Announcement

**Psycho101** · 05-09-2009, 09:05 AM

Re: G.Skill memory for EP45-UD3P

Vdrop is the difference in voltage set in the BIOS versus the voltage you see in CPU-Z.

Vdroop is the amount by which processor voltage "droops" when you load the CPU

It varies by motherboard, but usually the higher the voltages you set, the worse these two factors become, especially Vdroop. Load Line Calibration can virtually eliminate Vdroop, but often does nothing for Vdrop. Vdrop and Vdroop are both expected behaviours and are detailed in Intel spec sheets.

And 1.424v is absolutely nothing to worry about. If your temperatures are fine then that kind of voltage will not kill your CPU. Depending on cooling, 1.5-1.6v (high end air/water, up to phase change) would be fine.

**dclayw** · 05-09-2009, 06:59 PM

Re: G.Skill memory for EP45-UD3P

My cooling is good (Xigmatec S1283), during IntelBurnTest maximum stress gets both cores to 57 degrees. Needed 1.29375v in the BIOS to get to 3.78Ghz, actual (CPU-Z) is 1.248v and under load 1.232v so I guess vdroop is 0.016v. I've heard LLC is not good for 45nm precessors?, I don't know, I'm no expert. I have it disabled.

**kick** · 05-09-2009, 07:20 PM

Re: G.Skill memory for EP45-UD3P

the llc enabled "problem"is supposedly to do with earlier ep45 mbrds giving voltage spikes at lower cpu loadings.
In practice I go for whatever will hold the cpu closest to what I set it at in bios while cpu is loaded.
For me thats what I posted above.. I still get vdroop but not as extreme as without llc enabled.
Some people say you get better results at a few mhz diff with llc enabled or not,but hey if I changed every setting every time i tried a different speed,Id never actually get to use my pc

edit ive tried intel burn.psycho has too.
I got near 80C under load,indicated with 10 + passes for the cores but I dont know how much of a system check it is.It really seems to heat up the cpu and sounds scary,but feels more like"hey wanna see how hot your cpu will get and still run ok program"
nb while im typing this Im 50 min into an OCCT 1hr test
peak 69 core 0/overall cpu temp peak=57/system=41C/northbridge= 39C/gpu=46C//room temp showing 20 C/70 f.

yeh too much v is supposed to be the main component killer and then temperature.

test finished..ill try n save the graph readings.

**Psycho101** · 05-09-2009, 07:27 PM

Re: G.Skill memory for EP45-UD3P

LLC doesn't really do much harm. As a rule, if you can get a stable CPU without using it, up to 1.4v, don't use it. If you can't get stable at 1.4v, try 1.3500v and upward with it on. It would be really pointless to use if you can get stable without it on 1.3625v or less. Though you do see a fair few with it on, and running something like 1.2xxv.

As is usual with the internet, people seem to think in black and white. Once something is mentioned once, it can be taken as fact, and is then parroted by everyone on all sorts of forums and articles. things like "vdroop is evil, eliminate it before it assaults your wife and eats your babies", and the new in vogue "Using LLC or more than 1.4v CPU Termination will instantly and magically cause your CPU to degrade rapidly or even blow up".

As usual, the truth is somewhere in the middle.

You're much more likely to kill a CPU with high CPU Termination voltage than with LLC, that is true but I really do wonder who plucked the "No more than 1.4v VTT" thing out of their arse though. So anandtech killed a CPU... who cares, they must have been giving it some monster voltages all round for that to happen. If you need it to have the clock speed you want, and temperatures don't exceed 70c then it's all good, as long as you're sensible and don't go too far.

**Capt Proton** · 05-09-2009, 08:26 PM

Re: G.Skill memory for EP45-UD3P

I had a E8400 that started to lose it's overclock with 1.4+. I know one incident isn't a trend, but, I've read others posts who reported the same thing. It is not "out of someone's arse"

**dclayw** · 05-09-2009, 08:32 PM

Re: G.Skill memory for EP45-UD3P

Here are my current settings. All stress testing looks OK; Prime95 small FFTs 13.5 hours stable, Prime95 Blend 12 hours stable, IntelBurnTest 40 passes stable; but I still get memtest86+ errors. Simple answer is dodgy memory but it could be a memory timing or some other issue. As you can see I have all the subtimings on Auto. I have also included the actual values these get set to (see memset pic).

