Overclock.net banner

1 - 20 of 56 Posts

·
Registered
Joined
·
1,124 Posts
Discussion Starter #1 (Edited)
As the title suggests skew control seems important for overclocking memory if you want to push it a little

I spent the day testing different settings to see just how much difference the skew settings make and it seems it won't even boot with many of them

I can see now that it would be easy to think you were unlucky in the silicon lottery or your IMC isn't any good just because your skews aren't set correctly

There have been a couple of people I've seen playing with these settings, notably @OLDFATSHEEP and @Falkentyne off the top of my head

The 2 tests I'm using for stability are:

TestMem5 [email protected]
GSAT

Both of which there is a link to in my sig if you want to try this for yourself
I only did 2 minutes each initially as there was a lot of testing to do and the aim was to narrow down the possible final contestants
Now that I'm at this point it may be possible for me to stabilize a 4533MHz ram overclock with 2x16GB sticks, we'll see :)
My hardware is in my sig

So my first step was to find a setting I knew to be not quite stable
5.2/49 1.32v LLC7
4533-18-19-19-36
1.45vdimm
1.35vccio
1.35vccsa

4533-17.jpg

Next it was just a matter of going through all the possible combinations that booted and the ones around them and running the tests
Here are the results

SkewControl.jpg

These results laid out in this manor make it pretty clear what my optimum options are for this hardware
You can clearly see that GSAT has a much smaller range of acceptance

Open Office is a free download if anyone would like to try this at home and add your results to this thread for future reference
Be sure to list your parts in your sig or in your post, but it is worth doing a rig builder in your sig anyway
I look forward to seeing your results :)

You may find that you can boot a much higher ram overclock once you have done this

Edit:
After undertaking this testing, and setting my skews to 80-48-0 (I may try 80-48-34) I am now able increase my stable overclock from 4400-17-17-17-34 to 4500-18-19-19-36.
4533 is outside the ability of my hardware as far as I can tell. Still not a bad increase for 3600c16 sticks :)

SkewSettings.jpg

My .ods file is zipped below ready for downloading so you can just edit it for your own tests

 

Attachments

·
ٴٴٴ╲⎝⧹˙͜>˙⧸⎠╱
Joined
·
6,276 Posts
As the title suggests skew control seems important for overclocking memory if you want to push it a little

I spent the day testing different settings to see just how much difference the skew settings make and it seems it won't even boot with many of them

There have been a couple of people I've seen playing with these settings, notably @OLDFATSHEEP and @Falkentyne off the top of my head

The 2 tests I'm using for stability are:

TestMem5 [email protected]
GSAT

Both of which there is a link to in my sig if you want to try this for yourself
I only did 2 minutes each initially as there was a lot of testing to do and the aim was to narrow down the possible final contestants
Now that I'm at this point it may be possible for me to stabilize a 4533MHz ram overclock with 2x16GB sticks, we'll see :)
My hardware is in my sig

So my first step was to find a setting I knew to be not quite stable
5.2/49 1.32v LLC7
4533-18-19-19-36
1.45vdimm
1.35vccio
1.35vccsa

View attachment 2464126

Next it was just a matter of going through all the possible combinations that booted and the ones around them and running the tests
Here are the results

View attachment 2464128

These results laid out in this manor make it pretty clear what my optimum options are for this hardware
You can clearly see that GSAT has a much smaller range of acceptance

Open Office is a free download if anyone would like to try this at home and add your results to this thread for future reference
Be sure to list your parts in your sig or in your post, but it is worth doing a rig builder in your sig anyway
I look forward to seeing your results :)

You may find that you can boot a much higher ram overclock once you have done this

My .ods file is zipped below ready for downloading so you can just edit it for your own tests

I'm not really understanding what skew control is supposed to do here


the links in your github loops to several forum that ends up in a 404
 

·
Registered
Joined
·
1,124 Posts
Discussion Starter #5
I'm not really understanding what skew control is supposed to do here


the links in your github loops to several forum that ends up in a 404
I see what you mean. For something so important there is little information on it.
I don't know the technical side of things so am unable to explain all the ins and outs but I've seen some information from @OLDFATSHEEP explaining it. Maybe he can chime in :)

I just started testing in an effort to stabilize my own higher frequency overclocks and was very surprised to discover how narrow the band of usable numbers is :)
It makes me wonder how many people blame their IMC, when it's not their problem at all.
 

·
ٴٴٴ╲⎝⧹˙͜>˙⧸⎠╱
Joined
·
6,276 Posts
I see what you mean. For something so important there is little information on it.
I don't know the technical side of things so am unable to explain all the ins and outs but I've seen some information from @OLDFATSHEEP explaining it. Maybe he can chime in :)

I just started testing in an effort to stabilize my own higher frequency overclocks and was very surprised to discover how narrow the band of usable numbers is :)
It makes me wonder how many people blame their IMC, when it's not their problem at all.

doing this on an apex with maximus tweak enabled


not sure if I should bother with trying out skew settings to improve?

