Overclock.net banner
4881 - 4900 of 4995 Posts

·
Registered
Joined
·
5 Posts
This is the purpose of the Ring Down Bin feature. When Ring Down Bin is enabled, the CPU cache is always kept 300 MHz below the CPU core speed. If the core slows down, the cache will automatically slow down too. Leaving Ring Down Bin enabled can solve a lot of stability issues.

When Ring Down Bin is not enabled, the core and cache can end up running at the same speed. You will need a huge amount of extra voltage to get this stable.

The Intel CPU package temperature includes data from all of the individual cores. If one of the cores hits 75°C then the reported package temperature should be at least 75°C. If the core temperature is ever reported higher than the package temperature, that is a problem with the monitoring software.
Just out of curiosity: is it okay to have Ring Down Bin disabled when running fixed frequency on all cores and fixed uncore (assuming you keep at least 300 MHz between cores and uncore)? For example I've got 5.0 all core/4.7 uncore and 5.2 all core/4.9 uncore running stable on my 10900K with Ring Down Bin disabled, and I was under the impression that the 300 MHz gap between the cores and uncore would stay at all times since I've got everything running in fixed mode, but I wasn't sure. Should I enable Ring Down Bin just to be safe?
 

·
Registered
Joined
·
3,813 Posts
In fixed mode it doesn't do anything. It's more for overclocking in dynamic mode and EIST, SpeedShift and SpeedStep left enabled.
 
  • Rep+
Reactions: bluebaron

·
Registered
Joined
·
667 Posts

·
Gaming Overclocker
Joined
·
2,048 Posts
Yup that one. If you select a mount option (custom waterblock) it gives you washers to space the backplate from the motherboard to give it more mounting pressure / space for when there's no IHS but I didn't need them with the EK Supremacy mount.
... Cut ...
Hi, I'm quoting you again as I've only managed to have mine delidded and direct die installed yesterday and I may have some issues.
So I tried to install the frame like default using the square thin socket washer underneath the frame and there was no proper contact.
So I removed the 4 EK Velocity mount plastic washers from the top of the motherboard and placed them on the back (between the big rubber and the retention backplate.
And this way there was enough contact Imho (even with the four thumb nuts easily tighten and with little to zero tightening force).

But I do have this issues:
  • My 4200 ram oc bios profile won't work anymore, every time is giving me failed memory overclock, at 4400 cl 17 it works though with all the rest of the timings in Auto (not 100% stable as it then crashed while playing Hunt Showdown but maybe it wasn't tested enough since I don't use the 4400 ram freq).
  • The cpu temp (core max, package max) goes up to 88°C in Cinebench R20 (xmp ram, 5.2 fixed, 1.38v vcore) and also I'm noticing a 9° between the cores.
Even considering that yesterday evening was so hot as I was having 30C tamb, I still suspect something needs to be checked, adjusted (like removing the square socket washer to gain a little more pressure to the cpu pins and or something like I don't know, the old Supremacy Precise Mount for Ivy in order to get more tightening force on the Velocity block) but I'd like to hear your advice first.
Cheers
 

·
Registered
Joined
·
292 Posts
He has an SP104, that is like best Case Chip better as my one, such a chip scale also awesome in higher frequency,
my scaling end´s at 5,4ghz, 5,5ghz need´s more voltage as the step´s before.
SP100+ chip´s are that what everyone want to have and nobody found^^.
It give´s enough people me also who would pay for a SP100+ Chip but nobody sell such chip´s,
they are so rare like moon rooks on earth.
View attachment 2518222
Good stuff.
My SP74 and SP88 failed in asus realbech at 1.25v load voltage 5ghz.
The SP88 needed 1.26v and the SP74 the same.
So for me at 5ghz the same vcore was needed.

The olny changes i notice on the chips was the stock vcore.
My first SP63 was running very high stock vcore.
My SP74 less vcore and my SP88 even less.
SP63 was 10c hotter then the Sp88.
 

·
Registered
Joined
·
3,813 Posts
Hi, I'm quoting you again as I've only managed to have mine delidded and direct die installed yesterday and I may have some issues.
So I tried to install the frame like default using the square thin socket washer underneath the frame and there was no proper contact.
So I removed the 4 EK Velocity mount plastic washers from the top of the motherboard and placed them on the back (between the big rubber and the retention backplate.
And this way there was enough contact Imho (even with the four thumb nuts easily tighten and with little to zero tightening force).