Code:

Robust Graphics Booster ...............: [Auto]
CPU Clock Ratio .......................: [ 8]
Fine CPU Clock Ratio...................: [+0.5]
CPU Frequency .........................: 3.78GHz (445x8.5)
 
Clock Chip Control 
Standard Clock Control
CPU Host Clock Control.................: [Enabled] 
CPU Host Frequency (Mhz) ..............: [445Mhz]
PCI Express Frequency (Mhz) ...........: [100]
 
C.I.A.2 ...............................: [Disabled]
 
Advanced Clock Control [Press Enter]
CPU Clock Drive........................: [800mV]
PCI Express Clock Drive................: [900mV]
CPU Clock Skew (ps)....................:   [0ps]
MCH Clock Skew (ps)....................:   [0ps]
 
DRAM Performance Control
Performance Enhance....................: [STANDARD]
Extreme Memory Profile (X.M.P.)........: [Disabled]
(G)MCH Frequency Latch.................: [Auto]
System Memory Multiplier ..............: [2.40B] 
Memory Frequency (Mhz) ................: [1068]
DRAM Timing Selectable ................: [Manual]
 
Standard Timing Control
CAS Latency Time.......................:  [5]
tRCD ..................................:  [5]
tRP....................................:  [5]
tRAS...................................:  [15]
 
Advanced Timing Control
tRRD...................................:  [Auto]
tWTR...................................:  [Auto]
tWR....................................:  [Auto]
tRFC...................................:  [Auto]
tRTP...................................:  [Auto]
Command Rate (CMD) ....................:  [Auto]
 
Driving Strength Profiles
Driving Strength ......................: [1066]
 
Channel A
Static tRead Value.....................: [Auto]
tRD Phase0 Adjustment..................: [Auto]
tRD Phase1 Adjustment..................: [Auto]
tRD Phase2 Adjustment .................: [Auto]
tRD Phase3 Adjustment..................: [Auto]
 
Trd2rd(Different Rank).................: [Auto]
Twr2wr(Different Rank).................: [Auto]
Twr2rd(Different Rank).................: [Auto]
Trd2wr(Same/Diff Rank).................: [Auto]
 
Dimm1 Clock Skew Control...............: [Auto]
Dimm2 Clock Skew Control...............: [Auto]
DDR Write Training.....................: [Auto]
 
Channel B
Static tRead Value.....................: [Auto]
tRD Phase0 Adjustment..................: [Auto]
tRD Phase1 Adjustment..................: [Auto]
tRD Phase2 Adjustment .................: [Auto]
tRD Phase3 Adjustment..................: [Auto]
 
Trd2rd(Different Rank).................: [Auto]
Twr2wr(Different Rank).................: [Auto]
Twr2rd(Different Rank).................: [Auto]
Trd2wr(Same/Diff Rank).................: [Auto]
 
Dimm1 Clock Skew Control...............: [Auto]
Dimm2 Clock Skew Control...............: [Auto]
DDR Write Training.....................: [Auto]
 
Motherboard Voltage Control
 
CPU---------------------Normal-----------Current
 
Load Line Calibration..................: [Disabled]
CPU Vcore...............1.30000V.......: [1.29375V]
CPU Termination.........1.200V.........: [1.220V]
CPU PLL.................1.500V.........: [1.500V]
CPU Reference...........0.760V.........: [0.772V]  
 
MCH/ICH
MCH Core................1.100V.........: [1.240V]
MCH Reference...........0.760V.........: [0.772V]
MCH/DRAM Reference......0.900V.........: [0.900V]
ICH I/O.................1.500V.........: [1.500V]
ICH Core................1.100V.........: [1.100V]
 
DRAM
DRAM Voltage............1.800V.........: [1.900V ]
DRAM Termination........0.900V.........: [0.900V]
Channel A Reference.....0.900V.........: [0.900V]
Channel B Reference.....0.900V.........: [0.900V]
 
