new internal bios in da house
http://www.abload.de/img/zwischenablage01c4bi.jpg
this bios fixed an Q-Fan issue with R3E and has Gulftown improvements
Printable View
new internal bios in da house
http://www.abload.de/img/zwischenablage01c4bi.jpg
this bios fixed an Q-Fan issue with R3E and has Gulftown improvements
No, those are the newer (PSC based I believe) kits. I have the Hyper MGH-E kit, CMG6GX3M3A2000C8 + extended fins.
Sure I don't mind. You have to know though that every cpu is different so voltages will vary. And if you are trying to stay on a 2:10 memory multi then you will need to hit atleast 200BCLK. And on top of all that, if you have two sets coming, I'm assuming you're trying to run 12gb. Expect to either run a lot of voltage to get them stable or run into problems. Key is to not try and skip ahead. Start from stock and you'll get a better understanding of what's going on. My settings are:
Code:Target CPU Frequency 3819MHz
Target DRAM Frequency 2015MHz
LN2 Mode: Disabled***
QPI Loadline Calibration: Enabled***
Sync Mode [Enabled]
AI Overclock Tuner [Manual]
OC From CPU level Up [Auto]
CPU Ratio Setting [19.0]
CPU Turbo Power Limit [Disabled]
>CPU Configuration----------------------------
CPU Ratio Setting [19.0]
C1E Support [Disabled]
Hardware Prefetcher [Enabled]
Adjacent Cache Line Prefetcher [Enabled]
Intel(R) Virtualiyation Tech [Enabled]
CPU TM Function [Disabled]
Execute Disable Bit [Enabled]
Intel(R) HT Technology [Enabled]
Active Processor Cores [All]
A20M [Disabled]
Intel(R) SpeedStep(TM) Tech [Disabled]
Intel(R) C-STATE Tech [Enabled]
C State package limit setting [Auto]
----------------------------------------------
BCLK Frequency [201]
PCIE Frequency [100]
DRAM Frequency [2015]
UCLK Frequency [4029]
QPI Link Data Rate [7254]
Memory Configuration Protect [Disabled]
>DRAM Timing Control--------------------------
CAS# Latency 9 [7]
RAS# to CAS# Delay 9 [7]
RAS# PRE Time 9 [7]
RAS# ACT Time 24 [20]
RAS# to RAS# Delay 6 [AUTO]
REF Cycle Time 110 [110]
WRITE Recovery Time 15 [15]
READ to PRE Time 9 [9]
FOUR ACT WIN Time 31 [31]
Back-To-BackCAS# Delay 0 [Auto]
Timing Mode 1N [1N]
Round Trip Latency on CHA [AUTO]
Round Trip Latency on CHB [AUTO]
Round Trip Latency on CHC [AUTO]
WRITE To READ Delay(DD) 8 [Auto]
WRITE To READ Delay(DR) 8 [Auto]
WRITE To READ Delay(SR) 18 [Auto]
READ To WRITE Delay(DD) 7 [Auto]
READ To WRITE Delay(DR) 7 [Auto]
READ To WRITE Delay(SR) 7 [Auto]
READ To READ Delay(DD) 7 [Auto]
READ To READ Delay(DR) 6 [Auto]
READ To READ Delay(SR) 4 [Auto]
WRITE To WRITE Delay(DD) 8 [Auto]
WRITE To WRITE Delay(DR) 8 [Auto]
WRITE To WRITE Delay(SR) 18 [Auto]
----------------------------------------------
CPU Differential Amplitude [Auto]
CPU Clock Skew [Auto]
IOH Clock Skew [Auto]
------------ Extreme Engine Digi+ ------------
Digi+ PWR Mode [T-Balanced]
PWR Volt. Control [Auto]
Load-Line Calibration [Full]
CPU Voltege OCP [Enabled]
CPU PWM Frequency [AUTO]
Extreme OV [Disabled]
Extreme OC [Auto]
CPU Voltage 1.250 [1.13125]
CPU PLL Voltage 1.812 [1.80200]
QPI/DRAM Core Voltage 1.350 [1.36250]
DRAM Bus Voltage 1.654 [1.65625]
>DRAM REF Voltages----------------------------
DRAM DATA REF Voltage on CHA [+0mV]
DRAM CTRL REF Voltage on CHA [+0mV]
DRAM DATA REF Voltage on CHB [+0mV]
DRAM CTRL REF Voltage on CHB [+0mV]
DRAM DATA REF Voltage on CHC [+0mV]
DRAM CTRL REF Voltage on CHC [+0mV]
----------------------------------------------
IOH Voltage 1.244 [1.11300]
IOH PCIE Voltage 1.508 [1.51050]
ICH Voltage 1.111 [1.11300]
ICH PCIE Voltage [1.51050]
---------- Spread Spectrum Control -----------
CPU Spread Spectrum [Disabled]
DRAM Spread Spectrum [Disabled]
***on-board jumper dependant
Yea they are good looking but probably not as effective as say the Corsair or Kingston design, with the added direct air flow from the optional fan (available for kingston as well).