But I do have this issues:
  • My 4200 ram oc bios profile won't work anymore, every time is giving me failed memory overclock, at 4400 cl 17 it works though with all the rest of the timings in Auto (not 100% stable as it then crashed while playing Hunt Showdown but maybe it wasn't tested enough since I don't use the 4400 ram freq).
  • The cpu temp (core max, package max) goes up to 88°C in Cinebench R20 (xmp ram, 5.2 fixed, 1.38v vcore) and also I'm noticing a 9° between the cores.
Even considering that yesterday evening was so hot as I was having 30C tamb, I still suspect something needs to be checked, adjusted (like removing the square socket washer to gain a little more pressure to the cpu pins and or something like I don't know, the old Supremacy Precise Mount for Ivy in order to get more tightening force on the Velocity block) but I'd like to hear your advice first.
Cheers
9c delta is quite a lot for direct die and the temps sound kinda high for 1.38v. What paste did you use?

Also. I don't really know in what way the Velocity mount is different from the Supremacy (S1151) mount my block uses. The RAM not working properly could also indicate a bad mount / pin contact in the socket.
 

·
Gaming Overclocker
Joined
·
2,048 Posts
9c delta is quite a lot for direct die and the temps sound kinda high for 1.38v. What paste did you use?

Also. I don't really know in what way the Velocity mount is different from the Supremacy (S1151) mount my block uses. The RAM not working properly could also indicate a bad mount / pin contact in the socket.
They don't differ (Supremacy /Velocity)
TG Conductonaut used
Also I tried to follow this installation video (could be wrong given that was done for the 9900k?)


I will check everything back again this evening when I get back home.
 

·
Registered
Joined
·
3,813 Posts
They don't differ (Supremacy /Velocity)
TG Conductonaut used
Also I tried to follow this installation video (could be wrong given that was done for the 9900k?)


I will check everything back again this evening when I get back home.
Same video I used. It's fine. I used a different "paste", Coollaboratory Liquid Ultra which I strongly prefer over Conductonaut. And my block is mirror lapped to 3000 grit. Stock springs, stock thumb nuts.

My core delta's are usually around 2-4c max.
At those same frequencies and voltages I get around 74-76c in CB20 and 89-92c in Prime95 AVX Small (worst case load possible well over 340w).

Now at 5.1/4.7 1.270v I get 71c CB20, low 60's in games and 86-89c in Prime95 Small around 330-335w.
 

·
Registered
Joined
·
307 Posts
Good stuff.
My SP74 and SP88 failed in asus realbech at 1.25v load voltage 5ghz.
The SP88 needed 1.26v and the SP74 the same.
So for me at 5ghz the same vcore was needed.

The olny changes i notice on the chips was the stock vcore.
My first SP63 was running very high stock vcore.
My SP74 less vcore and my SP88 even less.
SP63 was 10c hotter then the Sp88.
Your forget one thing you have a other board.
The measurement of the vcore is die sense, if your board use socket sense you have higher shown value by same vcore.
Difference was 90-100mV or so.
The second thing is, is the vrm very strong and you drive a good LLC(100mV+ dropp) it´s possible to drive lower vcore stable.
It´s also easier to be stable with more power consumption.

I think your SP88 will be on a Apex/Hero under 1,15V load with wakü.
So don´t think that your chip is bad.
 

·
Registered
Joined
·
48 Posts
Anything for XIII? Not saying there is anything wrong, just wondering. :)
Hey i already uploaded :)

 

·
Registered
Joined
·
434 Posts

·
Registered
Joined
·
434 Posts
Hi @safedisk ,

I installed 2301 and so far so good...
Last 2201 I had some problems to stabilize my OC like I had tuned the 2103.
Let me ask you one thing... Do you have any new "working tool" for changing some bios parameter on the fly?

EDITED:
In 2301 BIOS PTT is enabled by default !
 

·
Optimal Pessimist
Joined
·
3,175 Posts
Hi @safedisk ,

I installed 2301 and so far so good...
Last 2201 I had some problems to stabilize my OC like I had tuned the 2103.
Let me ask you one thing... Do you have any new "working tool" for changing some bios parameter on the fly?

EDITED:
In 2301 BIOS PTT is enabled by default !
I wonder why... :)
 
  • Rep+
Reactions: RobertoSampaio

·
Registered
Joined
·
434 Posts
I'll give up to run 56x with these new BIOS...
With 2103 it runs fine, but 2201 and 2301 it's impossible...
The good news is I need only 1460mv to run 55x light loads...
I will tune all again for 55x.
And buy a cryo cooler... LOL
 

·
Optimal Pessimist
Joined
·
3,175 Posts
I'll give up to run 56x with these new BIOS...
With 2103 it runs fine, but 2201 and 2301 it's impossible...
The good news is I need only 1460mv to run 55x light loads...
I will tune all again for 55x.
And buy a cryo cooler... LOL
Seems like beyond 2103, the z490 BIOS may be better for 11th gen, but worse for 10th gen.
 
4881 - 4900 of 4995 Posts
Top