Overclock.net - An Overclocking Community - Reply to Topic
Thread: *Official* Intel DDR4 24/7 Memory Stability Thread Reply to Thread
Title:
Message:

Register Now

In order to be able to post messages on the Overclock.net - An Overclocking Community forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
If you do not want to register, fill this field only and the name will be used as user name for your post.
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:

Log-in


  Additional Options
Miscellaneous Options

  Topic Review (Newest First)
07-09-2020 01:40 PM
reflex75
Quote: Originally Posted by warbucks View Post
Quote: Originally Posted by reflex75 View Post
New G.Skill 4x8GB 4000Mhz CL15
Tested at 3700 CL13 for better latency.
Nice. I've got the same kit.
Yes they are very nice indeed, providing both high frequency and fast timings!
I wonder if they are the best binned RAM available for 32GB (4×8GB)...
07-09-2020 11:15 AM
Gregix So, there it is. Seems stable, I'll put it for night test for like24k%+ to be sure.
As you can see, this is almost exact copy/paste of @XGS-Duplicity from his 3900c14 tests.
Will try tune RTL, as it seems have some meaningful impact on scores/performance but from my test at the moment it is still faster than any previous setting I had. I think I polish it little more(damn RTL!) and skip any 4000+Mhz tests, as they are more stressful to my IMC/MB setup and just...meh, not worth it.

And yeah, thanks to @Roidsy I just set my son's TG 3200c14 at [email protected] and it seems stable too. Went through 1000% Ramtest so far, he is gaming since(3+hrs on) so it is success so far. His aida64 latency score went from 50ish to nice 42ns area, while boosting throughout so yeah. Seems faster in games he play(minimum FPS up).
07-09-2020 08:28 AM
warbucks
Quote: Originally Posted by reflex75 View Post
New G.Skill 4x8GB 4000Mhz CL15
Tested at 3700 CL13 for better latency.
Nice. I've got the same kit.
07-08-2020 08:23 PM
munternet
Quote: Originally Posted by Gregix View Post
Thx man, that gives me hope, I would be happy if sons 8700k plus Msi z370 plus will take like ~3800c15 level somehow. This CPU is dud, very bad bin, and MB isn't good either.

And, for my last results, you were right guys, more juice for CPU, Vcore from 1.265 >>1.275 did the trick, no freezes in game now. Karhu ramtest still stable. Heck, I had Karhu, plus GPU-Z Render test and Aida64 stability test running same time till 2000%+, no fuz, MB temp 29 degree constant. Gaming wise(WOT) was the same temps.
It is nice to see smooth game with boosted FPS like from ~240 to ~270+ area from tweaking memory only. Most gains and finally no stutter I had from proper setting RTL /IOLs (if everything fails, read manual...in this case formula is in fkn RTL description...). I am still thinking I could squeeze more at this frequency from them...
As reminder, this is Patriot Viper Steel 4500c19 running now 3800 14-14-14-32 t2 @1.52V, 1.288 SA 1.240 IO.
IF I get bored, I guess will try 4100c15...^^