What kind of timings have you got out of them? I also have a set thats why I ask
hum spoke too soon...i moved the gfx card from slot 2 to slot 1 and now it's unstable again. it was stable at 4.2 at only 1.3125 with 50% LLC, but now it's not stable even at 1.3250 with 50% LLC...weird. failing Linx after a few runs. anyone know if moving the gfx card would have affected things?
I posted the original message in this thread. Since I posted the message I've been through 3 further Rampage III Extreme boards for a total of 4 boards and all of them have malfunctioned!. The first 3 boards were connected to a PC Power & Cooling Turbo-Cool 1KW-SR 1000 watt PSU. I came to the conclusion that the PSU is defective and that's what is frying the boards. So I sent the PSU back to the manufacturer RMA and they sent me a refurbished one. The PC Power & Cooling technician that I talked to on the phone told me that they would test the unit that I returned to them and when I called a few days later to find out if the unit had tested defective they told me that the unit tested OK, There was nothing wrong with it. I still had suspicions about that PSU, Perhaps there was a incompatibility between this PSU and the motherboard because the Turbo-Cool 1KW-SR PSU is a 4 year old design . So I decided to buy a brand new Corsair HX1000 PSU which is on the Asus Qualified Vendors List (QVL) for 500 watt power supply and above that have been tested by Asus to work with this board, which is in the reference section in the back of the motherboard manual, Page A5.
I Paid $250 for it, but if it solves my problem I'll just keep going and not look back. When I got the 4th RMA RIIIE it tested ok outside and inside the case so I thought "At last, I've found the problem . So I just went ahead and started completing my build however I decided to power up just before I finished to make sure that everything installed so far was working properly and when I did NOTHING happened. The board wouldn't POST. i got the red light on the power on button on the board and the reset button light and in the back of the board in the I/O panel, the reset button light back there came on and on the bottom of the board the LED for the 1st BIOS came on after about 10 seconds. but no power to the CPU or fans so I'm sitting there scratching my heard trying to figure out what is wrong now. I decided that it must be the CPU although I really doubted that, I decided to RMA the CPU, an Intel I7 930 and try another one and when I got the replacement and tried it still nothing. I can't think of what else to try next.
I've tried 3 different PSU's two different CPU's. I only bench test with 1 memory module installed in the red socket closet to the CPU socket and both EPS12V power CPU Power lead connected and both EZ Molex plugs connected. I decided to send the board back to Asus in Fremont CA and let them figure out what is wrong with this motherboard. I didn't want another replacement, I just told them figure out what's wrong with the motherboard and send it back to me when it's fixed or send me a check for $429.00. The price I paid for it from Newegg.com. I have built many computers over the 15 years that I've been doing this but I have never had so many problems with a motherboard than I've had with this one. I'm a CompTIA A+ certified IT Technician but I'm stumped with this one. I'm about ready to pull the trigger on a EVGA Classified 3 way. I wish I didn't have to spend the extra money but I've already spent an extra $300 on top of what I paid for it just trying to get the Rampage III Extreme to work to no avail. Anybody got any ideas PLEASE let me know.
