Printable View
Maxxx, can I have your cpu please?:D
Looks good m8, in my experience the asus probe thing is not very accurate, but if you were registering 49/51 before tinkering then it could be possible, a bad mount can be disatrous to temps.
The only way to get a true reading of course is with an mm:(
but its looking good
Thats good to hear. Will save me having to invest in more cooling. I will give the stock cooling a shot first, but if im getting ridiculous temps then i'll watercool the NB with my S-Max, and then use the thermalrights for the mosfets.Quote:
Originally Posted by rob2k
To anyone that had this problem - its the same issue I had. It looks like the board isnt DOA. I sent mine back to scan, and it worked fine for them. They tested with corsair memory, where as I was using supertalent. I think it could be that this board just refuses to work with certain types of memory, and the technicians at scan seemed to backup this theory.Quote:
Originally Posted by REVHEAD
I have another on its way anyway, and my 2x2gb set of g-skill PI arrived yesterday, so hopefully that will get me up and running!
@bustamove44Quote:
Looks good m8, in my experience the asus probe thing is not very accurate, but if you were registering 49/51 before tinkering then it could be possible, a bad mount can be disatrous to temps.
Its why am not sure about that reading. 49/50c was before changing thermal paste. But after the change, it was reading way over 60c on idle, but after adding plastic washers to the screws, the temps drop. I didn't re-mount the heatpipe assemply. Just added plastic washers.
Oh, do you think the E8600 is a good chip? Its my first wolfdale and was trying to achieve the same voltage the other folks get.
@MikeMK: thats a bummer. Wasn't also sure about the XMS3 I have when I first built this, but they seem to be doing good.
What's the ambient in your room? Unless you are using some kind of exotic cooling the CPU must be above ambient temperature... 7 degrees seems way off. I would set the LCD Poster to report your temp from the BIOS options, and see what you get.Quote:
Originally Posted by maxxx
eternal_fantasy , sorry , just a quick question
are You using (CellShock MSC CS3222580) BLACK Kit ( DDR3-1800 / 2x1GB )
---
myself, currently got Maximus Extreme board, but still in
endless fight :lsfight: with whole ensemble /setup
to get Maximus Extrem stable with Quad and this ram kit :confused:
so im just keeping eye on that new Rampage Extreme board..
it would be good to know, if i can use this ram kit on this board and finally get them running at their nominated speed
TIA
:)
@EF: I also don't believe its 7c for the CPU. No exotic cooling for me. Just a TRUE120 and a 2000 rpm fan (plus an aircon at full blast). I would trust the reading of CoreTemp in this case. My concern is the NB. If I don't use PC Probe II, any alternate software we can us to read the temps? I got the temps from the BIOS showing in the LCD post but wanted a software as well (if there is one).Quote:
What's the ambient in your room? Unless you are using some kind of exotic cooling the CPU must be above ambient temperature... 7 degrees seems way off. I would set the LCD Poster to report your temp from the BIOS options, and see what you get.
Cheers....
yeah, the CPU reading in PC Probe II drops to 3~10C for me once in a while too. then the next time you open it up, it reads 'normal' again. just a software bug, rely on realtemp/coretemp anyway as the PC Probe reading sometimes stays pegged at 50C or something while realtemp goes from 45~50C during various Prime iterations.
@3oh6: hey man, been busy? Thanks for the tip. How about for the north bridge? I just followed EF's suggestion now to use the LCD poster but for a software to do it? Everest or something like that?
16-18 hour days at work all week...yeah, you could call it busy ;)
as for Probe II, the NB temp has always been consistent for me. when ambient temps raise, the NB temp raises appropriately, and then falls when i drop ambient again. i have no problem believing your temps went down by tightening the screws with washers. did you leave the springs in place or just remove them? might try it when i change the thermal paste as well. also, was it just for the NB mount holes or for all screws you added washers too?
idle temps of course are one thing, but load is another. you probably won't see as huge a drop, or maybe you will. but again, Probe II hasn't done any funky readings of NB temps like it does with CPU. load up the system with HCI Memtest, and see what temps are. when fully loaded, i have found HCI to bring the most heat to the NB.
Yeah it's as in my sig, the black DDR3-1800 2x1GB. It works fine at 1800 7-6-5-18 in my Maximus Extreme, and works fine at the same settings in the Rampage Extreme. So no idea why you can't get it to run at rated speed in the ME..Quote:
Originally Posted by i43
Maybe Everest Ultimate may detect the NB sensor, but I'm not sure. It's just BIOS and Poster report for me. Always found the PC Probe to be a little buggy.Quote:
Originally Posted by maxxx
Sure that increasing the contact pressure on the NB will improve temps, but down to 35 degrees on air cooling? I'm using water, direct waterblock contact with the northbridge, 26 ambient etc, and with 1.48V MCH and I idle also at 35 degrees? I'm finding that hard to believe. Hopefully more people can try adding washers etc and prove me wrong. :up:Quote:
Originally Posted by 3oh6
So what does the LCD Poster say on the NB temps maxxx?
@3oh6: I did add plastic washers to all the screws and the springs are still there. I'll try HCI Memtest and see the results....Quote:
also, was it just for the NB mount holes or for all screws you added washers too?
@EF: I did install Everest, but I have no clue which is the NB temps there..:am:
Let me get back to you on the temps from the LCD poster. Will put load to it and let it run for a fair bit.