Heh, I just booted same settings but 3900Mhz, testing in progress.
Ok, no go, performance worse...
After many recommendations from users here I settled on GSAT and TestMem5 with Extreme .config file for testing the ram after making sure the CPU is 100% with small fft's.
Only error I have had since using this method is a crash to desktop in BFV which was fixed with a smidge more vccsa
07-08-2020 04:38 PM
Jpmboy
Quote: Originally Posted by Gregix View Post
Yeah, that might be vcore.
I need to test more(gaming now) as u can see, ramtest goes flawlessly.
TBH That ram goes 3900c14, but with 1.63V or more, so it is kinda no go for 24/7 for me.
3800c14 results below. Again, cr1 required too much voltage for my liking...
Temps are no issue here, when CPU/MB temp spikes up, tornado in my case starts
yeah, ram stress tests are fine for testing the ram settings, but since during the test the IMC is experiencing a continuous (monolithic) load it really cannot reflect real world/gaming load variations. What trips up any of these rigs (multiple, aligned logic circuits with embedded checksums = "machine checks") is rapidly varying operations... exactly what gaming does. So, yeah it is really important to stress test the ram configuration, but that simply address that one domain many times revealing other areas that need to be tuned to the new ram freq and timing set. All that said... "practice like you're gonna play" kinda sums it all up.
07-08-2020 02:48 PM
Gregix
Quote: Originally Posted by Roidsy View Post
I used to have those, ran them at 4133 16-16-16-34 cr 2t in 2x8 and 3600 14-14-14-32 cr 1t in 4x8. Couldn't get 3733 c14 or 4266 c16 stable in any configuration (at 1.5v, that is). They would also post at 4400 cl19 in 2x8 but it would crash as soon as windows loaded so I didn't bother with that.
Thx man, that gives me hope, I would be happy if sons 8700k plus Msi z370 plus will take like ~3800c15 level somehow. This CPU is dud, very bad bin, and MB isn't good either.

And, for my last results, you were right guys, more juice for CPU, Vcore from 1.265 >>1.275 did the trick, no freezes in game now. Karhu ramtest still stable. Heck, I had Karhu, plus GPU-Z Render test and Aida64 stability test running same time till 2000%+, no fuz, MB temp 29 degree constant. Gaming wise(WOT) was the same temps.
It is nice to see smooth game with boosted FPS like from ~240 to ~270+ area from tweaking memory only. Most gains and finally no stutter I had from proper setting RTL /IOLs (if everything fails, read manual...in this case formula is in fkn RTL description...). I am still thinking I could squeeze more at this frequency from them...
As reminder, this is Patriot Viper Steel 4500c19 running now 3800 14-14-14-32 t2 @1.52V, 1.288 SA 1.240 IO.
IF I get bored, I guess will try 4100c15...^^

Heh, I just booted same settings but 3900Mhz, testing in progress.
Ok, no go, performance worse...
07-08-2020 07:26 AM
reflex75
Quote: Originally Posted by XGS-Duplicity View Post
@reflex75 Nice latency [IMG class=inlineimg]/forum/images/smilies/smile.gif[/IMG]. Is that 2 dimms or 4 dimms? Which test did you use for stability? have you tried cas 15 @ 4200?
Thanks.
It's 4 dimms (4x8) and just for fun and benchmark purpose, not completely stable.
I don't have fan on them, so temperature can increase pretty high during heavy gaming with my air GPU.
Stock primary timings (4000 CL15) are safer, and my motherboard doesn't like high frequency 4200+....
07-08-2020 05:12 AM
XGS-Duplicity I have a question. I started using gigabyte eztune software. I noticed using xmp automatically sets my VCCPLL OC voltage to 1.31v revealed by the eztuner software(not viewable in hwinfo64). Is this voltage something I should work on lowering or is it ok at 1.31v?

@reflex75 Nice latency . Is that 2 dimms or 4 dimms? Which test did you use for stability? have you tried cas 15 @ 4200?
07-08-2020 12:39 AM
reflex75 New G.Skill 4x8GB 4000Mhz CL15
Tested at 3700 CL13 for better latency.
07-08-2020 12:11 AM
Roidsy
Quote: Originally Posted by Gregix View Post
Second question, can't find any decent results OC Teamgroup 3200 14-14-14-31 on intel platform. I tried some OC but this pack seems different than other b-die. Any user here?
I used to have those, ran them at 4133 16-16-16-34 cr 2t in 2x8 and 3600 14-14-14-32 cr 1t in 4x8. Couldn't get 3733 c14 or 4266 c16 stable in any configuration (at 1.5v, that is). They would also post at 4400 cl19 in 2x8 but it would crash as soon as windows loaded so I didn't bother with that.
This thread has more than 10 replies. Click here to review the whole thread.

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off