^^What kind of case do you have, or are you even running one? There are several reported cases of problems with certain cases, and even particular CPU heatsinks (i believe 1 of them was the VenomousX).
Sounds like its time to move on to something new, the Classified seems to be a great board. You have more patience than I would in that situation.
That 2.xx is coming from the BIOS. I run Linux on my box and see the 2.xx in the output from dmidecode
Code:Handle 0x0002, DMI type 2, 15 bytes
Base Board Information
Manufacturer: ASUSTeK Computer INC.
Product Name: Rampage III Extreme
Version: Rev 2.xx
Hi Guys
I'm hoping that someone can help me with what ought (I think) to be an elementary problem. I don't seem to be able to get my R3E/980X to run in turbo mode. I thought it was meant to work without having to do anything special but no matter what combination of BIOS settings I use, I seem to be stuck at a maximum of 3334MHz.
I'm using BIOS 9124 and running Linux, kernel 2.6.34, with a Core i7 980X on an R3E with 6 x 2GB RAM. I don't have access to the BIOS settings while typing this but I have tried various combinations of the Turbo Mode setting together with the two that the help on the right hand side of the screen say must be disabled (is it C1E and TM?).
I have no great wish to overclock but I would like to get turbo mode functional if at all possible. I check the speeds by running a single copy of `openssl speed` then
$ grep MHz /proc/cpuinfo
cpu MHz : 1600.000
cpu MHz : 1600.000
cpu MHz : 1600.000
cpu MHz : 1600.000
cpu MHz : 3334.000
cpu MHz : 1600.000
Is there some operating system dependency on turbo mode? I thought it was purely hardware and that using only a single core would boost it to 3.6GHz.
I have the board working fine in the Corsair case with both the PC Power & Cooling unit and the HX1000. Based on your description and that of your post at the ASUS forum talking about smoke or sparks I would venture to say that board is shorting out against something in the case. PM me your RMA case number and I will have someone contact you shortly about it.
Update -
Rampage III Extreme
Beta BIOS 0010 -
1. Fix Fan Ramp Issue with previous Beta
http://www.mediafire.com/file/c22yya...-ASUS-0010.rar
The only version out is 1.04G, there is not a 2.xxx version of the board. That number being reported is based on CPU Microcode. As far as one board clocking better than another one that is due to a whole host of factors but as far as the boards are concerned, the most important difference is the quality of the IOH. The difference between a really good IOH and an average one is easily worth 10 Bclk.
Not part of this but another myth is that the board is tuned more for Gulftown than Bloomfield and that is false. The BIOS microcode takes care of the base settings for either CPU. The difference is that each processor family requires different tuning methods to reach certain levels.
I have been contemplating whether to test out UD7 v2 and see how it does vs R3E for bloomfield since some people think RE3 is no good for none other than GT processors.
Overall I think the RE3 did quite well for 24/7 oc on my bloomfield :shrug::)
Here is my new board
This is REV.1.04G.
http://namegt.com/570
http://cfile9.uf.tistory.com/image/1...4C69DF3632F75B
http://cfile2.uf.tistory.com/image/1...4C6CC65699FD8E
well i consider the rexIII not as good as a rexII for bloomfeild
not even close;)
both my classy 760 kickin its ass bigtime
i'll wait another week for a descent bloomfeild bios
hey Bingo any hope for bloomfeild?!!
i dont give a crap for subzero
i wonder if asus even cares
asus cares:up:
they all work and improving ...
