Nehalem 101 part1 - 1366 and X58

[saaya]

Hey guys,

thought id post some details so your already prepared for nehalem and how to clock it.
Ill try to post more details here bit by bit as i have the time :D

What cpu should i get?
well that depends on your budget of course :D
Here is WHO i would recommend WHAT cpu:

920: ~300US$
people who want to try the latest platform
people who want sli on an intel system or might want it in future
occasional gamers
hardcore gamers
workstation users
overclockers on air

940: ~600US$
workstation users
overclockers on water
overclockers on xtreme cooling
3dmark fetishists

965: ~1000US$
demanding workstation users
overclockers on xtreme cooling*
3dmark fetishists *
bandwidth fetishists
people with a big budget seeking the best of the best

*several people report they cant hit 5Ghz with the 965 chips on ln2, so you might want to wait and see if the extra 400$ for a 965 over a 940 is worth it... if most cpus max out below 5Ghz then the only advantage of the 965 is better memory performance thanks to unlocked memory and uncore multipliers which might not be worth an additional 400$

What Overclocks can I expect?
most 965 chips dont seem to hit 5G on ln2...
So far a 940 should be good enough to max out the current chips under ln2 (23x200=4600Mhz)
the gains from Ln2 are rather small for most chips, only around 300 mhz, and lower than -20 temps dont seem to gain much at all.
Most chips dont like higher voltages than around 1.5v and barely scale above that.

965:
~4.5Ghz 3d stable with all 4 cores on ln2
~3.8Ghz 3d stable with all 4 cores on air
~2100Mhz tri channel stable on air
~220Mhz Bclock
~200Mhz Bclock stable on air
~8GT/s or 4Ghz or below QPI speed, cooling doesnt really matter

940:
max 5Ghz (23x220)
~4.5Ghz 3d stable with all 4 cores on ln2
~3.8Ghz 3d stable with all 4 cores on air
~2100Mhz tri channel stable on air
~220Mhz Bclock
~200Mhz Bclock stable on air
~8GT/s or 4Ghz or below QPI speed, cooling doesnt really matter

920:
max 4.5Ghz (21x220)
~4.5Ghz 3d stable with all 4 cores on ln2
~3.8Ghz 3d stable with all 4 cores on air
~2100Mhz tri channel stable on air
~220Mhz Bclock
~200Mhz Bclock stable on air
~8GT/s or 4Ghz or below QPI speed, cooling doesnt really matter

as you can see all cpus seem to clock about the same so far, with some few 965 cpus clocking to 5g and higher.
cooling is definately a big limitation as core i7 cpus run very hot above stock voltage and speed.
similar to k10 most cpus dont like to run at very low temperatures however and the gain from ln2 over watercooling should be very small.
There are some good cpus that clock very well with cold temps of around -100, but those seem to be the big exception so far.
for most cpus there seems to be no or almost no scaling below around -40C, so phase change cooling will have a big revival it seems :D



What Performance gains can I expect?
In Audio/Video/CAD and other highly multithreaded applications higher clockspeeds should result in notable performance gains.
Unfortunately you wont see very high gains if you look at games...
Going from DDR3 1066 to DDR3 1600 barely gives any boosts in games or 3d benchmarks actually... its great for synthetic memory benchmarks, but thats about it. And according to this article on bit-tech.net (http://www.bit-tech.net/hardware/2008/11/06/overclocking-intel-core-i7-920/6) going from 1066 777 to 1600 777 doubled their memory bandwidth in synthetic tests, but this memory overclock plus going from 2.66ghz to 4ghz with a gtx280 gave them a boost of:

4fps in crysis (~10% boost)
6fps in far cry2 (~10% boost)
20fps in hl2 (~10% boost)

so a 50% boost in cpu clocks and a 50% boost in memory clocks results in only a 10% performance boost in games...
Their overclocked 920 was faster than a 965, yet you see there are only very minor gains in gaming performance...
So to spend 3x the money for a 965 over a 920 doesnt make any sense whatsoever if your main focus is gaming.

Clocks:
Instead of the FSB Nehalem uses a reference clock like AMD, but its at 133Mhz and not 200Mhz.
This reference clock is multiplied to create all clocks inside the CPU:

Bclock x CPU Multiplier = CPU Clock (12-256)
Bclock x QPI Multiplier = QPI Clock (2-30)
Bclock x Uncore Multiplier = Uncore Clock (2-30)
Bclock x Memory Multiplier = Memory clock (2-30)

available/working multipliers on 965XE
Bclock x CPU Multiplier = CPU Clock (12-256)
Bclock x QPI Multiplier = QPI Clock (18, 20, 24)
Bclock x Uncore Multiplier = Uncore Clock (10-30)
Bclock x Memory Multiplier = Memory clock (6-30) (max is 14x since Uncore Multiplier needs to be 2x or higher than Memory Multiplier)

available/working multipliers on 940
Bclock x CPU Multiplier = CPU Clock (22)
Bclock x QPI Multiplier = QPI Clock (18)
Bclock x Uncore Multiplier = Uncore Clock (16)
Bclock x Memory Multiplier = Memory clock (6,8) (retail chips might have higher memory multiplier unlocked)

available/working multipliers on 920
Bclock x CPU Multiplier = CPU Clock (22)
Bclock x QPI Multiplier = QPI Clock (18)
Bclock x Uncore Multiplier = Uncore Clock (16)
Bclock x Memory Multiplier = Memory clock (6,8) (retail chips might have higher memory multiplier unlocked)


default Multipliers on 965XE
133 x 24 = 3200MHz CPU
133 x 24 = 3200MHz QPI (6.4GT/s)
133 x 20 = 2666Mhz UnCore
133 x 10 = 1333Mhz Mem (DDR3 1333)

default Multipliers on 940
133 x 22 = 2933MHz CPU
133 x 18 = 2400MHz QPI (4.8GT/s)
133 x 16 = 2133Mhz UnCore
133 x 8 = 1066Mhz Mem (DDR3 1066)

default Multipliers on 920
133 x 20 = 2666MHz CPU
133 x 18 = 2400MHz QPI (4.8GT/s)
133 x 16 = 2133Mhz UnCore
133 x 8 = 1066Mhz Mem (DDR3 1066)


default voltages on 965XE:
default voltages on 940:
default voltages on 920:


Steppings:
retail Core i7 chips will be C0 stepping
some ES Core i7 and nehalem based Xeon chips are B stepping chips
B stepping chips clock a few hundred mhz higher than C stepping chips, at least on Ln2
C stepping chips can take more vdimm, vtt and vcore than B stepping chips without burning the chip, but clock slightly worse

Overclocking:
If you have an XE cpu its all easy, you just change the CPU Multiplier to increase the CPU Clockspeed, thats it...
If you dont have cpu multiplier access, then things get a bit tricky :D
You need to push up the Bclock, which is related to all other clocks though... so when you increase the Bclock you overclock every part of the processor.
QPI, Uncore and the memory controller/memory.

Crime Scene Investiagtion errr i mean QPI: :D
You dont really want to overclock QPI though...
Just like HyperTransport QPI is way overpowered for a Desktop system and only really useful for Servers, and you wont see any notable improvements from clocking the QPI bus higher. So while overclocking it wont really give you any benefits, thats what you have to do if you want to overclock your cpu via the Bclock.

It usually maxes out at around 8GT/s, which is around 4Ghz, and as you have seen from the details above, there are only 2 multipliers you can select unfortunately, which will limit your overclocks on Bclock for the 920 and 940 parts. The lowest multiplier you can select is 18. If you do the maths, that means you will be limited to a Bclock of 222. That is IF you processor and board can run 4Ghz QPI... some cant run that high...

Can I get an Uncore do you want more? :D
Now about Uncore, Uncore is the L3 cache and memory controller, and its powered not by Vcore, but by VTT. I guess Intel couldnt decide what to call this part of the cpu and then went for Uncore which pretty much means its not the core... not very creative, but it gets the job done :D
Im sure youve seen many if not all mainboard manufacturers using one or two or even 4 phases of a CPU PWM to power the VTT. While on Core2 VTT consumed around 10-15W and you could use a basic PWM to feed it, it now needs double that power, around 30W, and when your overclocking, even more, naturally... :D

3-some Memories: mhhh good times, good times :cool:
The Uncore multiplier needs to be 2x or higher than the memory multiplier.
Since Uncore is locked to 2133Mhz on the 920 and 940 higher memory multipliers than 8x are not available. (retail chips might have higher memory multiplier unlocked) So to reach a higher memory clock than 1066 you need an XE cpu which lets you overclock Uncore and use 10x, 12x and 14x memory multipliers. Or you overclock the Bclock which will result in higher memory clocks as well, but with the highest multiplier of 8x you will just barely make it over DDR3 1600 with a bclock of 200, so memory overclocking seems to be limited on the 920 and 940 chips! (retail chips might have higher memory multiplier unlocked) How big the impact of speeds beyond ddr3 1600 really is highly depends on the applications though...

