Jump to content

Comprehensive Memory Overclocking Guide

Thx for reply!

 

My pc parts are in my profile (8086k dellided, asus maximus x hero mobo, 32Gb 16x2 g.skill 3200CL14 memkit)

 

Last 2 days I made a several tests to fully isolate CPU OC/cache and memory OC:

1. I succesfully finished prime95 tests (sse/avx2/fma3) with my 8086K at 5GHz / 4.7GHz cache 1.35v and RAM on auto (2133-15-15-15)

each one 6hrs of 512-4096k range with 26624mb RAM usage, 12threads, 15mins per test settings. I think it may be stable even at lower vcore (1.3 for example).

highest temps were 70C. Now we know that CPU is capable of this and can handle 5GHz error-free.

 

2. Time to play with memory. I locked CPU at 4GHz, turned off turbo, left cache at 4.7GHz, set constant vcore to 1.25v (higher than default). I set manually 3200 freq, turned off 100:100 memory multipliers, left only 100:133. Set 14-14-14-34-2N primary timings, set mem voltage to 1.35v. VCCIO to 1.1v, SA to 1.15v, ASUS Maximus tweak mode 1 (mode 2 sets all 2-3 subtimings very aggressive)

and... no go. Errors in prime95 (FMA3 enabled by default) in about 1-2 hours. 1-2 threads stops with errors.

Give VCCIO 1.2v, SA 1.25v - survived 4 hours and 1 thread stopped with error.

 

I'll try to enable XMP and see if it can pass prime95 in that way.

The only difference I see at the moment, that asus mobo set tFAW to 34 on auto, while there is 39 in XMP profile. And tRRD_L is 7 on auto instead of 8 in XMP. tRRD_S is 8 (auto) / 8 (XMP).

 

At this point I wonder - if it will not pass tests on XMP or on manually corrected settings fully according to XMP - is it a bad mem indicator or at least bad binned mem? Of course it may be CPU IMC, but 3200CL14 seems to be too easy even on 16Gb modules. RAM is in QVL list, also my asus mobo is in QVL list of g.skill for this sticks. But it works on standard JEDEC 2133CL15.

 

When I tried to test the limits of this kit - I were able to work with 3466CL15 "stable" (errored in prime95 also). 3600CL16 - only 50% of success boot each time (random D5/B2/55 errors on Qcode) with errors in memtest86 test 6 (block move). 3733CL17 - only boot to BIOS in about 33% success rate, massive errors in memtest86 in each test.

Seems my mem doesn't like voltage higher than 1.35, on that tests I usually got "55" when tried voltages between 1.37-1.40. Also it don't like anything less, when overclocked. On 1.325-1.33v massive errors in 3466CL15 in memtest86 (all tests).

 

I read your recommendations about RTL and RTL init settings. On 3200CL14 mode my mobo seems to train and set them correct each time from boot to boot (checked several times turning PC off and on and comparing what is set on AUTO). It trains to 53/53 for ch A rank 1/2 and to 55/55 for ch B rank 1/2. latency offset defaults to 21 setting IOL to 6 for both channels. I tried to change offset to 23, it rebooted several times showing "55" errors and stuck & working at 23 now, IOL changed to 4 then. Seems "retrained". I played with RTL also, but anything less than 53 for first channel don't work, after several retries bios loads in safe mode saying "overclocking failed". 53/54 instead of 53/55 works, but I think that 55 is more correct for case with dual ranked memory. Most of the guides and tweaks are for 16Gb kits (single ranked).

According to information on different sources my memory kit use Samsung B-Die chips.

 

I'll return tommorow after testing XMP mode and manual-XMP. Also will post aida64 bandwidth tests, timings/subtimings screenshots etc. I'm at work now.

 

Link to comment
Share on other sites

Link to post
Share on other sites

Left prime95 for night with manually set timings and asus tweak mode1 (relaxed secondary/tertiary timings) and it failed in 2 hours (1 thread errored in 2048k test pattern)

CPU were locked at 4GHz and cache at 3.7GHz.

here is a settings screenshot:

3lY3nZJm.png

 

ok, I decided to load XMP profile and entered XMP mode. Surprise, but it doesn't lock any settings or timings. If some timings already set manually - it doesn't change them.