Don't hold out for a BIOS that makes Bloomfiled or Gulftown clock better. All of the factors that affect final clock speeds barring the basics of power delivery and memory sub-timings are automated (or in hardware, such as trace impedances), so the BIOS will not play a role in extending clocking margins past those factors at this point.
i got 3 920s here
1 will do 4.77 linx stable on rexII and e760
my best 920 maxed out 4.6 slow mode on rex III:rolleyes: i even tried 3x1 hypers
i dont need extra special clocking bios just a bois that will function properly on bloomfeid i dont give a crap for gulftown i sold my chip
i dont a crap what the board will do zubzero;) any cheap board will clock subzero
i'll even accept a bios as good as the rampage II gene :rofl:
if we not gonna get a good bios for gulftown i'd like to know and just dump the board
As I said above, a "magical" BIOS will NOT improve clocking margins (assuming the board has been set up correctly by the end user), you are wasting your time.
What you speak of is a hardware level limit, it won't change with a BIOS "that works properly for Bloomfield.". It's a myth that there is some kind of special Bloomfield or Gulftown tweak that makes the board lean in one direction.
You get a bad IOH and there's nothing you can do on boards like this in software (BIOS).
If you are unhappy, cut your losses and sell it. Best advice.
-Raja
Maybe you're already aware of this but I just want to make sure. If you're going to play with RAM voltages
NEVER let the difference between QPI/DRAM voltage and DRAM Bus voltage be more than 0.5v
You will Kill your CPU.
Ok, is this the BIOS that incorporates all the improvements made to BIOS 0002 and 0878 ??
This is from the other thread:
Beta BIOS 0010 -
1. Fix Fan Ramp Issue with previous Beta 0002/0009
2. Memory Table Fine Tuning
This board clocks Bloomfields just as well as it does Gulftowns, it just requires different tuning on the QPI/VTT side. This board requires different settings on Bloomfield than it does on the RIIE or other boards due to the PWM design. My 920 and 980 both happily run at 230 Bclk on air and once things get colder the Bclk rates certainly go higher.
The differences between boards at times come down to the quality of the IOH. Sometimes you get a really good one and sometimes you might get one that is average. The other problem is that if you ran your Bloomfield on a Gigabyte X58-Uxx series board then chances are you will need a lot more VTT as those boards had some problems that damaged processors (just do a search here and you will find the thread on it).
http://www.xtremesystems.org/forums/...d.php?t=256934 - One example of many with Bloomfields hitting high Bclks.
@bingo
i have Rampage III Extreme and beta bios 1003
why your newest bios are 0010 and 0878??
must an newer bios not named as 1004-1005-xxxx
can you tell me why this is so??
What's meaning of based on CPU microcode?
980X has some CPU microcode?:(
It's very strange what CPU-Z rev based on CPU microcode.
If so that 980X has some CPU microcode at 1 CPU model???
My rex3 with 980x reported rev 1.xx, some of my friends rex3 with 980x reported rev 2.xx.
Yes, I know all retail rex3 PCB rev is only 1.04G.:)
I didn't really want to post this, but I have seen only one RIIIE with a strong IOH. Did the binning just go out the window with these things?
I think it means good IOH chip.
Basically, under 240 - 250MHz, BCLK limitation based on IOH chip quality.
Good IOH chip can run over 240MHz with x36 QPI link speed.
Average IOH can run around 220 - 230MHz with x36 QPI link speed.
Bad IOH chip just only run under 210 MHz with x36 QPI link speed.
The first you check your IOH QPI limit.
You can check QPI link limit with x12 CPU multi, easilly.
(If x36 QPI link & slow mode limit are same, it's CPU BCLK limit.)
I'm limited to 215 before degradation kills my 3D scores and 220 before it stops booting... which is absolute bull crap considering how expensive this board is.
What kind of support does Asus have for extreme benchers? Can someone PM me a better way to go about getting this resolved (or a good contact) since on the front end "overclocking voids warranties"?
I have seen numerous boards with really good IOHs and so far only two that were what I would call average. Of the last 30 samples checked at Newegg, three went over 230 on air, 25 were all around 225~28, and two were at 220~221.