It seems to be quite useful as well to overclock the Uncore even if you dont want to run huge memory clocks btw. Remember it contains the L3 cache, so overclocking it will reduce L3 cache latencies and we all know that the higher the memory controller is clocked, the more efficient it works and the higher the bandwidth, even if the memory is still running at stock. So High Uncore clocks with low memory clocks at tight latencies will definately be a sweet combo! Again memory enthusiasts will NEED an XE to fully max out the memory potential even at low clocks, since only the XE cpus allow Uncore overclocking... (retail 920 and 940 chips might have higher uncore multipliers unlocked)

Does high Vdimm kill Core i7?
Some say yes, some say no...
for those of you who remember amd going IMC some years back, you might remember that the 90nm shrink brought some problems with it... high vdimm could kill or degrade the integrated memory controller. well how did we work around that back then? vcore had to be increased as well to keep the vcore vdimm ratio more or less the same and things were fine. Later when amd moved to K10, the same thing happened again with some imcs dieing at 1.9v ddr2 vdimm if vcore and other related voltages were kept low.
well for amds imc implementation the memory controller was actually powered by vcore, so thats most likely why the vcore vdimm ratio had to be maintained. for nehalem or core i7 the memory controller is powered by vtt...

Intel recommends a max vdimm of 1.65v, which is curiously 1.5x vtt...
Several people reported that running higher than default vtt plus higher than 1.65v vdimm works just fine. how come?
From what i know about manufacturing processes, you have to pick the target voltage you want to work with at some point, and then decide what transistor design to use. some transistors can take high voltages but switch slow and are rather beefy, others are small and can switch much faster but will degrade with higher voltages. which is exactly what people reported with vdimm damaged core i7 cpus. one way to work around this and stress the transistors less is by not grounding them to ground but to some other voltage.

I dont know why or how, but vdimm is definately related to vtt, and the fact that intel recommends a max vdimm of 1.5x vtt is not a coincidence if you ask me... its not clear whether this 1.5x ratio exists and if it has to be maintained, or if vdimm has to be within a certain voltage range to vtt, but in either case the resulting highest vdimm/vtt ratios are about the same... once we have relatively cheap 920 cpus available and people can risk burning a cpus im sure it will only be a matter of weeks until we know more about vdimm/vtt on core i7 :D

http://www.abload.de/img/nehalemvdimmvttratioorlhwi.png

high uncore and memory clocks are possiblewithout high vdimm, we hit ddr3 2200 with 1.7v vdimm on BloodRage a few days ago :D


Setfsb Note:
Once you boot at a certain Bclock you will only be able to bump it up by around 25Mhz before the board gets unstable.
Most likely its caused by the chipset and memory training during bootup, core i7 has a built in routine that automatically fine tunes the chipset and memory during bootup.




Here are some recommended settings for you i got a from a friend at intel.
if your reading this, cheers man :toast:

How to find max Bclock:
Bclock 133Mhz
CPU Multiplier 12x/14x
Uncore Multiplier 16x or higher
QPI Multiplier 18x
Memory Multiplier 8x or lower
Vcore 1.25v
VTT 1.45v
Vdimm 1.65v

push up the Bclock and increase VTT and QPI volts to get higher.
Max 100% safe VTT seems to be 1.65v


How to find max Mem clock:
Bclock 133Mhz
CPU Multiplier 12x/14x
Uncore Multiplier 16x or higher
QPI Multiplier 18x
Highest Memory Multiplier
Vcore 1.25v
VTT 1.45v
Vdimm 1.65v

push up the Bclock and increase VTT and Vdimm to get higher.
more vdimm and vtt than 1.65v at your own risk
1.8vdimm should be 100% safe though



How to find max CPU Clocks on i7-965:
Bclock 133Mhz or lower
CPU Multiplier 24x
Uncore Multiplier 16x or higher
QPI Multiplier 18x
Memory Multiplier 8x or lower
Vcore 1.25v
VTT 1.45v
Vdimm 1.65v

reduce Bclock, then increase the cpu multipliers step by step.
push up vcore to get higher clocks stable


How to find max CPU Clocks on i7-940/i7-920:
max Bclock
CPU Multiplier 15x
Uncore Multiplier 16x or higher
QPI Multiplier 18x
Memory Multiplier 8x or lower
Vcore 1.25v
VTT 1.45v
Vdimm 1.65v

increase CPU multipliers step by step, push up vcore to get higher clocks stable. if you cant get a multiplier stable, reduce Bclock until your stable.
Max bclock for high multis is slightly lower than for low multis.


The return of the Turbo-Button?
The max turbo multipliers are stored in seperate registers from the "normal" max allowed cpu multipliers, and they can not be changed (unless you hack the cpu :D) The turbo multipliers can not be forced to stay on all the time... at least not directly... you can fool the cpu into believing all the conditions that are necessary to run turbo mode are there, even if they are not :D
the highest boot multiplier is the highest "normal" multiplier...

for the i7-965 it works like this:

IF turbo is disabled
OR vcore is unstable
OR current is too high
OR Tdp is too high:

4 cores will run at 24x133=3200Mhz
3 cores will run at 24x133=3200Mhz
2 cores will run at 24x133=3200Mhz
1 cores will run at 24x133=3200Mhz

IF Turbo is enabled
AND vcore is stable
AND current is ok
AND the TDP is below the limit

4 cores will run at 25x133=3450Mhz
3 cores will run at 25x133=3450Mhz
2 cores will run at 25x133=3450Mhz
1 core will run at 26x133=3600Mhz


for the i7-940 it works like this:

IF turbo is disabled
OR vcore is unstable
OR current is too high
OR Tdp is too high:

4 cores will run at 22x133=2933Mhz
3 cores will run at 22x133=2933Mhz
2 cores will run at 22x133=2933Mhz
1 cores will run at 22x133=2933Mhz

IF Turbo is enabled
AND vcore is stable
AND current is ok
AND the TDP is below the limit

4 cores will run at 22x133=2933Mhz
3 cores will run at 22x133=2933Mhz
2 cores will run at 22x133=2933Mhz
1 core will run at 23x133=3050Mhz

so you see, turbo is really not that useful for most of the cpus, its only useful for the 965, where it basically auto overclocks the cpu... so basically the Core i7 lineup looks like this:

920 2800Mhz BUT if vcore is unstable, current is too high, or temps are too high, it will run slower.
940 3050Mhz BUT if vcore is unstable, current is too high, or temps are too high, it will run slower.
965 3600Mhz BUT if vcore is unstable, current is too high, or temps are too high, it will run slower.

What turbo really is, is a more advanced form of throttling the cpus from those clockspeeds down to a SAFE speed that will work even with high temps, fluctuating vcore and high current. instead of advertising a 3600mhz cpu that throtles down to 3400 or less depending on the situation, which would cause a lot of complains, the cpus are rated at 3200mhz and everything else above it is a BONUS, so nobody can complain :D


This is it or now, ill add more details later on. if you have any questions, just ask :toast:

[Ghigo]

Hmm the +2 on multiplier overclock from turbo mode can be forced always active ?

[HDCHOPPER]

whew :wth: nehalem going to be a time consumer for me :ROTF: not a AMD fan at all

but when my BLOODRAGE board gets here :worship::worship::D
it's gonna be fun :D
and all our new nehalem coustermers are going to get Foxconn aimed at them:up:

[Movieman]

Saaya: Cleaned your thread for you..;)

[eva2000]

Thanks Sascha for the 101 :)

Looks like memory oc'ing will be alot of fun finding out all the tweaks and sweet spots :D

[Movieman]

Thanks Sascha for the 101 :)

Looks like memory oc'ing will be alot of fun finding out all the tweaks and sweet spots :D

Now all I need is Sascha to stop over the house and stand behind me.
HMM, Taiwan to New Hampshire isn't that long a trip is it?:rofl:

[Oliver]

thx 4 posting

[zanzabar]

is that memory clock that u posted the actual clock or the ddr clock

and the 940 has a multi of 22 right so that would be a max of 4.88ghz befor a qpi wall thats a nice wall i expected much lower

[philbrown23]

wow so this is going to be like amd, man what a bummer. oh well i guess.