Advanced Settings
Limit CPUID Max. to 3..................: [Disabled]
No-Execute Memory Protect..............: [Enabled]
CPU Enhanced Halt (C1E)................: [Enabled]
C2/C2E State Support...................: [Disabled]
x C4/C4E State Support.................: [Disabled]
CPU Thermal Monitor 2(TM2) ............: [Enabled]
CPU EIST Function......................: [Disabled]
Virtualization Technology..............: [Disabled]
 
Integrated Peripherals
USB Storage Fucntion .............: [Disabled]

Attached Files

untitled.JPG (0, 127 views)

**Psycho101** · 05-09-2009, 10:45 PM

Re: G.Skill memory for EP45-UD3P

I stand by my statement. It most certainly, in its usual form, which is over hyped and exagerated: "pulled out of their arse". I have personally experienced no CPU degredation.

there have been cases where after a certain period of time, people have had to give a chip more voltage to maintain a certain core speed. Now is that the CPU? Is it a deterioration of motherboard components? Is it merely that someone set a large overclock in the winter with lower ambient temperatures, and come along with summer came instability.

These such variables are never mentioned. It's usually pure conjecture and an over excited overclocker shouting "Aaaagh! YOOT-WANID" (It Used To Work And Now It Doesn't). And when someone reasonably suggests that an increase in CPU temperature can cause a lesser overclock, even if that increase means the processor temp is still in the "safe zone", they still insist that the degredation fairy definately did it, because they saw it fleeing the scene.

The odds of your processor going **** up are worse than someone winning the lottery, and the brittish lottery jackpot odds are 14 million to 1.

And by the way, I'm talking reasonable voltages here, not someone using a five stage phase change cascade and pumping 1.8v through the thing. If their chip "degrades" then it's their own fault, and I'm sure they fully understand the risks involved anyway.

**Bee_Dee_3_Dee** · 05-10-2009, 03:19 AM

Re: G.Skill memory for EP45-UD3P

Originally posted by dclayw View Post

Here are my current settings. All stress testing looks OK; Prime95 small FFTs 13.5 hours stable, Prime95 Blend 12 hours stable, IntelBurnTest 40 passes stable; but I still get memtest86+ errors. Simple answer is dodgy memory but it could be a memory timing or some other issue. As you can see I have all the subtimings on Auto. I have also included the actual values these get set to (see memset pic).

Code:

Robust Graphics Booster ...............: [Auto]
CPU Clock Ratio .......................: [ 8]
Fine CPU Clock Ratio...................: [+0.5]
CPU Frequency .........................: 3.78GHz (445x8.5)
 
Clock Chip Control 
Standard Clock Control
CPU Host Clock Control.................: [Enabled] 
CPU Host Frequency (Mhz) ..............: [445Mhz]
PCI Express Frequency (Mhz) ...........: [100]
 
C.I.A.2 ...............................: [Disabled]
 
Advanced Clock Control [Press Enter]
CPU Clock Drive........................: [800mV]
PCI Express Clock Drive................: [900mV]
CPU Clock Skew (ps)....................:   [0ps]
MCH Clock Skew (ps)....................:   [0ps]
 
DRAM Performance Control
Performance Enhance....................: [STANDARD]
Extreme Memory Profile (X.M.P.)........: [Disabled]
(G)MCH Frequency Latch.................: [Auto]
System Memory Multiplier ..............: [2.40B] 
Memory Frequency (Mhz) ................: [1068]
DRAM Timing Selectable ................: [Manual]
 
Standard Timing Control
CAS Latency Time.......................:  [5]
tRCD ..................................:  [5]
tRP....................................:  [5]
tRAS...................................:  [15]
 
Advanced Timing Control
tRRD...................................:  [Auto]
tWTR...................................:  [Auto]
tWR....................................:  [Auto]
tRFC...................................:  [Auto]
tRTP...................................:  [Auto]
Command Rate (CMD) ....................:  [Auto]
 
Driving Strength Profiles
Driving Strength ......................: [1066]
 
