Post your Cinebench R20+15+R11.5+2003 Scores **Don't Read The OP PLZ**

[Brekmister]

I so love this CPU for all the abuse I have given it these last 3 years

[kestonlamers]

 

http://i.imgur.com/TCMR8IN.png
http://i.imgur.com/FTejdvg.png

[KeltonDSMer]

I know it's not really the right place for this, but I already tried to start another thread and didn't get a response yesterday. You guys are the most likely fellas here at LTT to be able to help me! From my thread yesterday:

 

"Hey it's getting cold again so I wanted to push this little i3 6100.

 

I am on an Asrock Pro4s Z170M, bios version L2.23 (I think it is the last version to support bus increases on locked chips without the multiplier getting killed)

 

Super frustrating, I can increase voltage at stock speeds as much as I want to. I have been doing that to find my comfortable temp limits before the sensor stops reading after increasing bus speed. Basically, I run benchmarks at stock speeds but over-volted to get an idea of what temps will look like once I bump up clocks and the sensor becomes unreadable if that makes sense.

 

Problem is, once I touch the base clock the max voltage caps at ~+100mv or 1.20V, wtf? Has anyone experienced this before? No difference in CPUz's shown voltage from +100mv to even +250mv if the base clock is above 100MHz

 

Current/power limits are all disabled, as well as Cstates and anything that looks like it might interfere with an OC, but nothing seems to allow me to increase the voltage beyond 1.2V once the bus is past stock 100MHz. Again, at stock base clock, the voltage offset works as expected and only caps voltage once the base clock is increased past 100MHz.

 

Anyone know what is going on? Is there a different bios version anyone is using to successfully crank up voltage and base clock on a locked chip with an asrock z170?

 

Any help would be greatly appreciated!"

 

Weird, right? Since the FIVR is not integrated into the package like Haswell I figure this has to be an issue on the mobo side?

 

Here is the best I could do on R15 with the stock cooler @ 1.184v (like 4 months ago D:) :

 

Spoiler

 

I think the memory controller on the chip is less than stellar, I can crank the divider and pull back timings at stock cpu clocks but once that base starts going up I have a hard time even booting much above 2400MHz

[Jumper118]

48 minutes ago, KeltonDSMer said:

I know it's not really the right place for this, but I already tried to start another thread and didn't get a response yesterday. You guys are the most likely fellas here at LTT to be able to help me! From my thread yesterday:

 

"Hey it's getting cold again so I wanted to push this little i3 6100.

 

I am on an Asrock Pro4s Z170M, bios version L2.23 (I think it is the last version to support bus increases on locked chips without the multiplier getting killed)

 

Super frustrating, I can increase voltage at stock speeds as much as I want to. I have been doing that to find my comfortable temp limits before the sensor stops reading after increasing bus speed. Basically, I run benchmarks at stock speeds but over-volted to get an idea of what temps will look like once I bump up clocks and the sensor becomes unreadable if that makes sense.

 

Problem is, once I touch the base clock the max voltage caps at ~+100mv or 1.20V, wtf? Has anyone experienced this before? No difference in CPUz's shown voltage from +100mv to even +250mv if the base clock is above 100MHz

 

Current/power limits are all disabled, as well as Cstates and anything that looks like it might interfere with an OC, but nothing seems to allow me to increase the voltage beyond 1.2V once the bus is past stock 100MHz. Again, at stock base clock, the voltage offset works as expected and only caps voltage once the base clock is increased past 100MHz.

 

Anyone know what is going on? Is there a different bios version anyone is using to successfully crank up voltage and base clock on a locked chip with an asrock z170?

 

Any help would be greatly appreciated!"

 

Weird, right? Since the FIVR is not integrated into the package like Haswell I figure this has to be an issue on the mobo side?

 

Here is the best I could do on R15 with the stock cooler @ 1.184v (like 4 months ago D:) :

 

  Reveal hidden contents

 

I think the memory controller on the chip is less than stellar, I can crank the divider and pull back timings at stock cpu clocks but once that base starts going up I have a hard time even booting much above 2400MHz

that does sound like a bad bios

did you get it from here>

http://overclocking.guide/asrock-z170-non-k-overclocking-guide/

if you didnt try the one for your board from there. 

[KeltonDSMer]

2 hours ago, Jumper118 said:

that does sound like a bad bios

did you get it from here>

http://overclocking.guide/asrock-z170-non-k-overclocking-guide/

if you didnt try the one for your board from there. 

 

I got the bios from the "beta" section on Asrock's site. Unfortunately, the page you linked doesn't have a bios for the z170m pro4s and the comments on the regular pro4s page on that site have a couple people asking about the matx board and the only replies link to the same beta page I got my bios from

 

I don't have much thermal headroom left (stock cooler), but I'm a little bummed that it seems like I'll only be able to play with 1.184v

 

Thanks for the link though, seems like a really handy source!