1.468v dram
1.232 sa/io
 

·
Just hangin' out
Joined
·
1,123 Posts
This is an Asus only BIOS option, I presume?
 

·
Registered
Joined
·
1,124 Posts
Discussion Starter #8

·
ٴٴٴ╲⎝⧹˙͜>˙⧸⎠╱
Joined
·
6,276 Posts

·
Registered
Joined
·
547 Posts
doing this on an apex with maximus tweak enabled
not sure if I should bother with trying out skew settings to improve?

1.468v dram
1.232 sa/io
I doubt that is TM5 Extreme1 anta777 or GSAT stable for very long ;)
 

·
ٴٴٴ╲⎝⧹˙͜>˙⧸⎠╱
Joined
·
6,276 Posts
I doubt that is TM5 Extreme1 anta777 or GSAT stable for very long ;)
did anta777 for 10 passes actually

max temp on memory 46c
 

·
Registered
Joined
·
1,124 Posts
Discussion Starter #12
maximus tweak is only on maximus series

I am not aware of any skew control on the apex's bios @munternet
It's in the Maximus XII Apex's BIOS 0088 :)
I'm fairly sure I've seen Buildzoid messing with it on the Gigabyte Master also
 

·
Registered
Joined
·
529 Posts
I'm not really understanding what skew control is supposed to do here


the links in your github loops to several forum that ends up in a 404
The skew control (or "slew control" according to Samsung), basically used to match the signal of your ram sticks. ODT is on the ram die, and you have to match the impedance to get the best signal quality. A range of the ODT numbers should all work, based on the quality of your dies. On the other hand, the "rising" and "falling" mean the signal changes from "low" to "high" and from "high" to "low". You have to match the characteristics of the ram die.

Skew control usually won't show any obvious effects, unless you have reached the limit of your sticks. BIOS usually provides the common numbers for a kind of ram dies (like b-die or DJR), but as an OCer, you can tune the numbers according to your specific ram sticks.
 
  • Rep+
Reactions: Veii

·
ٴٴٴ╲⎝⧹˙͜>˙⧸⎠╱
Joined
·
6,276 Posts
It's in the Maximus XII Apex's BIOS 0088 :)
I'm fairly sure I've seen Buildzoid messing with it on the Gigabyte Master also
I have to check that again, been a while without messing with that.

Under what setting is it?
 

·
ٴٴٴ╲⎝⧹˙͜>˙⧸⎠╱
Joined
·
6,276 Posts
The skew control (or "slew control" according to Samsung), basically used to match the signal of your ram sticks. ODT is on the ram die, and you have to match the impedance to get the best signal quality. A range of the ODT numbers should all work, based on the quality of your dies. On the other hand, the "rising" and "falling" mean the signal changes from "low" to "high" and from "high" to "low". You have to match the characteristics of the ram die.

Skew control usually won't show any obvious effects, unless you have reached the limit of your sticks. BIOS usually provides the common numbers for a kind of ram dies (like b-die or DJR), but as an OCer, you can tune the numbers according to your specific ram sticks.
ahh ok so it's the procodt / termination ohm settings, got it, thought it was something else


do you have any recommended settings for dual rank b-die on apex?
 

·
Registered
Joined
·
529 Posts
ahh ok so it's the procodt / termination ohm settings, got it, thought it was something else


do you have any recommended settings for dual rank b-die on apex?
Similar but the ProcODT is the ODT on the CPU, ODTs on the ram dies are like WR-NOM-PARK or such. I use 80-48-48 for CHA and 80-48-34 for CHB. For the rising and falling slopes, especially "data" slope, b-die can use 8, and 0 as the offset.
 
  • Rep+
Reactions: Veii

·
ٴٴٴ╲⎝⧹˙͜>˙⧸⎠╱
Joined
·
6,276 Posts
Similar but the ProcODT is the ODT on the CPU, ODTs on the ram dies are like WR-NOM-PARK or such. I use 80-48-48 for CHA and 80-48-34 for CHB. For the rising and falling slopes, especially "data" slope, b-die can use 8, and 0 as the offset.
I'll see if I actually tweaked this option, what improvements do you predict will happen with this?
 

·
Registered
Joined
·
529 Posts
It all depends, like easier to POST or can bear much higher ram temp...
 
  • Rep+
Reactions: munternet

·
Registered
Joined
·
1,124 Posts
Discussion Starter #19
ahh ok so it's the procodt / termination ohm settings, got it, thought it was something else


do you have any recommended settings for dual rank b-die on apex?
I am using 80-48-0 on the hardware in my sig or I could try 80-48-34 but I didn't test my sticks separately like OLDFATSHEEP did. At least one of my sticks errors quickly in GSAT at 80-48-48
 
1 - 20 of 56 Posts
Top