9900k Direct Die Overclock Help Needed - Page 4 - Overclock.net - An Overclocking Community

Forum Jump: 

9900k Direct Die Overclock Help Needed

Reply
 
Thread Tools
post #31 of 34 (permalink) Old 08-20-2019, 03:14 PM
New to Overclock.net
 
jasper34's Avatar
 
Join Date: Oct 2010
Posts: 12
Rep: 0
I don't delid, but had no trouble getting to 5.2GHz on all 16 threads at 1.268v, I game at 45C Just a 280mm AIO and Thermal Grizzly Kryonaut paste (the best non-conductive made). It's virtually silent and stable for gaming. I no longer punish my CPU's with unrealistic loads. If it never crashes and is 100% stable playing my most demanding game at 2160P, I'm happy.
jasper34 is offline  
Sponsored Links
Advertisement
 
post #32 of 34 (permalink) Old 08-22-2019, 06:19 AM
New to Overclock.net
 
Socks keep you warm's Avatar
 
Join Date: Feb 2011
Location: Melbourne, Australia
Posts: 3,104
Rep: 131 (Unique: 116)
Quote: Originally Posted by jasper34 View Post
I don't delid, but had no trouble getting to 5.2GHz on all 16 threads at 1.268v, I game at 45C Just a 280mm AIO and Thermal Grizzly Kryonaut paste (the best non-conductive made). It's virtually silent and stable for gaming. I no longer punish my CPU's with unrealistic loads. If it never crashes and is 100% stable playing my most demanding game at 2160P, I'm happy.
Wow you must have the best 9900K on the market. You should load HWInfo and let us know how many Windows errors occur.

I like to have fun.
https://hwbot.org/user/sockskeepyouwarm/
https://www.trueachievements.com/gamer/II+TH3PRO+II
16 Threads is Good
(10 items)
CPU
Intel Core I9 9900K 4.8Ghz
Motherboard
Gigabyte Aorus Elite
GPU
Aorus 2080Ti Xtreme Waterforce WB 11GB
RAM
Trident Z
Hard Drive
Samsung 250 SSD
Power Supply
Cooler Master Silent Pro Hybrid
Cooling
Watercooling Custom
Case
900D
Operating System
Windows 10 x64
Monitor
28" ACER 4K Monitor
▲ hide details ▲


Socks keep you warm is offline  
post #33 of 34 (permalink) Old 08-22-2019, 01:22 PM - Thread Starter
New to Overclock.net
 
Join Date: May 2019
Posts: 24
Rep: 0
Quote: Originally Posted by Falkentyne View Post
You are correct in some of your points. Extreme LLC should simply not be used at those voltages. Extreme Loadline calibration is 0.2 mOhms and Turbo is 0.4 mOhms and High is 0.8 mOhms. In "general", if you can get away with it, MORE vdroop is better for you, not less. But on fixed voltages, since AC Loadline won't help you (see below for full explanation), you need a balance between bios voltage and vdroop.

First of all, the below is based on "Auto" vcore being used, NO VRM loadline calibration used (Intel specified VRM loadline used-1.6 mOhms), and x49/x50 core multiplier, with AC Loadline limited to 1.6 mOhms.

The entire problem with the "1.520v" value is, the way Intel designs these processors (based on *AUTO* Vcore and Serial VID), it will *never* see a 1.52v idle voltage, with zero amps anyway. If you use the maximum AC Loadline of 1.6 mOhms (I found that at 4.9 and 5 ghz default VIDS, there is no difference in load VR VOUT between an ACLL of 1.6 mOhms and ACLL of 1.3 mOhms except at light loads), and use Auto vcore (x50 and x49 multiplier), your idle VR VOUT would be about 1.40v, not 1.520v. That's because the AC Loadline value is what is responsible for boosting the "Idle" voltage target and the "load" voltage target to the VRM. The "base" VID at x49 and x50 multiplier is about 1.175v to 1.250v, depending on the silicon quality of your processor. (you can get this value manually, on fixed voltage, by setting AC Loadline and DC Loadline to "1" (0.01 mOhms)). If you were using pure auto vcore at this point, and NO loadline calibration (LLC Standard/Normal), this is the Voltage "target" that would get sent to the VRM, and used by the voltage controller before vdroop.

An AC Loadline higher than 1 will raise the default idle voltage up. An AC Loadline of 1.60 (160) will put the idle VR VOUT somewhere around 1.40v. An AC Loadline of 2.10 (maximum Intel allows for 4/6 core processors) will put the idle VR VOUT at about 1.45v.

DC loadline affects the VID *AFTER* the VRM already has the voltage ID target programmed--meaning the DC Loadline has no effect on VR VOUT at all. The DC Loadline mOhms resistance value affects the VID in the exact same way that "VRM Loadline" (Loadline calibration) affects VR VOUT. However Intel specifies that DC Loadline is only used for power reporting, not for the CPU power supply. CPU Package Power is the result of VID * Amps, after DC Loadline affects the VID.