 

 

[done12many2]

3 hours ago, Jumper118 said:

that does sound like a bad bios

did you get it from here>

http://overclocking.guide/asrock-z170-non-k-overclocking-guide/

if you didnt try the one for your board from there. 

 

Where's that 5960x score at? 

[Jumper118]

4 minutes ago, done12many2 said:

 

Where's that 5960x score at? 

its erm. well i noobing can only get 1904 which is too slow. 

7 minutes ago, KeltonDSMer said:

 

I got the bios from the "beta" section on Asrock's site. Unfortunately, the page you linked doesn't have a bios for the z170m pro4s and the comments on the regular pro4s page on that site have a couple people asking about the matx board and the only replies link to the same beta page I got my bios from

 

I don't have much thermal headroom left (stock cooler), but I'm a little bummed that it seems like I'll only be able to play with 1.184v

 

Thanks for the link though, seems like a really handy source!

 

 

o yes i must have missed the M on the end of your earlier post  It is a strange problem and if there are any older or different bios you can find from asrock then you could try those. You could even contact asrock and ask them. 

 

Yes that is an excellent source. i have been linking it since bclk day 0 but nobody else seems to know about it. it has a separate guide for each manufacturer of motherboard all in the same place which is very helpful. 

[KeltonDSMer]

1 minute ago, Jumper118 said:

its erm. well i noobing can only get 1904 which is too slow. 

o yes i must have missed the M on the end of your earlier post  It is a strange problem and if there are any older or different bios you can find from asrock then you could try those. You could even contact asrock and ask them. 

 

Yes that is an excellent source. i have been linking it since bclk day 0 but nobody else seems to know about it. it has a separate guide for each manufacturer of motherboard all in the same place which is very helpful. 

 

Yeah I'll try another earlier bios, I just always get a little nervous flashing it so i figured I'd ask first before I start trying older versions willy-nilly. Thanks, I'll let you know how it goes!

[done12many2]

1 minute ago, Jumper118 said:

its erm. well i noobing can only get 1904 which is too slow. 

 

That's actually pretty damn good.  You aren't going to get a whole lot higher, maybe +/- 1950 with any form of ambient cooling.  Either way, I say you crank up the voltage and submit what you've got.  

[Jumper118]

Just now, KeltonDSMer said:

 

Yeah I'll try another earlier bios, I just always get a little nervous flashing it so i figured I'd ask first before I start trying older versions willy-nilly. Thanks, I'll let you know how it goes!

bios flashing is pretty safe as long as you dont turn it off half way through  

Just now, done12many2 said:

 

That's actually pretty damn good.  You aren't going to get a whole lot higher, maybe +/- 1950 with any form of ambient cooling.  Either way, I say you crank up the voltage and submit what you've got.  

its OK, but i need 2000 on ambient. it is possible but it just requires my to tweak it which i dont really have time to do atm but i will try anyway. i am currently running 1x7970 and 2x 7950 in cf with it and putting out some pretty beastly scores.  

[done12many2]

1 minute ago, Jumper118 said:

its OK, but i need 2000 on ambient.

 

I'd really like to see that happen, but I wouldn't count on it.

[Jumper118]

4 minutes ago, done12many2 said:

 

I'd really like to see that happen, but I wouldn't count on it.

its happening. i got 5.6ghz out of an i5 2500K under water  i know this is a whole different beast but it has to be done. i should only need 4.85ghz? 

[done12many2]

Just now, Jumper118 said:

its happening. i got 5.6ghz out of an i5 2500K under water  i know this is a whole different beast but it has to be done. i should only need 4.85ghz? 

 

Haha.  Much easier said than done bud.  I've run many chips at or above 5 + GHz.  Hell, I've even run my 5960x just over 5 GHz, but not with a full load on all 8 cores.  Sub ambient required for that.  I think you'll find that it falls flat on its face right around 4.8 with all cores under load, even with the best water cooling in the world.  My water doesn't even heat up at all, I just can't pull the heat off the package fast enough.

 

4.85 GHz will definitely break 2000cb.  I hope you pull it off!

[Jumper118]

2 minutes ago, done12many2 said:

 

Haha.  Much easier said than done bud.  I've run many chips at or above 5 + GHz.  Hell, I've even run my 5960x just over 5 GHz, but not with a full load on all 8 cores.  Sub ambient required for that.  I think you'll find that it falls flat on its face right around 4.8 with all cores under load, even with the best water cooling in the world.  My water doesn't even heat up at all, I just can't pull the heat off the package fast enough.

 

4.85 GHz will definitely break 2000cb.  I hope you pull it off!

well if it does fall flat on its face i may be forced to freeze it with the phase cooler. at -50c on 1.6v i think it would certainly do 5ghz+  

 

it would also be nice to give my phase cooler a proper work out for once as well. it normally gets c2d and pentium with the odd i5 and i3. i think the 3930K and fx8320 were the hardest on it. 