[Gendo]

Sounds exciting on one side; it's a new way of doing things which is always fun. On the other hand it seems that overclocking these cpu's is very limited. I doubt we'll see 6Ghz+ in this first generation of Nehalem:(

[HawaiianSupermn]

Wow, I was just getting used to overclocking current gen, I guess its back to the rear of the short yellow bus for me. :shrug:

[saaya]

i updated the first post with a lot more interesting infos :toast:


Hmm the +2 on multiplier overclock from turbo mode can be forced always active ?unfortunately no...
the highest boot multiplier is the highest "normal" multiplier...

for the i7-965 it works like this:

IF turbo is disabled
OR vcore is unstable
OR current is too high
OR Tdp is too high:

4 cores will run at 24x133=3200Mhz
3 cores will run at 24x133=3200Mhz
2 cores will run at 24x133=3200Mhz
1 cores will run at 24x133=3200Mhz

IF Turbo is enabled
AND vcore is stable
AND current is ok
AND the TDP is below the limit

4 cores will run at 26x133=3450Mhz
3 cores will run at 26x133=3450Mhz
2 cores will run at 26x133=3450Mhz
1 core will run at 27x133=3600Mhz


for the i7-940 it works like this:

IF turbo is disabled
OR vcore is unstable
OR current is too high
OR Tdp is too high:

4 cores will run at 22x133=2933Mhz
3 cores will run at 22x133=2933Mhz
2 cores will run at 22x133=2933Mhz
1 cores will run at 22x133=2933Mhz

IF Turbo is enabled
AND vcore is stable
AND current is ok
AND the TDP is below the limit

4 cores will run at 22x133=2933Mhz
3 cores will run at 22x133=2933Mhz
2 cores will run at 22x133=2933Mhz
1 core will run at 23x133=3050Mhz

so you see, turbo is really not that useful for most of the cpus, its only useful for the 965, where it basically auto overclocks the cpu... so basically the Core i7 lineup looks like this:

920 2800Mhz BUT if vcore is unstable, current is too high, or temps are too high, it will run slower.
940 3050Mhz BUT if vcore is unstable, current is too high, or temps are too high, it will run slower.
965 3600Mhz BUT if vcore is unstable, current is too high, or temps are too high, it will run slower.

What turbo really is, is a more advanced form of throttling the cpus from those clockspeeds down to a SAFE speed that will work even with high temps, fluctuating vcore and high current. instead of advertising a 3600mhz cpu that throtles down to 3400 or less depending on the situation, which would cause a lot of complains, the cpus are rated at 3200mhz and everything else above it is a BONUS, so nobody can complain :D


Saaya: Cleaned your thread for you..;)
thanks :D


Thanks Sascha for the 101 :)

Looks like memory oc'ing will be alot of fun finding out all the tweaks and sweet spots :Dyes, things are getting a lot more flexible now :D
im still worried about high vdimm though... :D


Now all I need is Sascha to stop over the house and stand behind me.i take that as a compliment, but no, im into females... but thanks anyways :D *takes cover from dave throwing bricks*

HMM, Taiwan to New Hampshire isn't that long a trip is it?:rofl:no, and with the climate and economy going down, i think ill actually walk :P :D


is that memory clock that u posted the actual clock or the ddr clock

and the 940 has a multi of 22 right so that would be a max of 4.88ghz befor a qpi wall thats a nice wall i expected much lowerwell thats IF you can get 222Bclock stable, more realistically is 200, which means your limited to 4400mhz... and yes, the likelihood of your cpu crapping out before 4.8ghz is actually there too, so its not that bad of a qpi limitation. but for the 920 its pretty bad... itll be possible to max out the cpu on air before qpi limits it, but under ln2 the 920 will be pretty worthless... unfortunately as its priced very nicely...

and if you really want to get at it with ln2, then you NEED the 965, the 940 will be qpi limited... if you got a good set of cores you can reach 5ghz+ on the cores... supposedly some guys are already at 5.5... so there you go, a non XE 3.2ghz core i7 that will probabaly follow later wont be useful for LN2... 24x200 is 4.8ghz and thats not the limit for a set of good cores under ln2...


Sounds exciting on one side; it's a new way of doing things which is always fun. On the other hand it seems that overclocking these cpu's is very limited. I doubt we'll see 6Ghz+ in this first generation of Nehalem:(suicide screenshot with one core, yeah, but 3d with 4 cores, no way :D

intel should have held back those 6.5g 3d stable 8600s :D


Wow, I was just getting used to overclocking current gen, I guess its back to the rear of the short yellow bus for me. :shrug:well if your familiar to K8 or K10... this is basically the same with DDR3 and an intel logo on it :D

[freshy]

thanx for the info saaya, good read :clap:, so with a core i7 920 should hit 4ghz on water ok with a good board and be stable ?

[saaya]

hmmmmm yes, im not 100% sure, but yeah, 4g on water should be doable for the 920
i havent tried many cpus tho, and i dont know how the retail cpus clock... so you might only get 3.8 or 3.9...

[Ghigo]

Thx golden infos :yepp:

[Slovnaft]

Wow thanks for that clockspeed data. 5.5ghz 3d stable already? i'll def be splurging on a 965. can't wait to see those results, i wonder whether 5.5ghz i7 run will beat out a 6.5ghz E86 run on benches that don't value multithreadedness as much as 06?

[naokaji]

*Cough cough* now all we need is a release date and price for the Bloodrage:p:

Nice thread Saaya, thx for all the info, will be helpful once the emperor ehh nehalem arrives.

[TheKarmakazi]

First off great thread and awesome info!! Thanks for the heads up, I will definitely be getting a 965.... just gotta do a fire sale here and drop all the S775 stuffz (like 6 mobos and 11 cpus lol!). Got plenty of DDR3 ready..



Wow thanks for that clockspeed data. 5.5ghz 3d stable already? i'll def be splurging on a 965. can't wait to see those results, i wonder whether 5.5ghz i7 run will beat out a 6.5ghz E86 run on benches that don't value multithreadedness as much as 06?

Should be pretty close... Nehalem is supposed to be about 15% faster clock per clock (right?)... So the maths:

5500Mhz x .15 = 825

5500 + 825 = 6325Mhz equivalent E8600

But quad cores have some advantages in non threaded benches also since you can run certain process on certain cores thereby distributing the load. Not too mention by disabling a core or 2 you should be able to gain more Mhz...

[saaya]

Thx golden infos :yepp:
your welcome! :toast:


Wow thanks for that clockspeed data. 5.5ghz 3d stable already? i'll def be splurging on a 965. can't wait to see those results, i wonder whether 5.5ghz i7 run will beat out a 6.5ghz E86 run on benches that don't value multithreadedness as much as 06?
some wont even hit 5G with all 4 cores 3d stable, so dont expect to much!
i said 5.5g and below, meaning i doubt we will see more than 5.5g anytime soon :D


*Cough cough* now all we need is a release date and price for the Bloodrage:p:soon and less than BlackOps at launch :D


Nice thread Saaya, thx for all the info, will be helpful once the emperor ehh nehalem arrives.your welcome :toast:



First off great thread and awesome info!! Thanks for the heads up, I will definitely be getting a 965.... just gotta do a fire sale here and drop all the S775 stuffz (like 6 mobos and 11 cpus lol!). Got plenty of DDR3 ready..heh, well, if you have a good 8600 or 775 board you might wanna keep it, price perf will beat nehalem and in some benches im not sure if nehalem will take the perf crown...


Should be pretty close... Nehalem is supposed to be about 15% faster clock per clock (right?)... So the maths:

5500Mhz x .15 = 825

5500 + 825 = 6325Mhz equivalent E8600weellll i would say thats an oversimplification :D
theres no linear scaling and the higher ipc doesnt really do much in 3d apps as we can see from all the benchmark results posted already... its all about the cpu tests and then the added bandwidth of the tri channel imc helps a little too...

[Tony]

i7 Bclock x CPU Multiplier = CPU Clock (12-256)

AMD HTT ref frequency x cpu multi

i7 Bclock x QPI Multiplier = QPI Clock (18, 24)

AMD HTT X HT Multiplier = HT clock

i7 Bclock x Uncore Multiplier = Uncore Clock (10-30) This is IMC core speed.