IOH quality certainly plays into overclocking as I have stated before, but the quality of your chip, memory, and power supply play a large part also. Discounting all of that, you have to have the BIOS setup properly based on your hardware combination. PM me and we can discuss further.
These BIOS codes are based on the 1003 code. They are named differently as I released them straight from R&D as service will not provide support on them. The next beta (through the qualification process but not finalized) or official BIOS will be in the 1000 series.
I have been buy over 50 pcs of X58 mobo, and select good IOH chip mobo for benching.
This IOH BCLK limitation does not different between vender to vender.
If you buy eVGA or GIGABYTE or ASUS, the result is same.
I think if you buy 5-6 same board, you can get good one.
This is IOH chip quality problems, not mobo design problems.
So please stop complain to vender.
I love H/W, especially mobo very much!!
So, when I find new rev or new mobo at my favorite shop, I buy it.:D
In fact, Bloomfield & Gulftown overclocking, mobo is very important factor, I think.
If you have good CPU, but mobo does not good one, you can not get good result.
So I need good CPU & mobo!!:)
At this time, my favorite mobo is REX3, especially REX3 BIOS is very nice.
Thank you for ASUS development team & bingo13!!:up::up::up::up::up:
I really doubt at cooling efficiency of Corsair's design, unless there is thermal paste between plates and upper fins, although they are great sticks. Kingston's design is plain ugly IMO with those tall heatsinks. :D
Currently I'm in a process of tweaking memory with Gulftown, but you can check these threads: Mushkin Redline 1600 6-7-6-18-1T@i4memory, Asus Rampage III Extreme - results thread@i4memory and Mushkin Redline 1600 6-7-6-18-1T@mushkin as they contain my settings and achievements.
I realize there's nothing wrong with the design. It's a great motherboard, but from what I've seen, they generally have trouble pushing high BCLK without QPI Slow Mode enabled. What I am saying is that I think for the high end boards, it would make sense for the IOH to be binned better. I accept that this might be asking for too much. It just seems unfortunate that I can have such a great chip and be held back by my board. I was able to push 230 on my old Gigabyte board. I can't get close to that on this one without QPISM enabled.
lol i wont run no gigabut x58:rofl:
thats exactly why i didnt get a ud7
i'll take the board out again n try again with another bios
hope gaming performance is fixed:rolleyes:
btw subzero high blk can be done on cheap boards too you know;)
i'm guessing the IOH on the board suks lol
I can hit high clocks, just not with the x36 multi on the QPI.
http://valid.canardpc.com/cache/screenshot/1346574.png
http://valid.canardpc.com/cache/screenshot/1297816.png
http://i747.photobucket.com/albums/x...erxx/SPi1M.png
http://i747.photobucket.com/albums/x...yerxx/4765.jpg
I want some of those speeds in 3D!
Hi
I read that it was possible to connect a connector "Internal USB" port on the RC Bluetooth but the pin is not in one place
is this possible or not?
thanks ;)
I can't run my card @16x on rampape 3 ,i used slot 2 ,help me
Haha yea the Kingston do look a bit odd, but any cosmetic 'impairment' in the sake of performance I can appreciate :). I actually run my Dominators with the entended fins, and they use thermal pads inbetween. But if you don't like tall heatsinks you may think they also look ugly :p: But that may be since the Kingston are blue; The GTs look damn sexy with the 'Ferrari Red' Corso Rosso accents :D
Will be checking out those links for sure, thanks don :up: Although seeing how well my GTs did, I may never look back at the Mushkin. Truth be told, I already sold them :ROTF:
I sent the board to Johnathan in Fremont CA. The RMA # is USA1082294
As far as the board shorting out against something in the case, I have thought of that possibility but I don't think that is what is causing the problems. I have installed one of my old EVGA nForce 4 motherboards in the case and it POST OK so why won't the RIIIE do the same thing? I wish I had known about the PC Power & Cooling Turbo-Cool 1KW-SR PSU working with the board as you say, that could have saved me $250
Figured I'd give an update on my part...