Maxxx -
In Everest, expand the "Computer" category on the left-side list. Highlight "sensor" under "Computer" and you'll see a "Motherboard" reading on the right side. I've always interpreted that reading as NB.
i'm hoping my initial thoughts are wrong too thinking something else happened but keep in mind, idle temps are more about ambient than anything...that's why i wanted to see load temps. if his ambient is aroun 21-22C then 35C isn't that difficult to achieve with the stock setup, i can do it too.Quote:
Originally Posted by eternal_fantasy
that is the generic MB temp sensor which is just a sensor somewhere on the board measuring the surface temp of the board in that area. the latest Everest build i have tried doesn't have hooks for NB, SB temps yet but perhaps an update will add those. your NB temp is going to be quite a bit higher than the MB reading from Everest, and the MB reading from Everest should match up to the MB reading in Prob II.Quote:
Originally Posted by cool_case
oohh.. this is the best board i ever had :eek:
after tweaking the VTT, CPU and NB GTL i think it run stable... (before tweaking i had some problems with large ffts / 896k & 768k) :)
http://www.abload.de/img/4.5ghz-600fsb-largefkga.png
@cool_case: 3oh6 is correct, it seems to be the mobo sensor reading there but thanks nonetheless....
@Chri$ch: nice one there.
Here's a screenie while running HCI Memtest. http://img178.imageshack.us/img178/6632/screen6dw0.jpg
LCD poster is showing exactly what ASUS PC Probe II is showing. My ambient is somewhere between 22c and 24c. Do you guys think I had destroyed or damage the sensor somehow? Isn't it inside the NB?
Btw, for those using the stock sink when you replace the Fusion block, make sure you put enough thermal paste. When I was putting mine back together, I applied a very thin layer of AS5 and it didn't even touch the stock sink. Replaced AS5 with MX-2 on that part.
maxxx whats your vNB?
it's in the screenshot, 1.32v.Quote:
Originally Posted by eternal_fantasy
that is still awfully low but your vNB is also quite low, i've never ran that little so i can't really compare. the sensor, hopefully, is in the actual die of the NB like a CPU but i don't know if the x48 even offers that. i do notice substantial swings from idle to load in a relatively small amount of time making me believe the NB temp sensor is on-die or at the very lest, involved in it direct somehow.
i would like to see your NB temps under load at 1.5v vNB. and when using multiple instances of HCI, you might have to go in and set the affinity of two instances to one core and the other two instances to the other core. it doesn't look like that fourth instance is getting any CPU cycles.
Good point. The fast response idle-to-load NB temps likely means the sensor is on-die.Quote:
Originally Posted by 3oh6
I just checked temps on my old P5N32-E Plus. There is no NB reading available in PC Probe, which means that the NB reading is provided by a special sensor. On the "plus", MB reads 33 (in Everest and Probe) and a wire temp sensor I just put under the NB heatsink reads 56 @ idle, though it doesn't go up under load.
Man... this board is killing me.
I don't want to spend any more money, but....damn!
Just installed my board today. What´s the lastest chipset driver for this chip ?
Where can i find it ?
Its actually 1.33872 as set in the BIOSQuote:
maxxx whats your vNB?
Am kind afraid to run it at 1.5v since we're not sure if the sensor is damaged or broken. I guess i'll have to try and see the results...Quote:
i would like to see your NB temps under load at 1.5v vNB. and when using multiple instances of HCI, you might have to go in and set the affinity of two instances to one core and the other two instances to the other core. it doesn't look like that fourth instance is getting any CPU cycles.
@cool_case, I see on your sig you had a Maximus Extreme, does it have a temp sensor on the NB?
@Turbodream: I just use the one provided with the DVD. Asus doesn't have much of a download yet for this mobo.
If you keep NB temp below 55c average should be fine. It's kinda touchy when run at higher temps than that. With a 40x40x20mm fan and AS5 my NB idles at 35c, load at 38c with 1.53V vNB with Asus heat pipe majig on the Rampage Formula. So I quite can't understand how others have such high temps besides the obvious poorly ventilated cases or hot ambient temps (ie. summer)
open bench setup, ambient as low as 22C as measured at the fan intake area and with the stock cooling, i can hit 55C anytime i want by loading up HCI with 1.50vNB. some people just have better contact than others out of the box and possibly it is the chipset itself just running hotter than others, lke CPUs. one sample doesn't mean all of them are like that. plus, you are talking about your Rampage Formula, aren't you?Quote:
Originally Posted by mikeyakame
I have a feeling mine would run higher on an open bench. I can't replicate the same airways on an open bench as I can in my 832 Stacker. With 1 120x38mm Delta at the back , 3 120mm door , 2 120mm front, 1 120mm top each group with diff rpm, cfm and varying direction it works better than I could have ever imagined.
2 front both draw air in, 1 fan on door on lower quadrant is intake which feeds dominator ram cooler and 40x40x20 NB fan. 110cfm Scythe in my tuniq tower which does a good job of removing the heat round the vrm area. Delta directly behind it spinning at 3700rpm with 130cfm (not too loud either, quieter than 4500rpm 230cfm delta i have sitting here), all other 120s are exhaust. My mobo is also installed opposite direction so cpu cooler is at bottom of case, gpu closer to PSU. Opposite of the normal Stacker 830/832 placement.
I was surprised when i removed the heatpipe majiggy that the NB surface was actually pretty flat, was tempted to lap it but finish was actually way better than I expected. Would have been too little to gain from doing that. SB temps did rise by 2-3c after putting on AS5 lol which surprised me, but hell 34 instead of 32c is hardly something to be concerned with. My tuniq tower on the other hand has a nice big warp on the contact surface, gave up after an hour of sanding it. When I get around to it I'm going to throw it on the mill and do it properly and quickly.
Eddit: Yep the formula. Difference should be for the most part the cooling setup and beefier voltage reg on the extreme right? Physically the X48 chips are still same revision I believe, so the differences between both should be minimal. DDR3 at high frequencies loads the NB a little more, but not a great deal more than DDR2 since it's more efficient and timings are generally looser overall.