AMD HTT x IMC Multi= IMC core clock speed

i7 Uncore clock / Memory Divider = Memory clock

This looks slightly different, probably due to it being DDR3 over DDR2

Overall though, its the same thing ;)

I maximised IMC core clock speed, this is where performance will be gained the most. If you increase IMC speed, keep ram speed sensible and keep tight timings there will be no need for massive vdimm and any risk to the IMC(cpu) overall.

[eva2000]

Tony how strong will the IMC be for overclocking headroom for 2x1GB vs 3x1GB x 6x1GB or 2x2GB vs 3x2GB vs 6x2GB ?

[Tony]

No idea yet Eva, i will get going next week although i still need boards. For now i would think 3x1 is easy, 3x2 is near as easy and 6x1 is like 3x2....6x2 will be a lot of load for clocking and 4gb dimms even more.

im going to concentrate my efforts on 3x2gb clocking as most will run 6GB with Vista 64 first then w7 later

[saaya]

yeah, very similar... the only difference is that on k10 you can actually ramp up the l3 cache clock independantly from the imc, right?
and yeah, 3x2gb is what everybody will run, except for benchers and xp fans who stick to 3x1gb.

the interesting part is that k10 cant get a ref clock much above 230 and i7 seems to be stuck at about that speed too, even lower actually... i wonder if there is some reason for this or its just a coincidence and the i7 bclock is really just limited by qpi and nothing else...
and if thats the case, then maybe amds ref clock is limited by some other clock that gets derived from the ref clock?

[WhiteFireDragon]

ahh it's learning to OC all over with so many terminologies. got too use to the old fsb, multi, and divider. i'm just gonna wait until i actually get nehalem to start fiddling with it if i can ever afford the CPU, mobo, and ddr3 :[

[SF3D]

Good thead Sascha!

This is really informative and helpfull thread for sure.

Overclocking Core i7 is not so hard or different compared to Core 2 Duo :)

[saaya]

thanks! :toast:

id like to add some more infos regarding impact of HT on cpu scores in 3dmark,
4 cores with ht vs 4 cores
4 cores vs 2 cores with ht

rough mhz differences between
max clock for 1 core
max clock for 2 core
max clock for 4 core

rough mhz differences between max suicide and fully stable for 1, 2 and 4 cores

[Tony]

yeah, very similar... the only difference is that on k10 you can actually ramp up the l3 cache clock independantly from the imc, right?
and yeah, 3x2gb is what everybody will run, except for benchers and xp fans who stick to 3x1gb.

the interesting part is that k10 cant get a ref clock much above 230 and i7 seems to be stuck at about that speed too, even lower actually... i wonder if there is some reason for this or its just a coincidence and the i7 bclock is really just limited by qpi and nothing else...
and if thats the case, then maybe amds ref clock is limited by some other clock that gets derived from the ref clock?

IMC and L3 same speed...as far as I remember

Regards Phenom, i have seen 280+ here...other have seen 300, its not the CPU that is the issue

[SAE]

K. Huge thanks. Think I'll bookmark this, Sascha.
Great read for n00bs like me. :D

[saaya]

Regards Phenom, i have seen 280+ here...other have seen 300, its not the CPU that is the issuehuh? really? then what is the limit and how do you get those high clocks?


K. Huge thanks. Think I'll bookmark this, Sascha.
Great read for n00bs like me. :Dlol :P

[Movieman]

I got 4226 but all with the multi..
Max I could do on the buss was 150 with any multi

[Slovnaft]

retail must clock better than ES
fugger's 5.5ghz chip is retail i believe.

[saaya]

I got 4226 but all with the multi..
Max I could do on the buss was 150 with any multi
is that the max stable for 4 cores loaded?
if yes then thats pretty good on air!
only 150 though is weird... did you lower the qpi divider to 18?

[Movieman]

is that the max stable for 4 cores loaded?
if yes then thats pretty good on air!
only 150 though is weird... did you lower the qpi divider to 18?

That was max SP1M stable with HT enabled.
What QPI divider? I don't see one in this bios..:rofl:

[saaya]

:0 no qpi divider?

[naokaji]

I got 4226 but all with the multi..
Max I could do on the buss was 150 with any multi

With what board though? it appears the Intel boards are heavily limited if it comes to Bclock oc.

[saaya]

i was wrong about memory dividers, its not derived from the uncore clockspeed but multiplier directly from the Bclock, my bad!
i corrected it and added a lot more details to the first post :)

[Praz]

Very nice writeup. Your post will save users a lot of time when first jumping to this platform. Also makes it easier to choose which processor should be bought based on their requirements and expectations.

[saaya]

Very nice writeup. Your post will save users a lot of time when first jumping to this platform. Also makes it easier to choose which processor should be bought based on their requirements and expectations.

thanks :toast:
and yes, that was the idea, to make sure people know what to expect.
around 4Ghz should be max with the 920 and good aircooling, which isnt bad at all, and despite everything beeing locked, DDR3 1600 is possible with the 8x memory multiplier and a bclock of 200. going from 1066 to 1600 barely gives any boosts in games or 3d benchmarks actually... its great for synthetic memory benchmarks, but thats about it.

according to this article on bit-tech.net (http://www.bit-tech.net/hardware/2008/11/06/overclocking-intel-core-i7-920/6) going from 1066 777 to 1600 777 doubled their memory bandwidth in synthetic tests, but this memory overclock plus going from 2.66ghz to 4ghz with a gtx280 gave them a boost of:

4fps in crysis (~10% boost)
6fps in far cry2 (~10% boost)
20fps in hl2 (~10% boost)

so a 50% boost in cpu clocks and a 50% boost in memory clocks results in a 10% performance boost in games... based on that id say:

920:
people who want to try the latest platform
people who want sli on an intel system or might want it in future
occasional gamers
hardcore gamers
workstation users
overclockers on air

940:
workstation users
overclockers on air
overclockers on water
overclockers on xtreme cooling

965:
demanding workstation users
overclockers on xtreme cooling
bandwidth fetishists
3dmark fetishists
people with a huge budget seeking the best of the best

[npk]

saaya any chance on fooling the cpu or mobo trick to be able to go over x20 with an i7 920 all 8 threads loaded with turbo mode?

[saaya]

yes, but only up to 21 with 4 cores and 8 threads, and 22 with 1 core and 2 threads...

[LinusTech]

fantastic thread.

[saaya]

it seems higher memory multipliers and uncore multipliers will be unlocked on retail 920 and 940 chips after all? :confused:
i hope this is true, after all thats what intel promised everybody at idf and several other occasions :D

[saaya]

it seems higher memory multipliers and uncore multipliers will be unlocked on retail 920 and 940 chips after all? :confused:
i hope this is true, after all thats what intel promised everybody at idf and several other occasions :D

[Praz]

it seems higher memory multipliers and uncore multipliers will be unlocked on retail 920 and 940 chips after all? :confused:
i hope this is true, after all thats what intel promised everybody at idf and several other occasions :D
Hopefully this will be prove to be true. Although as you have already wrote it won't make a big difference either way for most users but the option would be nice to have.

[HDCHOPPER]

anything being unlocked def makes anything more wantable to me :D

it's gonna be a fun the next copula years for shure

[G.Foyle]

it seems higher memory multipliers and uncore multipliers will be unlocked on retail 920 and 940 chips after all? :confused:
i hope this is true, after all thats what intel promised everybody at idf and several other occasions :D

Weird thing, of 5 motherboards I tested only MSI Eclipse allowed all memory dividers with i7-920. I think other boards will have them unlocked with new bios versions.

[LinusTech]

Hey Saaya, you haven't really touched on CPU PLL voltage. Any thoughts?

[LinusTech]

Another question: Is max BClk determined by the CPU or the motherboard?

[SAE]

I think, CPU maxes at about 220MHz and boards determine if you are able to get there. ;)

[saaya]

Weird thing, of 5 motherboards I tested only MSI Eclipse allowed all memory dividers with i7-920. I think other boards will have them unlocked with new bios versions.with the same cpu?
i checked our 920s and 940s and they have the memory multipliers locked.
basically the 920 and 940 cpus we have are double locked, uncore is locked AND the mem multis are locked.

its possible that you have a chip which doesnt have uncore and mem multis locked, but only the msi bios implemented uncore ratio changes for non 965 parts. and if the uncore doesnt change, then you cant use the higher memory multipliers...