So far (based on current season temps) I've gotten 4.4ghz @ 1866mhz w/ 12gb memory at cl7 timings (7-9-7-24-72-2T). I'm gonna shoot for 1T here soon and see if I can pull it off. At the same time I'm going to have to upgrade the cooling. Running a single loop on CPU/mobo/gpu is taxing even on a double deck quad radiator so I'm gonna have to fix that.
Something for those going for the higher bclk values, try going at slower mem speeds first before going too high. I've noticed (at least with my setup) the higher the bclk value I go, the slower the mem has to be to get it stable. Might just be with the amount of memory I'm running but once it hits a certain point the mem speeds end up creating your bclk value limitations regardless of how much memory you have (still dependable on amount, high the amount the smaller the range of allowed speed and the lower the bclk value will be).
Yeah, if the Kingstons were red or black it would be better. Dominators are very good looking and color combination is perfect match for R3E. Have you tried using thermal paste instead of pads?
If I could find here Corsair Dominators I would probably do the same, but at the moment Mushkin is best choice available. Luckily it's very good too. :)
hmm just got this together atm testing
as usual r3e screwed up fixed timings
RTL 50,52,54
if this was dfi/evga i could run it at 46,48,50
i7 930+Coolit Vantage ALC
Asus Rampage 3 Extreme bios 0010
Corsair Dom Gt 2k 8-9-8-24 3x2gb
Asus gtx480 + Asus GTX460 (physx)
Corsair AX1200
http://i287.photobucket.com/albums/l...1/ec9bd4b2.png
My R3E hits a BCLK limit at 221...
Rev 1.04G on board and 1.xxx in CPU-Z.
230 here rev 1.4g
Hello Guys...can someone post up a BIOS template to get me started on OCing: 980x, ATI 5970 4GB, 12GB CMT6X3M3A2000C8, Crucial C300 SSD, Enermax 1250w PSU. I have got it auto-OCed to 4.2 via BIOS already but want to see how far the CPU can go. Thank u very much in advance!
Here is some pics of 12GB of Cas 8 2000 GTs with the CMXTRFINKITRED fins. Got completely ripped of on them and would highly recommend that Canadians really dont bother as UPS will shred your backside.
http://img.photobucket.com/albums/v6...x/DSC_0359.jpg
http://img.photobucket.com/albums/v6...x/DSC_0357.jpg
http://img.photobucket.com/albums/v6...x/DSC_0352.jpg
http://img.photobucket.com/albums/v6...x/DSC_0351.jpg
Pics are crap but best I could do in a rush at the time :)
Edit: Also wanted to add a statement about how Corsair ships the fins. They send them in a Cardboard package and seperate each individual fin with cardboard as well. The thing that sucks about that is they do not cover the thermal tape on the bottom of the fin and it was impossible to keep the tape intact. Basically shredded my tape getting it off the cardboard. After paying more than $150 for $7 of painted alu, I destroyed the $2 of thermal interface. Actually quite disappointed in the whole thing really.
But, it still looks good.
that looks so sweet!
You're kidding right? Those close ups do those GTs with the extended fins more than enough justice! Awesome pics man looks like the paint is still wet :clap:
Very true. This was the same case with my extended fins. Only one of them had the entire two sections of thermal tape intact. On the other two, one of each of the two sections of thermal tape had come off onto the cardboard. I guess my kit wasn't as bad as yours as I got them right back on w/o much of a problem. But I get what you're saying, failing to pay attention to details such as that is pretty disappointing. Good thing I got the CMG2000C8s for cheap :yepp::yepp:
Lol, thanks! That was with no tri-pod and point and shoot. Could do much better with outside light and a tri-pod. Will give you guys some really nice pics when the rig is done :up:
Completely agree with you. It is the first time I have experienced this with Corsair so I guess that is why I was so shocked. Glad I got my C8 2000's cheap too LOL :D
Heh... probably my R3E has a bad IOH... same OC... 4.4GHz different value.
