ROG Crosshair VI overclocking thread - Page 389 - Overclock.net - An Overclocking Community

Forum Jump: 

ROG Crosshair VI overclocking thread

Reply
 
Thread Tools
post #3881 of 42763 (permalink) Old 03-22-2017, 11:41 AM
New to Overclock.net
 
CwStrife's Avatar
 
Join Date: Mar 2017
Posts: 90
Rep: 2 (Unique: 2)
Quote:
Originally Posted by warreng5995 View Post

Ok I've seen the core performance boost setting before, I'll turn it off.

This ram is rated for
15-15-15-35
Cas Latency 15
Voltage 1.35V

I've never really messed with ram timing. By "relaxed" do you mean I should set it to:

16-16-16-35?

If i remember right a few posts back one of these BIOS versions doesn't like odd ram timings? Someone correct me if i'm wrong. Try something like 18-16-16-16-38. You'll see in bios ram timings just follow those numbers
CwStrife is offline  
Sponsored Links
Advertisement
 
post #3882 of 42763 (permalink) Old 03-22-2017, 11:43 AM
 
Join Date: Mar 2016
Location: France/Armenia/US/Lebanon
Posts: 4,627
Quote:
Originally Posted by finalheaven View Post

He goes for near perfect stability. Hours and hours of testing.

What are your voltages/LLC? and what testing do you run for your chip?

Voltage on DMM is 1.238 for 3.8ghz, but he has a 1700 and i got a 1700x so slightly different binning. I haven't run a full stability test yet but been gaming on it all day without issues so far, and thats with ram at 2933hz as well.

The chip was stable on my G5 before it bricked and that was at 1.245 so pretty close in voltages. Was stable 12 passes of cinebench and 15mins of real bench, then 15mins of cpuz and gpuz stress test at the same time.
bluej511 is offline  
post #3883 of 42763 (permalink) Old 03-22-2017, 11:45 AM
Meddling user
 
gupsterg's Avatar
 
Join Date: Jan 2015
Location: Lurking over a keyboard
Posts: 7,076
Rep: 768 (Unique: 351)
Quote:
Originally Posted by bluej511 View Post

You need 1.35 for 3.8 stable really?

Please take no offense by this wink.gif , it is merely a discussion similar to one's we have had in the past my online buddy smile.gif .

My stability testing is far greater than yours.
Warning: Spoiler! (Click to show)
Quote:
Originally Posted by gupsterg View Post

Quote:
Originally Posted by bluej511 View Post

I love how the 1700x i got can do 3.8 stable at 1.245v if not even lower.
Not getting into how you stability test OC, etc smile.gif . Just curious a) what you do to be happy for stabilty? b) that OC on the past owned G5 or the C6H?
Quote:
Originally Posted by bluej511 View Post

15mins of real bench for base stability then a few cinebench runs one right after the other then game all day. Was on the g5. About to try docp standard to get my ram to 3200 on the ch6

For example I can run test of x264 for 10 loops at less (offset +106.25mV), 2 hrs RealBench, [email protected] 4-6hrs. Realbench on my CPU needs less of an offset for 3.8GHz than x264 wink.gif . So imagine the stability testing you posted you do, my CPU would do it at lower levels of offset than I have set now.

When you go for repeated testing / lengthy testing / various programs you will find what you thought was correct voltage point is not it. I have spent ~2-3 days perfecting my voltage point so I can run all these in a continuous run without turning off/rebooting etc. I arrived at ~1.350V VCORE (+137.5V offset LLC LVL 1).

Please:-

a) view my stability testing in the post (~38hrs run), do the same tests for same length and then tell me if what you set for a 15min run of RB and xx runs of CB works still for you wink.gif .

b) view my thread in sig on what went on with my testing for 3.7GHz, 3.8GHz, 3.9GHz and how I arrived where I am on voltage.

You will also see in my thread how finalheaven had thought he nailed 3.9GHz, on repeated testing he found he needed to increase VCORE far greater than his initial testing.
gupsterg is online now  
Sponsored Links
Advertisement
 
post #3884 of 42763 (permalink) Old 03-22-2017, 11:45 AM
Always Learning
 
finalheaven's Avatar
 
Join Date: May 2013
Posts: 1,118
Rep: 76 (Unique: 39)
Quote:
Originally Posted by bluej511 View Post

Voltage on DMM is 1.238 for 3.8ghz, but he has a 1700 and i got a 1700x so slightly different binning. I haven't run a full stability test yet but been gaming on it all day without issues so far, and thats with ram at 2933hz as well.