(Vdroop in millivolts is equal to R * I, based on Ohm's law...R=resistance (milliohms), I=Current (Amps).

At full load, the AC Loadline will boost the CPU VRM voltage target even higher than it does at idle. This VRM target cannot exceed 1.520v, but *will* be up to 1.520v on auto voltages. (easy way to see this safely on your own? Set AC Loadline to 160 and DC Loadline to 1, and watch the VID in windows. DC Loadline=1 prevents DC Loadline from dropping the VID for power reporting).

Now, this is where NOT having loadline calibration will help your processor's longevity. That 1.520v (1520mv) will be dropped by the VRM loadline value (in this case default is 1.6 mOhms), depending on resistance, so anywhere between 1.220v to 1.30v at full load, depending on amps.

Intel uses AC Loadline to help 'mitigate excessive vdroop' without needing high idle voltages, by raising the VRM voltage target at full load.

Dropping AC loadline below 1.6 mOhms (160) will reduce the idle VR VOUT in windows. Load VR VOUT (heavy amps) will remain the same depending on if AC Loadline is lowered slightly or a lot, because again this is a factor of resistance. You will have to lower AC loadline quite a bit to have the "VRM target" at full load below 1.520v, which will then drop your load VR VOUT also.

These settings are intel's maximum limits.


Now to answer your question, the problem with using 1.52v set in BIOS with NO loadline calibration at all, is that when you use a fixed voltage, AC Loadline can no longer help you. Fixed vcore bypasses AC Loadline and programs the VRM manually with a new voltage target. So your VR VOUT is going to be extremely high--probably 1.49v. You are not going to like that and even if it "may" be safe if NO LLC is used, you would not want to see slow degradation anyway. It makes far more sense just to use 1.3 mOhms AC Loadline or 1.6 mOhms AC Loadline (130-160) and use pure Auto vcore, and NO LLC, and get a lower idle voltage. It just makes sense.

Note: SVID Offset being enabled disables all manual and offset voltage control and allows VID to exceed 1.520v, by up to 200mv.
This can be useful to people on subzero, but a neat trick, since AC Loadline can now boost VID VRM target higher than 1520mv, is to lower AC loadline by a lot (like, 80-110, depending on your CPU quality) and then enable SVID offset. The lower AC Loadline will reduce your idle vcore by a lot, and SVID Offset enabled will allow VID VRM to exceed 1.520, but due to the lower AC Loadline, it will drop back down to 1520mv, then vdroop (VRM loadline) will drop it much lower on load.

Not sure if this makes sense. Your best bet is to actually test these things.

WARNING: NEVER enable SVID OFFSET ON AUTO VCORE if AC LOADLINE IS 1.6 MOHMS OR HIGHER! YOUR VR VOUT WILL BE TOO HIGH FOR SAFE AMPS LIMITS.

Okay so a little confusing. I dropped the LLC to auto. Dropped the VCORE to auto. Manually set the AC to 1.6 and DC to 1. It does not hit 1.52v anywhere and it does allow me to clock up to 5.2ghz with decent voltages, BUT, benchmark scores go DOWN, and I mean WAYYYYY Down. Not sure what math I am supposed to be calculating for correct voltages when messing with AC and DC and where to really move the testing
poopsLIVE is offline  
Sponsored Links
Advertisement
 
post #34 of 34 (permalink) Old 08-22-2019, 02:10 PM
New to Overclock.net
 
Falkentyne's Avatar
 
Join Date: Dec 2007
Location: Riverside
Posts: 5,987
Rep: 383 (Unique: 274)
Quote: Originally Posted by poopsLIVE View Post
Okay so a little confusing. I dropped the LLC to auto. Dropped the VCORE to auto. Manually set the AC to 1.6 and DC to 1. It does not hit 1.52v anywhere and it does allow me to clock up to 5.2ghz with decent voltages, BUT, benchmark scores go DOWN, and I mean WAYYYYY Down. Not sure what math I am supposed to be calculating for correct voltages when messing with AC and DC and where to really move the testing
here.
Attached Thumbnails
Click image for larger version

Name:	VID_VRVOUT_LOAD_DC_Loadline_1.jpg
Views:	26
Size:	871.5 KB
ID:	290858  

Click image for larger version

Name:	svid_Offset_DCLoadline_1.jpg
Views:	18
Size:	812.6 KB
ID:	290860  


[email protected] ghz, RX Vega 64, 32GB DDR4, Gigabyte Aorus Master, Seasonic Platinum 1000W, Corsair 760T
Alt: MSI GT73VR Throttlebook with 7820HK @ 4.7 ghz, GTX 1070 MXM TDP mod to 230W, 32 GB RAM

Last edited by Falkentyne; 08-22-2019 at 02:13 PM.
Falkentyne is offline  
Reply

Tags
9900k , direct die , Overclock , temperature , water cooled

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: 1 (0 members and 1 guests)
 
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