Hey Saaya, you haven't really touched on CPU PLL voltage. Any thoughts?havent played with it yet...


Another question: Is max BClk determined by the CPU or the motherboard?both somewhat, but its more a cpu than board thing i think.


I think, CPU maxes at about 220MHz and boards determine if you are able to get there. ;)yeah, but as far as i know every board can get there... :D and its not really 220Mhz that you gotta reach, its the 4Ghz QPI speed...

[LinusTech]

I am going to be doing a video version of "Nehalem Overclocking 101" for NCIX Tech Tips so this thread has been incredibly valuable.

Does anyone know how 12GB vs. 6GB of memory affects overclocking?

[tiro_uspsss]

this write-up is awesome! :slobber:

:bows: :bows: :bows: :bows: :bows:

:toast:

[saaya]

I am going to be doing a video version of "Nehalem Overclocking 101" for NCIX Tech Tips so this thread has been incredibly valuable.

Does anyone know how 12GB vs. 6GB of memory affects overclocking?

your welcome :toast:
some things might be wrong though, im having a hard time either dealing with un or misinformed engineers and intel reps here or for some reason the language problem causes me to understand the exact opposite of whats true :D

it seems 920s and 940s will have mem and uncore multis unlocked now, which doesnt make sense though, why would intel send every mainboard maker cpus that are locked, and then sell retail chips that are unlocked... the other way around it makes sense, but this is really weird... :confused:
maybe it was a last minute decision to unlocked the mem multis on the retail chips...

regarding memory configs, well it depends on the amount of chips and the density, plus amount of sticks, plus what chips you actually use etc.
in general id say:
1gb stick samsung =/> 1gb stick micron
2gb stick samsung > 2gb stick micron
3x1gb =/> 3x2gb > 6x1gb >/= 6x2gb > 3x4gb

thanks tiro_uspsss :toast:

[saaya]

Does high Vdimm kill Core i7?
Some say yes, some say no...
for those of you who remember amd going IMC some years back, you might remember that the 90nm shrink brought some problems with it... high vdimm could kill or degrade the integrated memory controller. well how did we work around that back then? vcore had to be increased as well to keep the vcore vdimm ratio more or less the same and things were fine. Later when amd moved to K10, the same thing happened again with some imcs dieing at 1.9v ddr2 vdimm if vcore and other related voltages were kept low.
well for amds imc implementation the memory controller was actually powered by vcore, so thats most likely why the vcore vdimm ratio had to be maintained. for nehalem or core i7 the memory controller is powered by vtt...

Intel recommends a max vdimm of 1.65v, which is curiously 1.5x vtt...
Several people reported that running higher than default vtt plus higher than 1.65v vdimm works just fine. how come?
From what i know about manufacturing processes, you have to pick the target voltage you want to work with at some point, and then decide what transistor design to use. some transistors can take high voltages but switch slow and are rather beefy, others are small and can switch much faster but will degrade with higher voltages. which is exactly what people reported with vdimm damaged core i7 cpus. one way to work around this and stress the transistors less is by not grounding them to ground but to some other voltage.

I dont know why or how, but vdimm is definately related to vtt, and the fact that intel recommends a max vdimm of 1.5x vtt is not a coincidence if you ask me... so sticking to this 1.5x vtt rule is what we should do if we want to use high vdimm.

http://www.xtremesystems.org/forums/attachment.php?attachmentid=88844&stc=1&d=1226390827

[Fr3ak]

What voltage do you refer to as VTT? Voltage of the integrated memory controller labeled as QPI/Vdimm on some boards?

[saaya]

only asus calls it that and its pretty confusing... :D
i think they call it that cause it helps clocking up qpi and memory...
but its not the voltage of the memory controller, its the entire uncore supply voltage.
it powers the memory controller AND the L3 cache and i think also a part of the qpi controller since more vtt helps to clock to higher qpi speeds.

[saaya]

most 965 chips dont seem to hit even 5G on ln2...
So far a 940 should be good enough to max out the current chips under ln2 (23x200=4600Mhz)

the gains from Ln2 are rather small for most chips, only around 300 mhz, and lower than -20 temps dont seem to gain much at all.

Most chips dont like higher voltages than around 1.5v and barely scale above that.

[Fr3ak]

Thanks for the explanation. Now a couple things are more clear to me =)

The 2 965 I tried seem to be way worse than some chips here at XS. I can run 215 MHz BCLK on a P6T with a CPU multi of 12, but cannot boot at 4 GHz. No way to hit 4.5 GHz on air with those CPUs. But there are so many new things for pverclocking with Nehalem, so it might need some fine tuning and a lot of trial and error to figure out which combinations work the best.

But nontheless, seeing what Nehalem CPUs can do on just aircooling at a relativly low voltage is very nice. I guess we will even see better results once it hits retail.

[saaya]

hmm whats the max you can hit then?
did you try high multi 133 bclock?
what vcore and vtt are you running?

[Fr3ak]

I can hit 385x MHz, but only 20 mins Prime stable before the PC resets. I tried up to 1.4V Vcore and VTT + 0.2 V max. Trying to find 24/7 settings here. BCLK below 133 MHz seems to be very buggy. 100 x anything does not work at all, 120 MHz works, 110 does not, etc

[G H Z]

Nice thread saaya :up:


I dont know why or how, but vdimm is definately related to vtt, and the fact that intel recommends a max vdimm of 1.5x vtt is not a coincidence if you ask me... so sticking to this 1.5x vtt rule is what we should do if we want to use high vdimm.

http://www.xtremesystems.org/forums/attachment.php?attachmentid=88844&stc=1&d=1226390827

That might be a hard rule to follow ;)

[bmg]

Nice thread saaya :up:

That might be a hard rule to follow ;)

I'd be pretty sure it's not a ratio thing, but more like a max difference between voltages. I'm guessing it's something more like Vdimm shouldn't be more than .5v-.6v greater than Vtt, so Vtt should be raised along with Vdimm. I imagine this will happen anyway, as Vtt will normally have to be increased to run higher memory speeds. It might also apply to Vcore, but again most of us will be increasing Vcore a bit anyway.

[G.Foyle]

I can hit 385x MHz, but only 20 mins Prime stable before the PC resets. I tried up to 1.4V Vcore and VTT + 0.2 V max.

It takes much more voltage to get Nehalems to 4 GHz than with Penryns. I could hit 4GHz only with 1.45V on the 965 and 1.475 on the 920. Seems like 920 can be maxed on air, at around 4100 MHz and 1.5 V.

[saaya]

Nice thread saaya :up:
That might be a hard rule to follow ;)
thanks mark! :)


I'd be pretty sure it's not a ratio thing, but more like a max difference between voltages. I'm guessing it's something more like Vdimm shouldn't be more than .5v-.6v greater than Vtt, so Vtt should be raised along with Vdimm.
but then youd need around 1.8v vtt for 2.3v vdimm, and i know of some people who supposedly ran 2.3v fine with much less than 1.8v vtt...
and youd need 1.5v vtt to run 2v vdimm, and again i know people who run 2v vdimm with less than 1.5v vtt...
as a matter of fact asus boards automatically set vdimm to 2v and vtt to 1.3v as soon as you set the 12x memory divider. and they do this without even showing that they bump up vdimm and vtt, it shows as "auto".
Im pretty sure asus isnt stupid and sets those voltages by default if they are not safe and will kill the cpus after some weeks... im pretty sure they tested this and found that 2v vdimm is stable with 1.3v vtt...

if its an offset then im sure its more than .55v...
asus seems to be pretty sure that 2v vdimm 1.3v vtt is safe
several people reported 1.8v vdimm 1.1v vtt is safe
that points to an offset of rather .7v, or max .7v to be more precise...
and in that case it doesnt make a big difference if we are dealing with an offset or ratio thing between vdimm and vtt at all, unless we look at vdimm voltages way above 2.25v we get almost the same vtt to vdimm ratio for high vdimm...

http://www.abload.de/img/nehalemvdimmvttratioorlhwi.png

i guess time will tell :)



It takes much more voltage to get Nehalems to 4 GHz than with Penryns. I could hit 4GHz only with 1.45V on the 965 and 1.475 on the 920. Seems like 920 can be maxed on air, at around 4100 MHz and 1.5 V.yeah but olli tried 1.5v right? and he couldnt even get 3.8ghz stable... def sounds like either a bad chip or something else holding you back olli. fire off a pm to kinc and ask him for some settings, maybe he can help :)


I imagine this will happen anyway, as Vtt will normally have to be increased to run higher memory speeds. It might also apply to Vcore, but again most of us will be increasing Vcore a bit anyway.i dont think its related to vcore... everbody came to the conclusion that vtt needs to be reaised for high vdimm to work ok, and many are automatically raising vtt if you set high vdimm or high mem ratios, nobody is touching vcore for that it seems... and vcore and the imc are on completely different power planes, there is no connection whatsoever from vdimm to the actual cpu cores, so i dont think vcore plays into this... but we will see :D

EDIT: i really dont think vcore has to do with this... just think about it, with dynamic vcore thanks to EIST intel would be in a LOT of trouble if low vcore would cause problems with high vdimm... cause in idle the cpu vcore will drop to 0.x and vdimm remains the same...