So I moved all to auto and loaded again. System set manually primary timings 14-14-34, left command rate at auto (defaults to 2N in my case). Moved VCCIO and SA to 1.3v and thats all XMP had done. I reverted VCCIO and SA to 1.2v and set them manually. Timings such as tFAW not set correctly by activating XMP as well as tRRD_S/L. I locked them manually on what XMP should do. Also I switched to asus tweak mode2 (more strict timings) but locked tRDWR_xx at 14 instead of 9 which mode2 sets in auto.

Prime95 survived for 4 hours. I know it's not enough but I prefer to use computer at day for something more usefull that prime95 looping :) I'll left it at night again.

Here is screenshot of current settings:

HChuKrCm.png

 

Finally. Now I'm testing mem overclock with CPU locked at 4GHz (it's even slower than default mode with turboboost etc enabled). Isn't it useless? I mean - I understand that CPU overclock should be tested first to be sure it's not a problem and then move to memory overclock. But overclocked CPU AND overclocked memory usually demands more. I'm not sure that if current memory settings will be 100% stable - they will remain stable when I'll apply CPU overclock with already tested settings (1.35v vcore, 4.7GHz cache).

Link to comment
Share on other sites

Link to post
Share on other sites

15 hours ago, Gre4ka said:

Left prime95 for night with manually set timings and asus tweak mode1 (relaxed secondary/tertiary timings) and it failed in 2 hours (1 thread errored in 2048k test pattern)

CPU were locked at 4GHz and cache at 3.7GHz.

here is a settings screenshot:

3lY3nZJm.png

 

ok, I decided to load XMP profile and entered XMP mode. Surprise, but it doesn't lock any settings or timings. If some timings already set manually - it doesn't change them.

So I moved all to auto and loaded again. System set manually primary timings 14-14-34, left command rate at auto (defaults to 2N in my case). Moved VCCIO and SA to 1.3v and thats all XMP had done. I reverted VCCIO and SA to 1.2v and set them manually. Timings such as tFAW not set correctly by activating XMP as well as tRRD_S/L. I locked them manually on what XMP should do. Also I switched to asus tweak mode2 (more strict timings) but locked tRDWR_xx at 14 instead of 9 which mode2 sets in auto.

Prime95 survived for 4 hours. I know it's not enough but I prefer to use computer at day for something more usefull that prime95 looping :) I'll left it at night again.

Here is screenshot of current settings:

HChuKrCm.png

 

Finally. Now I'm testing mem overclock with CPU locked at 4GHz (it's even slower than default mode with turboboost etc enabled). Isn't it useless? I mean - I understand that CPU overclock should be tested first to be sure it's not a problem and then move to memory overclock. But overclocked CPU AND overclocked memory usually demands more. I'm not sure that if current memory settings will be 100% stable - they will remain stable when I'll apply CPU overclock with already tested settings (1.35v vcore, 4.7GHz cache).

Reduce tWRWR_DR/DD to a value of 15 each for now and give it another go. Those tertiary timings are far too tight for your dual rank kit.

My (incomplete) memory overclocking guide: 

 

Does memory speed impact gaming performance? Click here to find out!

On 1/2/2017 at 9:32 PM, MageTank said:

Sometimes, we all need a little inspiration.

 

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

Left tests for this night with CPU overclock together.

Prime95 in FMA3 mode that were stable with CPU 5GHz cache 4.7GHZ & mem at 2133CL15 for 6 hours crashed first worker after 27mins and about a half in less than 2 hours.

Lowered cache to 4.5GHz, pushed more to VCCIO/SA - 1.25v and:

Prime95 in non AVX/FMA mode (SSE) survived for 3 hours, but 1 worker crashed:

U7EUhN0m.png

 

Adjusted tWRWR_DR/DD to 15. Prime95 FMA3 survived about 2 hours before 1 thread errored (seems better?) Maybe something other is too tight too?

 

7aMli1wm.png cX8eGz0m.png

 

...

 

switched to asus tweak mode 1 (relaxed timings), put 1.23v both on VCCIO/SA:

 

RuNg7bem.png KC8bCAbm.png

 

no go, errors arrives in 3-4 hours

 

put 1.25v both on VCCIO/SA (1.26 idle, 1.27 under load), put 0.01v more on vcore to 1.36v (1.375v under load):

 

SDq14idm.png

 

:(

 

I can test on XMP default that put 1.31v both on VCCIO/SA, but I don't feel it comfortable. It's only 3200MHz... not 4GHz

one my mate overclocked same memory kit to 4GHz on cheaper mobo (strix series) with CL17 and much tighter secondary/tertiary timings... and never set VCCIO/SA higher than 1.2v

 

I'm really confused. Unfortunately I don't have another CPU to test and verify that it's IMC limitation of my sample. 3200MHz seems to me too easy to each CPU even with dual ranked mem.

