PM #5

It seems like you're recalling a frustrating experience with a Samsung device that won't start up.

P
PinkSapphire
Junior Member
14
07-23-2019, 12:32 AM
#6
Checking if you're using 1:1 on Infinity Fabric or just RAM overclocking.
P
PinkSapphire
07-23-2019, 12:32 AM #6

Checking if you're using 1:1 on Infinity Fabric or just RAM overclocking.

M
ManMallow
Member
223
07-23-2019, 07:33 AM
#7
Sure, the CPU can't handle 1900 at that speed.
M
ManMallow
07-23-2019, 07:33 AM #7

Sure, the CPU can't handle 1900 at that speed.

I
193
07-25-2019, 08:17 AM
#8
Which BIOS version are you using? You need to revert four versions to access the 1900 feature.
I
itzMikaHackzZ_
07-25-2019, 08:17 AM #8

Which BIOS version are you using? You need to revert four versions to access the 1900 feature.

T
TheFallenRose
Senior Member
616
07-26-2019, 11:46 PM
#9
most recent updates show no previous version succeeded, but 1867 comes near. Tests weren't enough before; I switched to anta777's extreme1 testmem5 and used the large data set or memory test.
T
TheFallenRose
07-26-2019, 11:46 PM #9

most recent updates show no previous version succeeded, but 1867 comes near. Tests weren't enough before; I switched to anta777's extreme1 testmem5 and used the large data set or memory test.

M
MrSarx
Senior Member
375
07-27-2019, 05:50 PM
#10
Shadow of the tomb raider benchmark 2400 auto timings (16-16-16-16-junk) average frames per second: 98. GPU usage: 51% FPS | CPU game | CPU render | GPU min: 78 | 108 | 77 max: 153 | 340 | 172 Avg: 114 | 178 | 105 at 95% | 81 | 121 | 83 3000mhz xmp profile (15-16-16-16-) average FPS: 101. GPU bound 57% FPS | CPU game | CPU render | GPU min: 81 | 113 | 77 max: 166 | 354 | 177 Avg: 122 | 186 | 106 at 95% | 85 | 129 | 82 3733 (I'll take screenshots later when convenient) average FPS: 105. GPU bound 97% FPS | CPU game | CPU render | GPU min: 106 | 121 | 76 max: 197 | 371 | 182 Avg: 150 | 195 | 107 at 95% | 110 | 125 | 83 Geekbench score: 1124/7259 Ryzen master screenshots each profile Spoiler
M
MrSarx
07-27-2019, 05:50 PM #10

Shadow of the tomb raider benchmark 2400 auto timings (16-16-16-16-junk) average frames per second: 98. GPU usage: 51% FPS | CPU game | CPU render | GPU min: 78 | 108 | 77 max: 153 | 340 | 172 Avg: 114 | 178 | 105 at 95% | 81 | 121 | 83 3000mhz xmp profile (15-16-16-16-) average FPS: 101. GPU bound 57% FPS | CPU game | CPU render | GPU min: 81 | 113 | 77 max: 166 | 354 | 177 Avg: 122 | 186 | 106 at 95% | 85 | 129 | 82 3733 (I'll take screenshots later when convenient) average FPS: 105. GPU bound 97% FPS | CPU game | CPU render | GPU min: 106 | 121 | 76 max: 197 | 371 | 182 Avg: 150 | 195 | 107 at 95% | 110 | 125 | 83 Geekbench score: 1124/7259 Ryzen master screenshots each profile Spoiler