[Movieman]

I can hit 385x MHz, but only 20 mins Prime stable before the PC resets. I tried up to 1.4V Vcore and VTT + 0.2 V max. Trying to find 24/7 settings here. BCLK below 133 MHz seems to be very buggy. 100 x anything does not work at all, 120 MHz works, 110 does not, etc

Max I could get with the Intel board was 4226 on air(9.812s)
vcore : 1.4825
vdimm:1.62
31x136
SP1M stable at those settings but not much else.

max 24/7 100% load speed of 3733
vcore:1.35
vdimm:1.52
28x133
TRUE w/87cfm fan
Load temps at 60C
Very stable at that speed, had it at 100% load for the last 10 days.

[kiwi]

920:
max 4.5Ghz
~4.5Ghz 3d stable with all 4 cores on air

huh? :D

[Movieman]

huh? :D

yea, I want to see that also..

and room temps..maybe his air had a "chill" to it..mine was 72F..:D
Maybe a bit more..There was a 8 core clover right behind it at 100% load and they toss some heat..;)

[Fr3ak]

In fact I tested up to 1.7V (by accident, more or less). I have sen 4.2 GHz on air, but being completely stable is a different story.

I am doing a mainboard OC roundup, so I am interested in voltages, max BCLK and 2D/3D performance. No need to mess around with high CPU clocks for now.

[saaya]

those numbers are mostly based on that people told me, not my own results...
yeah i should edit that 4.5g on air thing :D

60C with true on 3.8 with 1.3v dave?
daym! :eek:

[Fr3ak]

Temps are pretty good for me too. Around 60-65C under load with 1.35V set in BIOS using a Noctua NH-C12P (in case Core Temp displays the right temperatures).

[dinos22]

great thread man

i emailed Richard over at bit-tech about his claim to keep uncore within 0.5v of vdimm which will keep the CPU safe

Richard confirmed on email that Francois from Intel told him this about uncore & vdimm relationship during his interview

maybe something you should add in first post

here is what Richard wrote about it

http://www.bit-tech.net/hardware/2008/11/06/overclocking-intel-core-i7-920/3


QPI/DRAM Voltage - 1.35V: This is poorly worded by Asus - it should read uncore or QPI/memory controller voltage so not to confuse it with the actual memory voltage. Increasing this is also necessary as it helps overclock the base frequency as the uncore area overclocks increase in relation to the CPU core overclocks. This voltage is tied to actual DRAM voltage - the two are directly connected on the motherboard. You'll need to increase this to keep the CPU safe.

While Asus and Intel (rightly) scare everyone (read: uneducated) into thinking that 1.65V on the DRAM voltage should be the absolute limit before you reach for the fire-blanket, all that's really needed it to obey this: keep the CPU uncore voltage within 0.5V difference of the DRAM voltage and there's no problem. Over this potential difference and you’ll greatly increase the chance of CPU death, but it certainly won't happen instantly in a big ball of fail fire if you make a mistake.

DRAM Voltage - 1.66V: This is the closest to the 1.65V the Corsair Dominator DIMMs wanted and it's within the 0.5V Uncore difference.

[Fr3ak]

Wasn't it kind of the same back in the AMD days? I don't remember what chipset it was, but A64 if I am not mistaken.

I take it that "uncore voltage" means VTT or QPI voltage? Pretty messed up naming =/ Intel should have published a "Nehalem naming and functions for dummys"-article. Seeing mainboard manufacturers using different wording, I am pretty sure, I am not the only one being a little confused here ;)

But it seems Richard is wrong about one thing:
"BCLK Frequency - 200(MHz): BCLK = Base clock. This affects the QPI and uncore (L3 cache, northbridge) frequency too - some motherboards like the Intel Smackover allow the ratios to be adjusted, but the Asus does not. Here the ratio is 18x for the QPI and 16x for the uncore (northbridge/L3 cache). The uncore frequency must also always be below the QPI, we’re told."

Either he was using a different BIOS or he was testing the Intel Smack Over with a 965, because with anything non-XE, QPI and uncore cannot be changed with any board I have tested so far.

Edit: Might be a BIOS issue. The beta BIOS of the X58-Extreme also alows QPI and uncore clock changes on a 920.

[dinos22]

Edit: Might be a BIOS issue. The beta BIOS of the X58-Extreme also alows QPI and uncore clock changes on a 920.

say what

that IS interesting :D

[Oliver]

say what

that IS interesting :D
indeed

[Fr3ak]

Changing QPI and uncore multiplier on a 920 works on the Smack Over and Gigabyte EX58-Extreme =)

On the P6T and Rampage 2 Extreme changing those values is not possible onm non-XE CPUs.

Edit: Uncore multiplier cannot be changed with the Smack Over, only QPI multiplier.

Being able to just change the QPI multiplier might be beneficial for perfoamance, but not for overclocking as you set it to the lowest multiplier anyway to get the max BCLK out of the board.

[Kensek]

Have you tried R2E 0602 Bios version.

ftp://ftp.asus.com.tw/pub/ASUS/mb/LGA1366/RampageII_Extreme/



P6T is 0804 ftp://ftp.asus.com.tw/pub/ASUS/mb/LGA1366/P6T_Deluxe/

[G.Foyle]

Have you tried R2E 0602 Bios version.

I tried it today. No difference whatsoever in max BCLK, max clock or SuperPI performance. OC settings from onboard switches seem to apply faster, but thats just my impression.

[bmg]

Gary Key's (bingo13 on this site I believe) blogpost on Anandtech addresses the voltage isssue http://www.anandtech.com/weblog/showpost.aspx?i=521

To quote the key paragraph: "Keeping VCore, VDimm, and VQPI in proper sync is the key to a stable overclock at voltages that will not shorten your processor's lifespan, well not by that much from what we can tell at this point. Intel's guidance continues to be not setting VCore past 1.55V, VDimm past 1.65V and VQPI (uncore to us) past 1.315V when overclocking. Kris will go over this in detail shortly, but in my experiences so far, getting these three voltages too far out of sync will quickly cause problems as one of our i965 processors can attest to now. Keeping them in sync and tweaking a few other settings will result in a finely tuned system that is capable of running settings higher than Intel's guidance although it is a warranty buster. At this point we like to keep VCore and VQPI within 0.0250~0.0375V of each other as you clock up, generally speaking, once you exceed about 1.3V on VCore it is time to start syncing these settings and others in the BIOS. "

[slash777]

Thx man for the infos !

[saaya]

great thread man

i emailed Richard over at bit-tech about his claim to keep uncore within 0.5v of vdimm which will keep the CPU safe

Richard confirmed on email that Francois from Intel told him this about uncore & vdimm relationship during his interview

maybe something you should add in first post
here is what Richard wrote about it

http://www.bit-tech.net/hardware/2008/11/06/overclocking-intel-core-i7-920/3
yeah ive seen that, but... hes got quite a few things wrong so i didnt really want to quote any of that and dont see that as a reliable source :P



I take it that "uncore voltage" means VTT or QPI voltage? Pretty messed up naming =/ Intel should have published a "Nehalem naming and functions for dummys"-article. Seeing mainboard manufacturers using different wording, I am pretty sure, I am not the only one being a little confused here ;)
they actually did and its called the white papers/yellow papers :D
the problem is that asus and others invent new names for those voltages instead of using the voltages that are common in all the technical documents...


But it seems Richard is wrong about one thing:
"BCLK Frequency - 200(MHz): BCLK = Base clock. This affects the QPI and uncore (L3 cache, northbridge) frequency too - some motherboards like the Intel Smackover allow the ratios to be adjusted, but the Asus does not. Here the ratio is 18x for the QPI and 16x for the uncore (northbridge/L3 cache). The uncore frequency must also always be below the QPI, we’re told."