The chip was stable on my G5 before it bricked and that was at 1.245 so pretty close in voltages. Was stable 12 passes of cinebench and 15mins of real bench, then 15mins of cpuz and gpuz stress test at the same time.

When you run RealBench (for at least 30 mins though), can you check if HWiNFO64 reports any errors? It states errors on the way bottom of the summary list if you didn't hide that option. I realized that the errors disappear if you feed enough voltages and the more errors there are, the more likely you'll crash soon.
finalheaven is offline  
post #3885 of 42763 (permalink) Old 03-22-2017, 11:52 AM
New to Overclock.net
 
warreng5995's Avatar
 
Join Date: Mar 2017
Posts: 83
Rep: 1 (Unique: 1)
Quote:
Originally Posted by CwStrife View Post

If i remember right a few posts back one of these BIOS versions doesn't like odd ram timings? Someone correct me if i'm wrong. Try something like 18-16-16-16-38. You'll see in bios ram timings just follow those numbers

Ok, I'll try those ram timings, and see what happens.

"if you can't fix it... you just need a bigger hammer..."
warreng5995 is offline  
post #3886 of 42763 (permalink) Old 03-22-2017, 11:53 AM
 
Join Date: Mar 2016
Location: France/Armenia/US/Lebanon
Posts: 4,627
Quote:
Originally Posted by gupsterg View Post

Please take no offense by this wink.gif , it is merely a discussion similar to one's we have had in the past my online buddy smile.gif .

My stability testing is far greater than yours. For example I can run test of x264 for 10 loops at less (offset +106.25mV), 2 hrs RealBench, [email protected] 4-6hrs. Realbench on my CPU needs less of an offset for 3.8GHz than x264 wink.gif . So imagine the stabilty testing you posted you do my CPU would do it at lower levels.

When you go for repeated testing / lengthy testing / various programs you will find what you thought was correct voltage point is not it. I have spent ~2-3 days perfecting my voltage point so I can run all these in a continuous run without turning off/rebooting etc. I arrived at ~1.350V VCORE (+137.5V offset LLC LVL 1).

Please:-

a) view my stability testing in the post (~38hrs run), do the same tests for same length and then tell me if what you set for a 15min run of RB and xx runs of CB works still for you wink.gif .

b) view my thread in sig on what went on with my testing for 3.7GHz, 3.8GHz, 3.9GHz and how I arrived where I am on voltage.

You will also see in my thread how finalheaven had thought he nailed 3.9GHz, on repeated testing he found he needed to increase VCORE far greater than his initial testing.

Oh none taken i was just curious haha.

I saw your post yea. Mine is mostly for gaming until i get fiber next month then i can start editing and posting youtube videos. Then we'll see if its stable or not. My first 4690k was a beauty golden chip that did [email protected] with an hr real bench no problems, went 6months without a single crash. Second one was [email protected] and again was solid for around 6months benched the same way.

I may even try to get lower voltages for this at 3.8 but no sure it will go lower. Both my 4690k have never crashed, and no joke but i can go 8-10hrs of gaming solid without a problem lol. The second 4690k crashed at 4.3 and 1.20 so i added 10mv then it was rock solid.

I don't fold or render yet but when i do we shall see. I usually start my OCs about the middle of what ive seen most samples and either go up or down with voltage or just up in multiplier. I may try an hr of real bench but honestly over benching is also ridiculously bad for the chip believe it or not. 100% usage 24/7 doesnt do it any favors lol.
Quote:
Originally Posted by finalheaven View Post

When you run RealBench (for at least 30 mins though), can you check if HWiNFO64 reports any errors? It states errors on the way bottom of the summary list if you didn't hide that option. I realized that the errors disappear if you feed enough voltages and the more errors there are, the more likely you'll crash soon.

You mean WHEA errors? Oh i always check for those before anything else.
bluej511 is offline  
post #3887 of 42763 (permalink) Old 03-22-2017, 11:55 AM
Hardware Rep
 
Gary-ASUS's Avatar
 
Join Date: Jan 2017
Posts: 5
Rep: 0
Quote:
Originally Posted by CwStrife View Post

I'm simply saying it's bull**** what was said here by Raja about all these problems. Shoudln't be this way on a released board.
What Raja stated is factual about the current state of the platform. It is immature at this point in a lot of areas, especially on the memory side. That said, AMD is working on the AGESA code updates and improvements will come, but do not expect miracles overnight. As for me, running completely stable on EFI 1001 with a 1700 at 3.9, memory (2x8GB) at 3200, (4x8GB at 2667), basically auto settings with memory timings tightened up a little. I have three CPUs at the moment, each one behaves totally different with the same settings and components. I have more coming for additional comparison testing and will be working on some basic tips and tricks for air/water overclocking over the next couple of weeks.