[done12many2]

2 minutes ago, Jumper118 said:

well if it does fall flat on its face i may be forced to freeze it with the phase cooler. at -50c on 1.6v i think it would certainly do 5ghz+  

 

it would also be nice to give my phase cooler a proper work out for once as well. it normally gets c2d and pentium with the odd i5 and i3. i think the 3930K and fx8320 were the hardest on it. 

 

That would be fun to hear about.  The 5960x puts out a metric shit ton of heat.  I'd imagine that at 1.6v with a full load, you'd be putting well over 400w through that chip.

[iHardware Shelden]

[Jumper118]

2 minutes ago, done12many2 said:

 

That would be fun to hear about.  The 5960x puts out a metric shit ton of heat.  I'd imagine that at 1.6v with a full load, you'd be putting well over 400w through that chip.

my phase is rated for -50c at 400w load so thats perfect  the fx8320 with 1.7v going through it was managing to get it to -48c at some point  

[done12many2]

8 minutes ago, Jumper118 said:

my phase is rated for -50c at 400w load so thats perfect  the fx8320 with 1.7v going through it was managing to get it to -48c at some point  

 

Straight from the Asus team.  Grant it, you won't be running Prime95, but you will be over 5 GHz compared to their observed 4.4 GHz draw.

 

 

[Jumper118]

5 minutes ago, done12many2 said:

 

Straight from the Asus team.  Grant it, you won't be running Prime95, but you will be over 5 GHz compared to their observed 4.4 GHz draw.

 

 

 

lol. at what voltage? is that cpu only? i know a 22core with an unlocked multi on ln2 can literally blow up any vrm  

[done12many2]

8 minutes ago, Jumper118 said:

lol. at what voltage? is that cpu only? i know a 22core with an unlocked multi on ln2 can literally blow up any vrm  

 

4.4 at 1.3v.  Do you see why I say you'll be well over 400w at 5.0 with 1.6v?  Actually, if you load all those cores just right, you might by at 500 before you know it.  That's scary.

 

The biggest issue I have with cooling my 5960x is not the capacity of my loop because it's actually way more than I need.  It's the fact that the heat from the IHS can't transfer into the waterblock fast enough.

[Jumper118]

7 minutes ago, done12many2 said:

 

4.4 at 1.3v.  Do you see why I say you'll be well over 400w at 5.0 with 1.6v?  Actually, if you load all those cores just right, you might by at 500 before you know it.  That's scary.

 

The biggest issue I have with cooling my 5960x is not the capacity of my loop because it's actually way more than I need.  It's the fact that the heat from the IHS can't transfer into the waterblock fast enough.

yeah. i am thinking my water block is not the best. mine is running at ~55-60c with a massive 1080 rad and about 15l of water in my loop. my cpu block is just a phobya uc-2 lt so its not the greatest ever.  

[done12many2]

3 minutes ago, Jumper118 said:

yeah. i am thinking my water block is not the best. mine is running at ~55-60c with a massive 1080 rad and about 15l of water in my loop. my cpu block is just a phobya uc-2 lt so its not the greatest ever.  

 

Another thing to consider is the additional draw from cache overclocking.  It's a lot more than I initially understood.  Here's an example of what Asus learned when overclocking a BWE chip's cache.  Pretty shocking how much power and heat it added to the overall equation.  Now image a HWE chip at 4.5+ on the cache with slightly higher cache voltage.  

 

 

[Dabombinable]

3 minutes ago, done12many2 said:

 

Another thing to consider is the additional draw from cache overclocking.  It's a lot more than I initially understood.  Here's an example of what Asus learned when overclocking a BWE chip's cache.  Pretty shocking how much power and heat it added to the overall equation.  Now image a HWE chip at 4.5+ on the cache with slightly higher cache voltage.  

 

 

I might limit my 4790K's cache to 4GHz. I've been running it at 4.7GHz for a while....

[Jumper118]

4 minutes ago, done12many2 said:

 

Another thing to consider is the additional draw from cache overclocking.  It's a lot more than I initially understood.  Here's an example of what Asus learned when overclocking a BWE chip's cache.  Pretty shocking how much power and heat it added to the overall equation.  Now image a HWE chip at 4.5+ on the cache with slightly higher cache voltage.  

i am only running mine at 4ghz on 1.25v atm i think  

[done12many2]

Just now, Dabombinable said:

I might limit my 4790K's cache to 4GHz. I've been running it at 4.7GHz for a while....

 

Yeah, I used to run mine high, but gains for daily use are negligible.  Now I just run it at 4.1 GHz @ 1.2v.  No need to crank it up unless I'm benching.