Channel A
Static tRead Value.....................: [Auto]
tRD Phase0 Adjustment..................: [Auto]
tRD Phase1 Adjustment..................: [Auto]
tRD Phase2 Adjustment .................: [Auto]
tRD Phase3 Adjustment..................: [Auto]
 
Trd2rd(Different Rank).................: [Auto]
Twr2wr(Different Rank).................: [Auto]
Twr2rd(Different Rank).................: [Auto]
Trd2wr(Same/Diff Rank).................: [Auto]
 
Dimm1 Clock Skew Control...............: [Auto]
Dimm2 Clock Skew Control...............: [Auto]
DDR Write Training.....................: [Auto]
 
Channel B
Static tRead Value.....................: [Auto]
tRD Phase0 Adjustment..................: [Auto]
tRD Phase1 Adjustment..................: [Auto]
tRD Phase2 Adjustment .................: [Auto]
tRD Phase3 Adjustment..................: [Auto]
 
Trd2rd(Different Rank).................: [Auto]
Twr2wr(Different Rank).................: [Auto]
Twr2rd(Different Rank).................: [Auto]
Trd2wr(Same/Diff Rank).................: [Auto]
 
Dimm1 Clock Skew Control...............: [Auto]
Dimm2 Clock Skew Control...............: [Auto]
DDR Write Training.....................: [Auto]
 
Motherboard Voltage Control
 
CPU---------------------Normal-----------Current
 
Load Line Calibration..................: [Disabled]
CPU Vcore...............1.30000V.......: [1.29375V]
CPU Termination.........1.200V.........: [1.220V]
CPU PLL.................1.500V.........: [1.500V]
CPU Reference...........0.760V.........: [0.772V]  
 
MCH/ICH
MCH Core................1.100V.........: [1.240V]
MCH Reference...........0.760V.........: [0.772V]
MCH/DRAM Reference......0.900V.........: [0.900V]
ICH I/O.................1.500V.........: [1.500V]
ICH Core................1.100V.........: [1.100V]
 
DRAM
DRAM Voltage............1.800V.........: [1.900V ]
DRAM Termination........0.900V.........: [0.900V]
Channel A Reference.....0.900V.........: [0.900V]
Channel B Reference.....0.900V.........: [0.900V]
 
Advanced Settings
Limit CPUID Max. to 3..................: [Disabled]
No-Execute Memory Protect..............: [Enabled]
CPU Enhanced Halt (C1E)................: [Enabled]
C2/C2E State Support...................: [Disabled]
x C4/C4E State Support.................: [Disabled]
CPU Thermal Monitor 2(TM2) ............: [Enabled]
CPU EIST Function......................: [Disabled]
Virtualization Technology..............: [Disabled]
 
Integrated Peripherals
USB Storage Fucntion .............: [Disabled]

[ATTACH]1063[/ATTACH]

Try Disabling CPU Enhanced Halt (C1E). After that, bump up CPU voltage another +.01V.

**kick** · 05-10-2009, 06:19 AM

Re: G.Skill memory for EP45-UD3P

No real experience of o/c dual cores(aside from my D820 which is more 2 p4's stuck together and no seperate core readings).

Is 1.30v the processor VID?
Seems stable enough altho the memory secondary timings do seem very "loose".
Who cares on that tho if the bandwidth is good. Did you do any bandwidth tests/comparisons with the memory at different settings?

**dclayw** · 05-10-2009, 07:28 AM

Re: G.Skill memory for EP45-UD3P

Originally posted by Bee_Dee_3_Dee View Post

Try Disabling CPU Enhanced Halt (C1E). After that, bump up CPU voltage another +.01V.

Thanks for the suggestion, I'll give that a go.

**dclayw** · 05-10-2009, 07:45 AM

Re: G.Skill memory for EP45-UD3P

Originally posted by kick View Post

No real experience of o/c dual cores(aside from my D820 which is more 2 p4's stuck together and no seperate core readings).

Is 1.30v the processor VID?
Seems stable enough altho the memory secondary timings do seem very "loose".
Who cares on that tho if the bandwidth is good. Did you do any bandwidth tests/comparisons with the memory at different settings?