4.4GHz (Multi 27 x BCLK 164) pass LinX with all RAM.
4.4GHz (Multi 22 x BCKL 200) no success... stop at 3rd pass. Just stop with error, no bsod or hangs...
Same Uncore and QPI Link... in both... grrr...
try 35x129 @ 1.46-1.48v vcore and VTT around 1.3-1.35v
also if you're running 2:1 ratio for your uncore you don't need to for a 980 *assuming that's what you have reading your sig*. i can push to that point just fine but again, it wont do a full 20 pass on linX but your uncore should only be around 1.5-1.7:1 ratio so aim low on that area.
Has anyone been able to run a xeon 5600 cpu on the r3e??
1st oc 20 Linx passed and 1st post from me :p:
All thanx to this thread :)
http://valid.canardpc.com/cache/banner/1347660.png
http://i249.photobucket.com/albums/g...ove/Passed.jpg
This usin the Official 901 a week ago, messing around with 0010 atm got an xtra gflop thou been to busy with other stuff to run 20 Linx. Forgive me my bad memory sticks, thou their getting job done atm :rolleyes:
CPU: 1.28125v
CPU PLL: 1.65625v
QPI: 1.32500v
DIMM: 1.60325v
700mV
500KHz
OC: Mode 1
Cpu spread spec: Disabled
Edit: forgot to mention LLC is OFF and HT ON
@turtletrax
I can't imagine how the pics can get any better but I'll take your word for it :) Can't wait to see em!
I don't know much about how running 12gb affects max frequencies and timings but try and see if you can get them to run 7-7-7-20-1T at 1.65v. I'm running only 6gb though but got them to run 2033MHz cl7 1.65v Super Pi 32M stable :D Good luck with your setup :)
Also to everyone else, drop your multis and find your boards max BCLK :p:
Ok, I know that this goes back a long way now but, I ended up having to take my video cards out and put them in slots 1 and 3.
And let me tell you that it was hell and I never want to go through that again ! Small case, and 4 HDD's, had to
rearrange everything and redo the cabling. Can we say :banana::banana::banana::banana::banana:!!! here ??! (Ok , so I guess not).
Anyways... so I also tried the QPI_LL_SW jumper. I set it on. And the board did boot and function
normally for me.
But I don't see any BIOS setting related to it at all yet I see some people referring to
a BIOS setting of
QPI Loadline Calibration: [Enable, Disable]
Did you get yours to boot with QPI_LL_SW jumper ?
I am using BIOS 0002, what BIOS did you use ?
Does it even help in anything ? Any issues having this jumper on for 24/7 use (with no overclocking) ?
Running 12 GB at 7-7-7-20-1T at 2Ghz would be almost impossible as far as I know. If it is possible it would take a cherry IMC and non-24/7 voltages on just about every aspect. Realistically (when I get the rig up and running) I will be seeking 1600 CAS 7 to about 1800 CAS 8. Anywhere in that range would make me quite happy. This rig will be doing some hard duty and instability is out of the question.
I will have some nice shots for sure of this rig when it is done as I will never be doing another anywhere near as expensive or powerful.
No BSOD and error code... just LinX says in the bar "Stopped fo error".
vDimm is at 1.64v, VTT at 1.34v for 27x164 and it pass.
And get LinX error with 22x200 and vDIMM at 1.64v VTT at 1.375v
My uncore are not running at 2:1 It is running at 3000'ish MHz in both configuration.
Any 920/930 owners hitting 4.5GHz+ with the R3E? If so, what BIOS, hardware config, cooling, settings, etc? Thanks.
Samson
with bios 0901 and the jumper manually set, LLC is on the first page (extreme tweeker) scroll down near the d-ram settings, option auto 50% or 100%
Take some from this thread: Asus Rampage III Extreme - results thread
Looking great! :up: Those sticks are just made for R3E! :yepp:
Sorry to hear that, but at least you'll be more satisfied with your rig now.