Either he was using a different BIOS or he was testing the Intel Smack Over with a 965, because with anything non-XE, QPI and uncore cannot be changed with any board I have tested so far.

Edit: Might be a BIOS issue. The beta BIOS of the X58-Extreme also alows QPI and uncore clock changes on a 920.hes wrong with several things he wrote. there is no limitation between qpi, uncore, and cpu clock.
they can all be higher or lower than the other, there seems to be no limitation whatsoever. i didnt test stability though so maybe one is unstable or x has to be higher than y to push it to higher clocks...

and yes, its just a bios thing if you can change uncore and mem multipliers and qpi multipliers or not. im sure everybody will add it soon...

[Charles Wirth]

To correct Gary "VQPI (uncore to us)"

Yes, Vdimm and uclk are directly related and raising vqpi doesnt scale uclk, vdimm does. That is the secret to free performance, that is how I hit 25K and 18.9ns latency.


they can all be higher or lower than the other true

Ill disclose the voltages I am running at on each board.

[Charles Wirth]

Rampage/EX58/EX58 Extreme/Smackover/P6T
3 Stage cascade -100c
Corsair 2133 and 1866 Dominator DDR3 - Samsung Rev E

Voltages

vcore ~1.5v, this is a "leaky" high IDV chip, scaling up with higher voltage is possible at a cost of runaway amp phenomena.
vmem ~2.2v for benching 2v on Vapo, compensate voltage with cooling to be safe.
vqpi ~1.35v depending on the board/bclock, upwards of 1.55v if I am pushing bclock, I run a low bclock so I usually leave it on stock.
vmem is the secret to uclk, 2.2v gets you 4.7Ghz ulck. uclk is also known as uncore.

general ratios

Stock QPI on i7 965 is x48 @ 133 bclock, if you raise the bclock to 160ish you would need to drop qpi ratio to X44 to continue raising the bclock to 180ish, drop qpi again to X36 again and that will get you out the theoretic max bclock of 220. That is not set in stone, some chips/board combos have exceeded these numbers but this rule is pretty close to what you will see on C stepping.

The uclk ratio I prefer to use is (the lowest possible) 1:2 of my ram speed (uclk runs at twice the speed of memory) and at lower speeds there are plenty more ratios to scale higher without running the sick memory speeds that I do, this is a huge gain in performance but for me it requires a lot of vmem and high end memory to run upwards of 2300Mhz (4700Mhz uclk) the memory controller has exceptional is its stability at this speed, I have not reached top speed and it will take even faster ram and more vmem to achieve that.

Runaway amp phenomena, I hook an amp meter to the four yellow wires on eight pin power cable that connects near the CPU. Watching this you will see that yes you can keep scaling up with more voltage but your chip has a limit with all cores loaded that it will runaway on amps till it crashes. I know exactly what range to stay in with cores loaded to know not to hit this runaway threshold. Keeping vcore down and disabling HT lets me run at the chips max speed, HT has an added power load on the CPU, disabling it helps gain critical Mhz.

[saaya]

Ill disclose the voltages I am running at on each board.cool, thanks :toast:

is more leaky better like with netburst or does it have no impact?

vdimm is needed for high uncore clocks, cause the imc is limiting, not the L3 cache or anything else. But whats the performance gain of 200mhz uncore clock more or less in your experience? it only helps for mem bandwidth, right?


Stock QPI on i7 965 is x48 @ 133 bclock, if you raise the bclock to 160ish you would need to drop qpi ratio to X44 to continue raising the bclock to 180ish, drop qpi again to X36 again and that will get you out the theoretic max bclock of 220. That is not set in stone, some chips/board combos have exceeded these numbers but this rule is pretty close to what you will see on C stepping.the actual qpi speed is half of he GT/s so the multipliers are 24, 22 and 18, but yeah... weird thing is intel told me lower than 18x qpi multis can be used, but if you read the register values for lowest qpi multiplier even on retail 965xe cpus it says lowest is 18x... :/


Runaway amp phenomena, I hook an amp meter to the four yellow wires on eight pin power cable that connects near the CPU. Watching this you will see that yes you can keep scaling up with more voltage but your chip has a limit with all cores loaded that it will runaway on amps till it crashes. I know exactly what range to stay in with cores loaded to know not to hit this runaway threshold. Keeping vcore down and disabling HT lets me run at the chips max speed, HT has an added power load on the CPU, disabling it helps gain critical Mhz.yeah, i wonder what actually happens inside the cpu at that point, when current suddenly increases more and more...

[LinusTech]

Hey Saaya,

Here's that video I said I would do up loosely based on your guide (tuned for beginners more than anything else). Hopefully you like :)

http://www.xtremesystems.org/forums/showthread.php?p=3445555#post3445555
(http://www.xtremesystems.org/forums/showthread.php?p=3445555#post3445555)
Linus

[Marco André]

Hey Saaya,

Here's that video I said I would do up loosely based on your guide (tuned for beginners more than anything else). Hopefully you like :)

http://www.xtremesystems.org/forums/showthread.php?p=3445555#post3445555
(http://www.xtremesystems.org/forums/showthread.php?p=3445555#post3445555)
Linus

Nice video but very superficial imo :up:

[Xello]

That's a great video, it might only go over the basics but this is what a lot of people will need just to get the ball rolling.

Well done mate :)

[saaya]

updated the first post with some more infos:

Once you boot at a certain Bclock you will only be able to bump it up by around 25Mhz before the board gets unstable.
Most likely its caused by the chipset and memory training during bootup, core i7 has a built in routine that automatically fine tunes the chipset and memory during bootup.

high uncore and memory clocks are possible without high vdimm, we hit ddr3 2200 with 1.7v vdimm on BloodRage a few days ago :D

[[EvQ]RheA]

2200 Mhz @ 1.7v is almost unconceivable :O

[boblemagnifique]

updated the first post with some more infos:

Once you boot at a certain Bclock you will only be able to bump it up by around 25Mhz before the board gets unstable.
Most likely its caused by the chipset and memory training during bootup, core i7 has a built in routine that automatically fine tunes the chipset and memory during bootup.

high uncore and memory clocks are possible without high vdimm, we hit ddr3 2200 with 1.7v vdimm on BloodRage a few days ago :D

Yes i have see the screen , impressive !!!

Chips Samsung (low voltage) ? or Micron D9JNL ? or Elpida Or .... :D

[msgclb]

Fr3ak asked this



What voltage do you refer to as VTT? Voltage of the integrated memory controller labeled as QPI/Vdimm on some boards?"


saaya your answer was this



only asus calls it that and its pretty confusing...
i think they call it that cause it helps clocking up qpi and memory...
but its not the voltage of the memory controller, its the entire uncore supply voltage.
it powers the memory controller AND the L3 cache and i think also a part of the qpi controller since more vtt helps to clock to higher qpi speeds.


Maybe Fr3ak understood your answer but I'm still confused. So on the Asus P6T Deluxe is the QPI/DRAM the VTT voltage that's so often mentioned? From your answer I got the impression it's ...:confused:

[saaya]

Yes i have see the screen , impressive !!!

Chips Samsung (low voltage) ? or Micron D9JNL ? or Elpida Or .... :D

i just checed and removed the heatspreader, its samsung! :D
K4B1G0846D



So on the Asus P6T Deluxe is the QPI/DRAM the VTT voltage that's so often mentioned? From your answer I got the impression it's ...:confused:

yes, qpi/dram voltage on asus boards is VTT, which is the uncore voltage.
they called it that way cause more vtt helps reaching higher memory and qpi clocks... somewhat...

thats very common, many engineers what a voltage actually is called like and give it their own name based on what they either KNOW or THINK it helps with... so if a certain voltage gets you higher pciE clocks some engineer might call it pciE voltage even though its actually the southbridge or IO voltage or PLL voltage or something else :)

[Harshal]

I have only done DDR3-2000 CL8 (A-Data) on P6T with 1.96V. VCORE and VDIMM was the only thing I played with.
Great therad so need to pick up few points and start again if intel allows me to keep chip for few more days :D

O/T: Where is my board saaya??? :P

[saaya]

right here, waiting for you to pick it up :D

[Oliver]

this board is for sure going to scream

cant wait to see some public numbers Sascha

keep pushing it

[boblemagnifique]

i just checed and removed the heatspreader, its samsung! :D
K4B1G0846D

Thank my friends :)

The chips samsung is good at low vddr on cas 8/9 (and a little cas 7) :)

Next Time for see the news Tests :up:

[Gendo]

right here, waiting for you to pick it up :D

And mine?:p:

[Amurtigress]

Hi there,

I am currently using a ASUS P5Q3 Del. board with 4 pcs. of Corsair DDR3-1600 CL9 RAM.