Core VID as shown by RealTemp is min 1.10 and max 1.30. At stock settings (everything on AUTO) but with speedstep and all that stuff turned off the actual core voltage (CPU-Z) is 1.28v. No bandwidth testing. Secondary timings are lose cause that's what they get set to when on AUTO, but still getting memtest errors even with those loose timings so no point in tweaking them just yet, but I expect it's probably the timings I need to tweak to get rid of the memtest errors, I dunno.

To be honest I don't think I have run a memtest out of the box, I probably should have done that as I'm trying to rule out faulty memory or memory compatability problem. I will run memtest o'night with everything at stock settings. I have seen a few people around with this board, memory and cpu, but as they say every board is different.

**dclayw** · 05-16-2009, 03:44 AM

Re: G.Skill memory for EP45-UD3P

Well, again after much testing, below are the settings I've settled on. My main aim was to simply get 3.6GHz with the memory running at or as close as possible to it's rated speed of 1100Mhz. So I was trying 450FSB with 2.40B strap (1080Mhz). (I've also tried various settings around 450, e.g. 444, 449 etc.) Seriously, (within safe voltage limits) I tried pretty much every combination of MCH Core/CPU Termination/CPU PLL/DDR Voltage, including low values and high values. Every single combination gives memtest errors. As mentioned in earlier posts, nothing and I mean nothing, works on the D strap. So assuming the memory isn't faulty (time will tell but I don't think it is), then this board/memory combination is exceedingly picky. As Lsdmeasap said, P45 is a tricky beast indeed.

I don't know why I am getting memtest errors with almost any configuration, I know, it suggests bad memory, but the few configurations that don't give memtest errors seem perfectly stable with Prime95 testing. Again, I guess time will tell.

All testing was done at 5 5 5 15 (spec) and I repeated all testing with and without loose sub-timings. When testing the various MCH Core/CPU Term. combinations, I tested by using default values for the reference voltages (default, not AUTO) and I also tested by using the generally accepted method of setting reference voltages for this board/cpu, i.e. with CPU Termination at 1.20v leave CPU reference at 0.76v and set MCH reference to around 0.66-0.70v, then let them adjust automatically as CPU Term. is adjusted.

So you can see below I've had to settle on FSB 400 with 2.40B, running my 1100Mhz memory at 960Mhz. I'm not so much concerned at running the memory at 140Mhz under spec, more concerned about the hassle I had to go through to even get to this stage and why I'm getting memtest errors. I'm open to suggestions as to what I might be able to try from here on. At least voltages are fairly low across the board and timings not too bad.

Here are my current settings. I get no memtest86+ errors with these settings (one of the very combinations where that is the case). IntelBurnTest is OK on these settings at 30 passes (several times). Prime95 blend 17 hours with no errors, Prime95 small FFTs 12 hours with no errors.

Code:

Robust Graphics Booster ...............: [Auto]
CPU Clock Ratio .......................: [ 9]
Fine CPU Clock Ratio...................: [+0.0]
CPU Frequency .........................: 3.60GHz (400*9)
 
Clock Chip Control 
Standard Clock Control
CPU Host Clock Control.................: [Enabled] 
CPU Host Frequency (Mhz) ..............: [400Mhz]
PCI Express Frequency (Mhz) ...........: [100]
 
C.I.A.2 ...............................: [Disabled]
 
Advanced Clock Control [Press Enter]
CPU Clock Drive........................: [800mV]
PCI Express Clock Drive................: [900mV]
CPU Clock Skew (ps)....................:   [0ps]
MCH Clock Skew (ps)....................:   [0ps]
 
DRAM Performance Control
Performance Enhance....................: [STANDARD]
Extreme Memory Profile (X.M.P.)........: [Disabled]
(G)MCH Frequency Latch.................: [Auto]
System Memory Multiplier ..............: [2.40B] 
Memory Frequency (Mhz) ................: [960]
DRAM Timing Selectable ................: [Manual]
 
Standard Timing Control
CAS Latency Time.......................:  [5]
tRCD ..................................:  [5]
tRP....................................:  [5]
tRAS...................................:  [15]
 
