F5F Stay Refreshed Hardware Desktop 9Gen memory controller (IMC) design

9Gen memory controller (IMC) design

9Gen memory controller (IMC) design

Pages (3): Previous 1 2 3 Next
P
Poseidon3131
Junior Member
14
03-27-2025, 12:00 PM
#11
Thanks for the feedback! I might try 3733-3800 next and observe the results. I recall reading that all Gygabyte Z390 boards use a T topology, though I'm not entirely sure. Just checked—my Corsairs are running at 2933Mhz with 1.23v, both auto, VCCIO at 0.950v and agent voltage at 1.050v. Would increasing them affect CPU temps?
P
Poseidon3131
03-27-2025, 12:00 PM #11

Thanks for the feedback! I might try 3733-3800 next and observe the results. I recall reading that all Gygabyte Z390 boards use a T topology, though I'm not entirely sure. Just checked—my Corsairs are running at 2933Mhz with 1.23v, both auto, VCCIO at 0.950v and agent voltage at 1.050v. Would increasing them affect CPU temps?

M
MoutazGamer
Junior Member
36
03-27-2025, 03:54 PM
#12
It's B-die too, just a lower bin. 3200 CL14. The system is running at 17-18-18-36 with fully customized secondary and tertiary settings. I haven't adjusted the IOL or RTL yet, but I expect changes soon. A few people say it works, but I haven't really felt much difference myself. I'm sure it does, though.
M
MoutazGamer
03-27-2025, 03:54 PM #12

It's B-die too, just a lower bin. 3200 CL14. The system is running at 17-18-18-36 with fully customized secondary and tertiary settings. I haven't adjusted the IOL or RTL yet, but I expect changes soon. A few people say it works, but I haven't really felt much difference myself. I'm sure it does, though.

T
taconiebre
Senior Member
506
04-09-2025, 11:00 PM
#13
This discussion is worth exploring for more RAM overclocking insights. It often gets overlooked since it isn't flashy, but the results are impressive.
T
taconiebre
04-09-2025, 11:00 PM #13

This discussion is worth exploring for more RAM overclocking insights. It often gets overlooked since it isn't flashy, but the results are impressive.

M
Mr_HorseS
Junior Member
21
04-11-2025, 01:55 PM
#14
Topology should outperform a simple daisy chain with four components when executed properly, though this becomes less significant around 3800-4000MHz. Reducing their numbers also decreased CPU temperatures slightly in my setup, but a drop to 0.3V is quite noticeable. Your auto-voltage settings are acceptable; modest adjustments for stability likely won’t affect performance much.
M
Mr_HorseS
04-11-2025, 01:55 PM #14

Topology should outperform a simple daisy chain with four components when executed properly, though this becomes less significant around 3800-4000MHz. Reducing their numbers also decreased CPU temperatures slightly in my setup, but a drop to 0.3V is quite noticeable. Your auto-voltage settings are acceptable; modest adjustments for stability likely won’t affect performance much.

J
JokerFame
Senior Member
670
04-11-2025, 07:49 PM
#15
Are you sure about challenging the 2T Brigade? I’m here to provide a serious perspective. For me, 2T is essential for maintaining optimal performance. Pushing 1T actually harms my tertiary timings and disrupts RTL/IO-L accuracy. Thanks for the read—hope it was helpful. I’m still working on refining it with more data. I asked them not to make the guide sticky since I’m short on time and think people can easily find answers online.
J
JokerFame
04-11-2025, 07:49 PM #15

Are you sure about challenging the 2T Brigade? I’m here to provide a serious perspective. For me, 2T is essential for maintaining optimal performance. Pushing 1T actually harms my tertiary timings and disrupts RTL/IO-L accuracy. Thanks for the read—hope it was helpful. I’m still working on refining it with more data. I asked them not to make the guide sticky since I’m short on time and think people can easily find answers online.

W
Wither01
Member
201
04-19-2025, 07:27 AM
#16
I possess two configurations. One runs at 3600 and 4267mhz, both rated B. The other operates at 2700x with good performance up to 4000mhz and 8700K only at 4300mhz. On the Intel board, during benchmarking at 4000mhz with a 14-14-14 configuration on the 4267mhz board, the 3600 model supports XMP timings at 4000mhz but requires slightly higher voltage—typically not exceeding 1.6v for memory.
W
Wither01
04-19-2025, 07:27 AM #16

I possess two configurations. One runs at 3600 and 4267mhz, both rated B. The other operates at 2700x with good performance up to 4000mhz and 8700K only at 4300mhz. On the Intel board, during benchmarking at 4000mhz with a 14-14-14 configuration on the 4267mhz board, the 3600 model supports XMP timings at 4000mhz but requires slightly higher voltage—typically not exceeding 1.6v for memory.

S
SPIKEBALL21
Member
135
04-21-2025, 05:28 PM
#17
I avoid benching and prefer constant performance. I've reached up to 1.6V, but it seems my setup doesn't adjust well to higher voltages. At 1.6V I still can't match what works at 1.4V, and I'd rather stay below 1.5V for regular use.
S
SPIKEBALL21
04-21-2025, 05:28 PM #17

I avoid benching and prefer constant performance. I've reached up to 1.6V, but it seems my setup doesn't adjust well to higher voltages. At 1.6V I still can't match what works at 1.4V, and I'd rather stay below 1.5V for regular use.

K
KiotPlays
Junior Member
14
04-22-2025, 08:48 AM
#18
Daily 3733 operates well around 1.375-1.40v with scaling adjustments. You're likely using Cas 10-12 at XMP speeds and adding a voltage boost.
K
KiotPlays
04-22-2025, 08:48 AM #18

Daily 3733 operates well around 1.375-1.40v with scaling adjustments. You're likely using Cas 10-12 at XMP speeds and adding a voltage boost.

_
_FreeZe_YT_
Member
145
04-23-2025, 07:00 PM
#19
This setup seems efficient but has some trade-offs. I’ll monitor the voltages and temperatures closely after they arrive. I noticed the airflow was subpar because I only used two 120mm fans instead of four, which might affect performance. The cooler also had fewer fans than intended—only one 120mm versus the recommended two 140mm. Right now it’s capped at 4.7Ghz with a BIOS setting of 1.265v, AVX disabled, and AVR temperatures between 55-60°C for gaming. However, under stress testing for over 30 minutes, it risks overheating and throttling, so I want to avoid excessive scaling.
_
_FreeZe_YT_
04-23-2025, 07:00 PM #19

This setup seems efficient but has some trade-offs. I’ll monitor the voltages and temperatures closely after they arrive. I noticed the airflow was subpar because I only used two 120mm fans instead of four, which might affect performance. The cooler also had fewer fans than intended—only one 120mm versus the recommended two 140mm. Right now it’s capped at 4.7Ghz with a BIOS setting of 1.265v, AVX disabled, and AVR temperatures between 55-60°C for gaming. However, under stress testing for over 30 minutes, it risks overheating and throttling, so I want to avoid excessive scaling.

J
jake8897
Junior Member
15
05-08-2025, 06:38 AM
#20
Is this the maximum allowed? It’s often wise to avoid setting it too high. Vdroop performs well and keeps the CPU in good shape. This could explain why your temperatures spike during stress tests.
J
jake8897
05-08-2025, 06:38 AM #20

Is this the maximum allowed? It’s often wise to avoid setting it too high. Vdroop performs well and keeps the CPU in good shape. This could explain why your temperatures spike during stress tests.

Pages (3): Previous 1 2 3 Next