No, there is no BIOS settings for QPI LLC, just an indication of on-board jumper setting.
I have been using it since day one, no problems throughout all the BIOS'es. IMO, it helps.
If you think CPU LLC is helpful and using it, then you will probably think the same for QPI LLC once you get it working.
Sometimes you have to live and learn. Glad you found a solution for the pcie x16 slot problem.
And as far as QPI LLC goes, you cannot set anything in the BIOS. It is strictly a jumper setting..only shows in the BIOS whether it is enabled or disabled.
Hope that helps :up:
Thanks guys, I see it now, at the top of the page of extreme tweeker
Yes. A very smart and simple answer. I agree. It is about the hardware makeup, not the little BIOS that's going to do miracles.
It's also a game of luck as bingo13 has often repeated about getting a lucky chipset that can clock well.
And since Asus does not make the x58 chipset themselves, well... what can you do.
To some degree a BIOS can optimize or enhance some features but certainly limited by the hardware available.
And also, your own choice of hardware that you put on, and how well you put it together and how well you
set your BIOS settings and what OS you run (ex. 32-bit, 64-bit) and how well you understand the OS and what tweaks you did to
the OS to optimize performance, etc, are all factors too.
Now if the board's engineering layout is bad, or if other boards seem to statistically have more "lucky" chipsets,
well, I don't know what to say about that except agree with raju, if you're not happy use the one you really like.
ooops,
still learning !!
http://www.xtremesystems.org/forums/...&postcount=823
vCPU 1.57
vQPI 1.45
Water cooled
Again, judging by your sig, if your going for 1600mhz on memory and you uncore is about 3000mhz, that's nearly a 2:1 ratio. You can get away with running with 2300-2700 alot easier.
At the same time, linx likes to ask silly voltages just for a good pass. My rig has been 24/7 stable at 4.515ghz for a while now (aside from down clocking due to recent summer temps being nearly 100 degrees with no wind... 4.4ghz). It'll pass not just hours but days of prime and mem tests without any hickups and hasn't crashed on any game at all yet it still wont pass linx because linx wants me to throw 1.5-1.51 vcore to it even though its stable at 1.47-1.48... because of that it wont pass linx and I'm not really willing to see all 6 cores on my 980 to be in the mid 90*c range...
I discovered this info a while back and thought I'd share it with everyone.
I usually test the memory side of the CPU with Super PI, then overall CPU stability with 20 passes of LinX. I was getting a lot of crashes and BSODs with LinX so I did some research. I found out that the linpack included with LinX 0.6.4 is outdated, and actually doesn't not have updated code for Gulftown CPUs; therefore causing false errors. Even on bloomfields it crashes and requires a lot more voltage. With the newer linpack, the cpu gets stressed more (temps will go higher) but it will pass at the lower voltages it did for Super PI.
Go to this link here to download the latest linpack, and just go to your LinX folder and copy/paste to replace. If you also go into the properties of the two different versions of linpack, you will see one version (latest) is verified and the older (original included w/ LinX 0.6.4) is not.
Nice job with the 920/930 overclocks, guys. Thanks!
What setting are you guys using for 'PWM Frequency'? Also if anyone can shine some light on the setting's function and role in overclocking I'd appreciate it.
QPI drops to about 1.25v, but i think its my funky memory sticks using 0.05v less than most others at highest frequency.
Funny part (NB Frequency) is at 3370 MHz at default voltage so QPI prbly be less volt at 3200MHz or lower i think :)
500 KHz "CPU PWM Frequency" seems to lower my CPU voltage by alittle :p:
700mV "CPU Differential Amplitude" and "Extreme OC" Mode 1,
seemd to help me thou not 100% what they do was something i picked up from one o these 90 pages :rolleyes:
Think Cpu Differential Amplitude makes it easier to up the BCLK...
I bet someone else here can explaine those better ;)