Hence my question for my upgrade to a nehalem system:

1) What memory modes except for triple channel are supported? Single channel? Dual? I am considering to run my system temporarily with 4x1 GB in dual channel mode if possible, and if that doesn't mean too much of a performance impact.

2) I read a bit on the memory controller/Memory module voltage not being separate yet and Intel recommending app. 1.65V for the RAM subsystem. My Corsairs are rated for 1.8V tho....any recommendations? I don't want to ditch my RAMs.

I'll see how the prices are developing around XMas...and then decide if I am changing or now. My Q9550 at 3.91 GHz is pretty neat so far.

[mike]

Best article I have read in a while here! SUre it'll get ripped left and right!
Thanks so much for your insight - helped me a lot!

[saaya]

1) What memory modes except for triple channel are supported? Single channel? Dual? I am considering to run my system temporarily with 4x1 GB in dual channel mode if possible, and if that doesn't mean too much of a performance impact.core i7 supports single dual and tripple channel.
each channel only has 2 slots though, while some server boards MIGHT have up to 4 per channel. running dual channel is perfectly fine, and actually shows slightly better performance in some scenarios than tripple channel! :D


2) I read a bit on the memory controller/Memory module voltage not being separate yet and Intel recommending app. 1.65V for the RAM subsystem. My Corsairs are rated for 1.8V tho....any recommendations? I don't want to ditch my RAMs.1.8v should be fine as long as you increase VTT as well.
this voltage is called QPI/Dram on asus and QPI/VTT on gigiabyte.
its the uncore voltage, and the memory controller is part of the cpu uncore.
so this is basically the chipset voltage... if you increase it, then it seems even memory voltages of up to 2v and higher are fine.
the rule of the thumb seems to be to keep vdimm and vtt within .5v of each other. so 1.3v vtt for 1.8v vdimm...
you will find that vtt can be as high as 1.5v without any problems for 24/7, and higher vtt means you can run higher memory clocks, so you might want to run more than 1.3v vtt.


I'll see how the prices are developing around XMas...and then decide if I am changing or now. My Q9550 at 3.91 GHz is pretty neat so far.Well, unless you have a really powerful vga, or even 2 or 3, or unless you really see that your maxed out by cpu power and you need more, for example for vide en/de coding or other cpu intensive tasks, then there is no reason to upgrade yet.

If you want the fastest of the fastes
if you have one or several powerful vgas
if you are a workstation user working with professional tools
if you are a virtualization fan
if you want to benchmark
if you want to play with a new plattform

then go for core i7


if you want to play games
if you are short on money
if you care about price/perf

then go for an E8400/E8500/E8600...


Best article I have read in a while here! SUre it'll get ripped left and right!
Thanks so much for your insight - helped me a lot!
as long as i get a single feedback like yours that what i did was helpful in some way, ill do my best to continue :toast:

[freshy]

merry christmas saaya all the best to you and your family :toast:

[Raptor8888]

yes, but only up to 21 with 4 cores and 8 threads, and 22 with 1 core and 2 threads...

@Saaya: GREAT THREAD. The best Nehalem writeup that I have seen on the web..

What steps are necessary to "fool" the 920 into X21 consistently with heavily threaded workloads?

[saaya]

thanks :toast:

to get 21x under any condition on BloodRage just set CPU Turbo to "Always On" in BIOS :D
I dont know if other boards have a similar feature... afaik they dont... maybe they will copy us again after some time and add that option :P

[Crankyhobo]

Thanks for the info.

Regarding gaming, a little offtrack but i wanted to comment since they brought it up.
From my experience their performance are mostly held back by:

1) GPU
2) HDD/DVD
3) CPU

And the conventional wisdom is that its better to get the next generation single card than SLI an older card. SLI in general only provides valuable gains at high resolution and very high anti-aliasing/filtering type settings. Install/Crack your games to the HDD so that you dont have to use the DVD, use an SSD or 10k+ RPM drive for the games you play most often. Watercool and overclock your graphics card(s).

[Amuro]

965XE goes up to around 3330mhz with turbo mode on 4 cores, and 3460mhz for single core.

[Touge180SX]

:shrug:
965XE goes up to around 3330mhz with turbo mode on 4 cores, and 3460mhz for single core.

What do you mean? That is the max you got with OC or running everything stock? :shrug:

[Amuro]

:shrug:

What do you mean? That is the max you got with OC or running everything stock? :shrug:
Stock. The OP made a mistake in post #1. 25 x 133mhz <> 3450 and 26x133mhz < > 3600

[Touge180SX]

Oh, got it! Sorry about that. :up:

[fubarswe]

.., keep returning to this thread.., very good stuff Saaya.. :-)

[Kanasatake]

For the second time..thanks Saaya for posting this and thank you to Linus for sharing your great video.... its nice to actually see the Bios in action rather than paragraph after paragraph..it can get confusing...


Merci, gneowan:up:

Hermes

[saaya]

Thanks for the info.

Regarding gaming, a little offtrack but i wanted to comment since they brought it up.
From my experience their performance are mostly held back by:

1) GPU
2) HDD/DVD
3) CPU

And the conventional wisdom is that its better to get the next generation single card than SLI an older card. SLI in general only provides valuable gains at high resolution and very high anti-aliasing/filtering type settings. Install/Crack your games to the HDD so that you dont have to use the DVD, use an SSD or 10k+ RPM drive for the games you play most often. Watercool and overclock your graphics card(s).

yeah, and get the last gen mainstream cpu or current gen entry level cpu and clock it up and get the latest high end card. thats the best for gaming, at least so far its been like that... if you do more than gaming on your pc, are a heavy multi tasker or want all apps to run and pop up quickly, then a fast hdd/ssd is good for you :D
top end cpus are only good for benching and professonal media tools, for gaming its really not needed.


965XE goes up to around 3330mhz with turbo mode on 4 cores, and 3460mhz for single core.oh yeah, my bad... :D
originally the max multi for single core was actually 27x on 965... at least on some ES cpus, but they apparently changed that for the retail chips.


.., keep returning to this thread.., very good stuff Saaya.. :-)thanks, its a bit outdated by now tho :D


For the second time..thanks Saaya for posting this and thank you to Linus for sharing your great video.... its nice to actually see the Bios in action rather than paragraph after paragraph..it can get confusing...
Merci, gneowan:up:

Hermes
it was only meant to introduce people to i7 before it was launched, i wanted to spill some beans so there wont be any bad rumors or misunderstandings about i7 spreading. some people still think qpi is the new name for fsb and think vtt is qpi voltage, but overall i think almost everybody understood i7 by now :D

this thread has inspired several i7 guides by now, which are all very nice and easier to read and understand and more complete, so it has served its purpose :D

anandtech wrote a guide in a similar style, xbitlabs has a nice guide and several other sites wrote good guides too :)

[dave510]

Thanks for posting this! I'm still nervous to touch OC settings. Still much more reading to do first!

[saaya]

theres not much that can go wrong really, just play around with the settings :D
i highly recommend the tool cpu tweaker btw, great to keep track of where the different clocks are at under windows :toast:

[dave510]

I just don't want to kill my i7 with too much voltage, etc. I'm only going to be looking for 24/7 stable. Probably won't be benching much.

[HDCHOPPER]

Saaya is the BloodRage considered a 2 ounce board ? (copper layer)

[Shibumi77]

Excellent link for first timers like me!! That video explained ALOT!!:up::up::up:

[saaya]

2 ounce copper? no, its 2 pounds of friggin platinum! :D

its great somebody tries something new and different and if they really use 2ounce on the power and ground layers inside the pcb and not on top, then they must even have tweaked their manufacturing process cause most pcb vendors cant do that. i havent heard of or seen any notable improvements though... all i saw was a questionable review at thg that concluded there was no difference and a more than questional review from asrock which showed it can make things worse in some situation...

i wish somebody would make a propper analysis and compare a 1ounce and 2ounce board of the same design like thg did.
until that happens and theres really an advantage im categorizing this as a marketing stunt and ignore it :D