ASUS Technical Marketing
Gary-ASUS is offline  
post #3888 of 42763 (permalink) Old 03-22-2017, 11:56 AM
Always Learning
 
finalheaven's Avatar
 
Join Date: May 2013
Posts: 1,118
Rep: 76 (Unique: 39)
Quote:
Originally Posted by bluej511 View Post

You mean WHEA errors? Oh i always check for those before anything else.

You don't get any errors even if you run it for 30 mins? Also whats your bios voltage for CPU set at?
finalheaven is offline  
post #3889 of 42763 (permalink) Old 03-22-2017, 11:57 AM
New to Overclock.net
 
Typhaeon's Avatar
 
Join Date: Mar 2017
Posts: 16
Rep: 3 (Unique: 1)
OK, so with 1001, my Fury Nitro is back to 16x mode in the PCIE X16 slot (according to the BIOS), but still won't output anything above 90hz in Low Freesync mode. High Freesync mode still doesn't work, and I can't select 100hz or higher from the Display Adapter Properties panel - still get No Signal from the DP connection.

DVI-D with Pixel Clock Patcher & HDMI have produced no improvements. I have no other card in the PCIE 3.0 lane, so what the hell?

Does anyone have any insight whatsoever on what setting I can even adjust to ~potentially~ get [email protected] output properly from the GPU? Is the card not getting enough power for some reason? Both VGA power connectors are hooked up properly, and I'd be seeing a lot more inconsistent behavior if it were an issue of my 1600W Platinum-rated eVGA PSU not working properly.

Ryzen 1800X, G.SKILL TridentZ F4-3200C14Q-64GTZ (4x16) running at DDR4-2667.
Quote:
Originally Posted by Typhaeon View Post

I posted a while back in this thread that I couldn't get my R9 Fury Nitro to output anything past [email protected] on this board. I wonder if anybody can give me a hand if they've had GPU issues they've solved via BIOS tweaks, because this is the last component I have to upgrade, and I don't want similar results with a more powerful/expensive GPU.

If I use my monitor's Low Freesync mode (Eizo Foris FS2735 - it has two: 35-90hz is Low; the 91-144hz High mode doesn't even work), I can get the card to render up to 90hz, but everything above just fails to output a signal over DisplayPort.

The GPU is the only PCIE 3.0 card in the system, installed in the 16x slot, and yet runs at 8x Native according to the BIOS. I have a Samsung 960 Pro in the M.2 Slot 3, and an Essence STX II in the PCIE 1x 3 slot. No other cards installed.

I've confirmed that this is not the fault of the OS (reimaged), the Crimson driver version (tried multiple), the BIOS switch on the card, the DisplayPort cable used, the DisplayPort input mode on the monitor (can switch between Low/High/DP 1.1/DP 1.2 in the OSD; switching off Freesync mode does nothing) or anything else I can think of offhand.

My PCIE settings in the BIOS are Auto - I've tried setting the M.2 slot to GEN 2, and forcing GEN 3 on the x16 slot, but that doesn't seem to do anything. I'm on 0902.

Anybody have similar experiences with a single GPU not going past x8 mode?
Typhaeon is offline  
post #3890 of 42763 (permalink) Old 03-22-2017, 12:01 PM
Meddling user
 
gupsterg's Avatar
 
Join Date: Jan 2015
Location: Lurking over a keyboard
Posts: 7,076
Rep: 768 (Unique: 351)
@bluej511

I'm glad we had the discussion wink.gif .

I was the same with my i5 4690K as Ryzen. The i5 4690K in my sig used the 4.9GHz OC 24/7, I set it ~1yr+ ago with same relentless stability testing. I had 0 issues, I posted similar data as for Ryzen in Devil's Canyon owners club.

I have a Q6600, owned since launch, again same relentless stability testing. Only after 1yr use did need like some silly little VCORE increase. Even in 2017 it can do the threads for a GPU to [email protected] for ~175hrs in one run wink.gif .

gupsterg is online now  
Reply

Tags
Bios5 , profiles to usb

Quick Reply
Message:
Options

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



Currently Active Users Viewing This Thread: 32 (6 members and 26 guests)
Kildar , mvmiller12 , RossiOCUK , SebaDimmu , xxela , yswai1986
Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off