 

Maybe 8Gb X 4 kit could work & overclock better? When I bought my kit I were able to get 16Gb x 2 (3200 CL14) or 8Gb x 4 (3200 CL14), I decided to buy 16x2. I'll try to write to G.Skill...

Edited by Gre4ka
info add
Link to comment
Share on other sites

Link to post
Share on other sites

I decided to test memory on lower frequency but with maximum tight timings.

 

2666 - 12-12-30-1N works perfect. With ASUS tweak mode 2 and all subtimings on auto I passed Prime95 FMA3 test for 12 hours w/o any errors. Also passes Memtest Pro 600% and RAMTest 6000% coverage. On such speed mobo trains RTL timings with a difference at 1 point (e.g. 47/48, 45/46 etc).

Here is my bandwidth and latency results.

 

2666-12-12-30-2N Mode2 default timings (all auto):

Q6PojR1m.png wft7yzzm.png

 

2666-12-12-30-1N Mode2 only tRFC & tREFI adjusted, other - all auto:

mIMQeIJm.png E2VEt8Tm.png

 

2666-12-12-30-1N Mode2 all timings tightened:

L8ye2SHm.png IqlxbcYm.png

 

I don't see & fell difference between maximum tightened timings and the option, where is only tRFC & tREFI adjusted. Tiny -0.01ns in latency and a ghost increase in memory copy, but also ghost decreasy in memory read... Seems that ASUS tweak mode2 already tight them best taking in mind stability & performance together.

Prime95 finishing 11hrs test with latest screenshot settings at the moment, no errors.

 

I'll try to go a bit higher (2800/2933/3000) with CL12 or 13.

Link to comment
Share on other sites

Link to post
Share on other sites

  • 1 month later...

Hi! So i decided to squeeze out the remanings of my old system:

I5 4690k @ 4.5ghz 1.38v

msi z97-g45 gaming

kingston hyperx fury 4x4gb 1866 ddr3

 

After some research it seems that my ram should be able to do around 2400mhz @ 1.65v and up to 2600mhz @ 1.85v, so i tried setting the volt to 1.85 and it didn't even post at 2000mhz, no matter how lose i set the timings. I tried 1.65v and now i could get it to boot at 2133mhz and rock stable at 2000 mhz, but it wouldn´t post at 2200. At 1.5v i can get it stable at 2400mhz cl 11, 12, 11, 30. It gets unstable at 1.515v and have trouble posting at 1.53v

 

At stock it can do 1.85v, but at 2400mhz it wont even post with 1.53v. Why will it manage lower and lower voltages the higher speed i set, seems like somethings wrong??

 

Thanks in advice!

Link to comment
Share on other sites

Link to post
Share on other sites

So I think this has been mentioned before, but I would've liked for this thread to be stickied.  I was looking into overclocking my RAM earlier this week, but didn't see this thread until today.  Thanks to @MageTank  for the post.

Link to comment
Share on other sites

Link to post
Share on other sites

  • 1 month later...

Im new in this world. What should i do when my dual ram is unstable with its xmp profile 2?

corsair vengeance lpx 2x4GB 2667mhz rated speed.

right now only 2400mhz which is profile 1 working.

mobo msi b450m bazooka plus. bios ver 1.0
ryzen 3 2200G 

Link to comment
Share on other sites

Link to post
Share on other sites

  • 4 weeks later...

Hey everyone, 

Is the "IMC Smackdown" test is still the 6. test in memtest (Block move, 64 moves)? I mean - since this guide was written - the order of the test still the same, just like in 2017?

 

Thanks in advance

Link to comment
Share on other sites

Link to post
Share on other sites

On 2/22/2019 at 1:02 PM, steve2563 said:

Hey everyone, 

Is the "IMC Smackdown" test is still the 6. test in memtest (Block move, 64 moves)? I mean - since this guide was written - the order of the test still the same, just like in 2017?

 

Thanks in advance

Yeah it's still the same as far as I remember.

CPU: Intel i7 7700K | GPU: ROG Strix GTX 1080Ti | PSU: Seasonic X-1250 (faulty) | Memory: Corsair Vengeance RGB 3200Mhz 16GB | OS Drive: Western Digital Black NVMe 250GB | Game Drive(s): Samsung 970 Evo 500GB, Hitachi 7K3000 3TB 3.5" | Motherboard: Gigabyte Z270x Gaming 7 | Case: Fractal Design Define S (No Window and modded front Panel) | Monitor(s): Dell S2716DG G-Sync 144Hz, Acer R240HY 60Hz (Dead) | Keyboard: G.SKILL RIPJAWS KM780R MX | Mouse: Steelseries Sensei 310 (Striked out parts are sold or dead, awaiting zen2 parts)

Link to comment
Share on other sites

Link to post
Share on other sites

Question for those (which I'm assuming most are these days) fiddling with DDR4, what's a "tight but safe" timing I can use for my tRFC? Right now I haven't pushed below 400 (and been stable) yet, I know the lower you can get stable on that the better.  Also a bit of an odd one, but one of my IO's is 2 different than the rest & from what I read of the Original Post so far, that's both not normal, and not supposed to be stable O_o lol

 

Note: The only timings I've manually set here so far are the primaries 17-17-17-38, tRFC to 400, and the voltage.  Haven't messed with anything else, the rest of that is auto.

 

Screenshot of my current timings that tested stable attached.

 

RAM-PassedMemTest64-1hr-SettingsnTemps.png

Save up to 10% on eVGA.com purchases, use my Associates Code: 47AQQ6KPU2IZNXY

My system on ModRigs: https://www.modsrigs.com/detail.aspx?BuildID=42686 

Primary Rig:

CPU: AMD 5950X at 4.65GHz 1.275V - Mobo: Asus Crosshair VIII Hero - PSU: eVGA P2 1200W

CPU Cooler: EK Quantum Velocity Block (480mm XE Radiator with push/pull EK Vardar D-RGB fans)

2 x 16GB G.Skill Trident Z Royal 3600MHz CL14 RAM (14-15-15-35-1T at 1.45V)

GPU: eVGA RTX 3090 Kingpin Hybrid - Core @ 2160MHz @ 1068mV, VRAM +1000MHz

Case: Thermaltake View 91 - SSDs/HDDs: Too many to list; Samsung 980 Pro 1TB & 2TB M.2s, Samsung 4TB SATA SSD

Monitor: Samsung Odyssey G9 49" Super-Ultrawide 240Hz Monitor

Link to comment
Share on other sites

Link to post
Share on other sites

  • 2 weeks later...
On 3/1/2019 at 11:51 AM, HoneyBadger84 said:

Question for those (which I'm assuming most are these days) fiddling with DDR4, what's a "tight but safe" timing I can use for my tRFC? Right now I haven't pushed below 400 (and been stable) yet, I know the lower you can get stable on that the better.  Also a bit of an odd one, but one of my IO's is 2 different than the rest & from what I read of the Original Post so far, that's both not normal, and not supposed to be stable O_o lol

 

Note: The only timings I've manually set here so far are the primaries 17-17-17-38, tRFC to 400, and the voltage.  Haven't messed with anything else, the rest of that is auto.

 

Screenshot of my current timings that tested stable attached.

 

 

Is that 1.6+v in your ram?

Link to comment
Share on other sites

Link to post
Share on other sites

I cant get my RAM stable, they're rated at 3200mhz 16-18-18-38 but if i even dare increase the speed one strap it won't boot at all, if i tighten up my timings it all goes wrong too. My cpu is IBT and x264 stable but no matter what I try I cannot get memory to overclock. What am I doing wrong? Also in this thread its suggested VCCIO and SA voltages not exceed 1.25v but when they're on auto they go above that themselves?

Link to comment
Share on other sites

Link to post
Share on other sites

10 hours ago, Benjeh said:

Is that 1.6+v in your ram?

No, it's reading out at 1.38-1.404V (DDR readout top right quadrant), set at 1.4V in BIOS. 

Save up to 10% on eVGA.com purchases, use my Associates Code: 47AQQ6KPU2IZNXY

My system on ModRigs: https://www.modsrigs.com/detail.aspx?BuildID=42686 

Primary Rig:

CPU: AMD 5950X at 4.65GHz 1.275V - Mobo: Asus Crosshair VIII Hero - PSU: eVGA P2 1200W

CPU Cooler: EK Quantum Velocity Block (480mm XE Radiator with push/pull EK Vardar D-RGB fans)

2 x 16GB G.Skill Trident Z Royal 3600MHz CL14 RAM (14-15-15-35-1T at 1.45V)

GPU: eVGA RTX 3090 Kingpin Hybrid - Core @ 2160MHz @ 1068mV, VRAM +1000MHz

Case: Thermaltake View 91 - SSDs/HDDs: Too many to list; Samsung 980 Pro 1TB & 2TB M.2s, Samsung 4TB SATA SSD

Monitor: Samsung Odyssey G9 49" Super-Ultrawide 240Hz Monitor

Link to comment
Share on other sites

Link to post
Share on other sites

7 hours ago, Benjeh said:

I cant get my RAM stable, they're rated at 3200mhz 16-18-18-38 but if i even dare increase the speed one strap it won't boot at all, if i tighten up my timings it all goes wrong too. My cpu is IBT and x264 stable but no matter what I try I cannot get memory to overclock. What am I doing wrong? Also in this thread its suggested VCCIO and SA voltages not exceed 1.25v but when they're on auto they go above that themselves?

Set your VCCIO and VCCSA manually so your system doesn't kill them. I recommend a 1.2V starting point on those. What chipset are you on and what processor are you running? For Z390/9900K, anything  at or under ~1.35V is fine for daily use according to most overclockers, ram voltage anything at or under 1.5V. 

Save up to 10% on eVGA.com purchases, use my Associates Code: 47AQQ6KPU2IZNXY

My system on ModRigs: https://www.modsrigs.com/detail.aspx?BuildID=42686 

Primary Rig:

CPU: AMD 5950X at 4.65GHz 1.275V - Mobo: Asus Crosshair VIII Hero - PSU: eVGA P2 1200W

CPU Cooler: EK Quantum Velocity Block (480mm XE Radiator with push/pull EK Vardar D-RGB fans)

2 x 16GB G.Skill Trident Z Royal 3600MHz CL14 RAM (14-15-15-35-1T at 1.45V)

GPU: eVGA RTX 3090 Kingpin Hybrid - Core @ 2160MHz @ 1068mV, VRAM +1000MHz

Case: Thermaltake View 91 - SSDs/HDDs: Too many to list; Samsung 980 Pro 1TB & 2TB M.2s, Samsung 4TB SATA SSD

Monitor: Samsung Odyssey G9 49" Super-Ultrawide 240Hz Monitor

Link to comment
Share on other sites

Link to post
Share on other sites

5 hours ago, HoneyBadger84 said:

Set your VCCIO and VCCSA manually so your system doesn't kill them. I recommend a 1.2V starting point on those. What chipset are you on and what processor are you running? For Z390/9900K, anything  at or under ~1.35V is fine for daily use according to most overclockers, ram voltage anything at or under 1.5V. 

Hi mate, Z170/6700K is what i'm playing with, bios is up to date, maybe my cpu overclock is the actual culprit I'm not really too sure now, maybe the passes in my stress tests was a fluke.

Link to comment
Share on other sites

Link to post
Share on other sites

35 minutes ago, Benjeh said:

Hi mate, Z170/6700K is what i'm playing with, bios is up to date, maybe my cpu overclock is the actual culprit I'm not really too sure now, maybe the passes in my stress tests was a fluke.

I would do initial memory overclocking with the rest of the system at stock. This rules out everything else as being the issue, do if you encounter a failure, blue screen or the like, it has to be the memory.  That's how I've done mine so far. Then you combine the two (CPU & RAM) OCs and adjust for stability of both. 

Save up to 10% on eVGA.com purchases, use my Associates Code: 47AQQ6KPU2IZNXY

My system on ModRigs: https://www.modsrigs.com/detail.aspx?BuildID=42686 

Primary Rig:

CPU: AMD 5950X at 4.65GHz 1.275V - Mobo: Asus Crosshair VIII Hero - PSU: eVGA P2 1200W

CPU Cooler: EK Quantum Velocity Block (480mm XE Radiator with push/pull EK Vardar D-RGB fans)

2 x 16GB G.Skill Trident Z Royal 3600MHz CL14 RAM (14-15-15-35-1T at 1.45V)

GPU: eVGA RTX 3090 Kingpin Hybrid - Core @ 2160MHz @ 1068mV, VRAM +1000MHz

Case: Thermaltake View 91 - SSDs/HDDs: Too many to list; Samsung 980 Pro 1TB & 2TB M.2s, Samsung 4TB SATA SSD

Monitor: Samsung Odyssey G9 49" Super-Ultrawide 240Hz Monitor

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, HoneyBadger84 said:

I would do initial memory overclocking with the rest of the system at stock. This rules out everything else as being the issue, do if you encounter a failure, blue screen or the like, it has to be the memory.  That's how I've done mine so far. Then you combine the two (CPU & RAM) OCs and adjust for stability of both. 

It was an unstable CPU clock, basically my CPU can't go further than 4712mhz and the uncore speed is utter pants too. I'm working my way up with the ram now, the good thing that's come out of this is I've dropped from 1.5+v to 1.39-1.42v I can probably go lower, will see, this just further cements my believe that the chip I have is utterly useless.

Link to comment
Share on other sites

Link to post
Share on other sites

  • 2 months later...

hey how about diving into elaborating on those tertiary timings a bit more with more experience :) 

Link to comment
Share on other sites

Link to post
Share on other sites

4 hours ago, Wonderin said:

hey how about diving i to elaborating on those tertiary timings a but more with more experience :) 

I can try, though truth be told I don't understand much about them. That being said, I can certainly point out which have yielded the most performance benefits from my personal tests. This guide is mostly comprised of trial and error, with most of the terminology coming from those that taught me most of what I know about memory overclocking. 

 

The funny part is that I looked back at this guide and completely forgot to add the Ryzen part, despite buying my Ryzen components well over a year ago. I guess the silver lining is that it might come in handy for Ryzen 3000.

My (incomplete) memory overclocking guide: 

 

Does memory speed impact gaming performance? Click here to find out!

On 1/2/2017 at 9:32 PM, MageTank said:

Sometimes, we all need a little inspiration.

 

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

So it seems that the most important part is primary timings.. My pc is oced except for the memory and i can't handle my OCD without trying to speed memory even a little bit :D.

I'm using broadwell-e processor and 3000 mhz hunix-mfr based memory kit (Kingston Predator (HX430C15PB3K2/32). Timings are a bit boring (15-17-17-36-2T) huh. I know it might be even not worth trying but hey, I should calm down a bit after i will see tighter timings or/and some extra 200 mhz frequency.

Link to comment
Share on other sites

Link to post
Share on other sites

4 hours ago, Wonderin said:

So it seems that the most important part is primary timings.. My pc is oced except for the memory and i can't handle my OCD without trying to speed memory even a little bit :D.

I'm using broadwell-e processor and 3000 mhz hunix-mfr based memory kit (Kingston Predator (HX430C15PB3K2/32). Timings are a bit boring (15-17-17-36-2T) huh. I know it might be even not worth trying but hey, I should calm down a bit after i will see tighter timings or/and some extra 200 mhz frequency.

From what I remember mage telling about hynix kits is they either clock high or they get very tight timings but never both, that's a trait of Samsung memory to have the best of both.

CPU: Intel i7 7700K | GPU: ROG Strix GTX 1080Ti | PSU: Seasonic X-1250 (faulty) | Memory: Corsair Vengeance RGB 3200Mhz 16GB | OS Drive: Western Digital Black NVMe 250GB | Game Drive(s): Samsung 970 Evo 500GB, Hitachi 7K3000 3TB 3.5" | Motherboard: Gigabyte Z270x Gaming 7 | Case: Fractal Design Define S (No Window and modded front Panel) | Monitor(s): Dell S2716DG G-Sync 144Hz, Acer R240HY 60Hz (Dead) | Keyboard: G.SKILL RIPJAWS KM780R MX | Mouse: Steelseries Sensei 310 (Striked out parts are sold or dead, awaiting zen2 parts)

Link to comment
Share on other sites

Link to post
Share on other sites

I just don't want to spend month adjusting secondary and tertiary timings without any guidelines for my memory type :(

 

ok seems like i got it wrong and i have afr memory, not mfr..

Link to comment
Share on other sites

Link to post
Share on other sites

  • 4 months later...

From you article: "I'll be frank here. I have no idea what SG or DG means..."

"SG" is "Same Group" (meaning the same memory page)

"DG" is "Different Group" (the same way that "DR" is "Different Rank" and "DD" is "Different DIMM")

 

Thoroughly enjoyed your guide, though.

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×