Advanced Timing Control
tRRD...................................:  [3]
tWTR...................................:  [3]
tWR....................................:  [4]
tRFC...................................:  [48]
tRTP...................................:  [3]
Command Rate (CMD) ....................:  [2]
 
Driving Strength Profiles
Driving Strength ......................: [1066]
 
Channel A
Static tRead Value.....................: [8]
tRD Phase0 Adjustment..................: [Auto]
tRD Phase1 Adjustment..................: [Auto]
tRD Phase2 Adjustment .................: [Auto]
tRD Phase3 Adjustment..................: [Auto]
 
Trd2rd(Different Rank).................: [Auto]
Twr2wr(Different Rank).................: [Auto]
Twr2rd(Different Rank).................: [Auto]
Trd2wr(Same/Diff Rank).................: [Auto]
 
Dimm1 Clock Skew Control...............: [Auto]
Dimm2 Clock Skew Control...............: [Auto]
DDR Write Training.....................: [Auto]
 
Channel B
Static tRead Value.....................: [8]
tRD Phase0 Adjustment..................: [Auto]
tRD Phase1 Adjustment..................: [Auto]
tRD Phase2 Adjustment .................: [Auto]
tRD Phase3 Adjustment..................: [Auto]
 
Trd2rd(Different Rank).................: [Auto]
Twr2wr(Different Rank).................: [Auto]
Twr2rd(Different Rank).................: [Auto]
Trd2wr(Same/Diff Rank).................: [Auto]
 
Dimm1 Clock Skew Control...............: [Auto]
Dimm2 Clock Skew Control...............: [Auto]
DDR Write Training.....................: [Auto]
 
Motherboard Voltage Control
 
CPU---------------------Normal-----------Current
 
Load Line Calibration..................: [Disabled]
CPU Vcore...............1.30000V.......: [1.250V]
CPU Termination.........1.200V.........: [1.220V]
CPU PLL.................1.500V.........: [1.500V]
CPU Reference...........0.760V.........: [0.772V]  
 
MCH/ICH
MCH Core................1.100V.........: [1.220V]
MCH Reference...........0.760V.........: [0.772V]
MCH/DRAM Reference......0.900V.........: [0.850V]
ICH I/O.................1.500V.........: [1.500V]
ICH Core................1.100V.........: [1.100V]
 
DRAM
DRAM Voltage............1.800V.........: [1.800V ]
DRAM Termination........0.900V.........: [0.850V]
Channel A Reference.....0.900V.........: [0.850V]
Channel B Reference.....0.900V.........: [0.850V]
 
Advanced Settings
Limit CPUID Max. to 3..................: [Disabled]
No-Execute Memory Protect..............: [Enabled]
CPU Enhanced Halt (C1E)................: [Enabled]
C2/C2E State Support...................: [Disabled]
x C4/C4E State Support.................: [Disabled]
CPU Thermal Monitor 2(TM2) ............: [Enabled]
CPU EIST Function......................: [Disabled]
Virtualization Technology..............: [Disabled]
 
Integrated Peripherals
USB Storage Function .............: [Disabled]

**profJim** · 05-16-2009, 05:21 AM

Re: G.Skill memory for EP45-UD3P

Which version of Memtest have you been running?
Some have reported problems with the 3.x versions.
Memtest86+ (version 2.11) is solid.

**dclayw** · 05-16-2009, 05:36 AM

Re: G.Skill memory for EP45-UD3P

I was running with memtest 86+ 2.01 initilally, then swapped to 2.11, didn't make any difference.

**profJim** · 05-16-2009, 06:35 AM

Re: G.Skill memory for EP45-UD3P

The obvious test (previously mentioned) would be to swap memory with a friend and see if the other memory works better in your system and if your friend has any issues with your memory.

This is a long shot, but I've been running my memory for the past month using slots 2 & 4. I spent a week testing with this configuration to assist a poster that had different issues with possible bad memory slots. It turned out that his UD3P was defective. While my current memory setup isn't recommended, I've had absolutely no problems using the #2 & #4 memory slots.

I wouldn't do a lot of testing using these memory slots, just a couple of tests using conservative settings that failed in your earlier tests.

Announcement

G.Skill memory for EP45-UD3P

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment