The Official OC (OverClocking) Thread!

Hi just wondering if anyones tried an Aerocool Dominator cooler

looks quite impressive and the db level is low too
http://www.cluboverclocker.com/reviews/heatsinks/AeroCool/dominator/index.htm

 

i use an artic cooling freezer 7 pro on mine and its great when you consider the price.

@ck5134

you might want to change you sig so its all on the same line b4 the mods see it.

 

Ah Im after as quiet as possible with a bit of clocking, the zalman i have is good, but does need a regular dust out, just thought about the Aerocool Dominator after reading the review and them saying its quieter than the Evercool WC-202 and looking at the temps only 2ºc warmer on an oc/d processor underload than the Evercool WC-202. now when air is that comparible to a water kit im interested )

thanks for the kick about my sig too

 

ck5134,

I don't much like the design. it only has 3 heat pipes and no support on one side. Vibration could be a problem since it's so massive and has such a huge fan! BTW, it's 5C warmer (40C vs 35C) under load than the WC-202. I'll stick with the arctic Freezer 7 Pro, for an Intel anyway! Got 2 Freezer 7s and 1 freezer 64! Love em all! Doesn't do quite as good a job on the AMDs, but still very good!

Clock On,
theone

 

As far as I was aware, the Freezer 7 Pro only had 3 heatpipes. Zalman coolers are OK, but only because you can vary the voltage of the fans, at 12V I find them intrusive. Essentially what you're after is a heatsink that can cool better than you need, so that you can have the fan speed low. I recommend looking into the Thermalright ultra 120, and using a Nexus or Scythe Case fan with it. Nexus fans have the smoothest sound and are almost the quietest fans available on the market, yet push reasonable airflow. Some people choose faster fans for better cooling, but my system is a pretty hot one, and Nexus fans could do the job perfectly. They're not cheap, but I recommend them if you're looking for quietness.

 

theonejrs
Ah but when oc/d the temp difference has dropped to 2ºc difference and idle overclocked its the same, seems similar to the results i had with my Asetek VapoChill Micro, the more you clocked the better it worked, yeh i get your point on the vibration front though, shame they dont just incorporate a few struts on things like these to minimise it.

sammorris

Im thinking of moving over to a Noctua NH-U12F for my next heatsink, I mentioned the Aerocool Dominator as I wasnt sure if anyone here had any experience of it, but as theonejrs mentioned vibration could be a problem with it, I guess ill stick with the Noctua NH-U12F.

)

 

sammorris,

Nope, 3 on each side and not the shared 3 like the zalman 9500. I don't like the Dominator because of all the weight being supported by the 3 heat pipes with the opposite side hanging in the air. Any vibration could lead to cracking of the heat pipes at the heat sink, especially as heavy and massive as it is. The design of the Arctic is much more solid as it's weight is evenly distributed! Add to that the airflow is straight at the rear case fan instead of being blown into the case the way the Dominator does it. I'll stick with what I got!

Clock On,
theone :}

 

how much harm can be done if i oc a amd 3200+ from 2.0 to 2.5 just using the stock heatsink?

 

ck5134,
I just looked up that Noctua and it's one "HUGE" S.O.B. at 155mmx126mmx95mm (with fan)! The 155mm height would cause a lot of problems with a number of cases, especially if it has a window fan! I don't even think it would fit on my MB or in my case. A Scythe "Ninja" while not near as huge would also be a reasonable choice. Still large but not on the scale of the Noctua! The fan for the Noctua would wind up in the trash as the blade design is poor. Put a silverstone on it and get some real airflow without all the noise. Just letting you know what I think!

Clock On,
theone :}

 

theone

The supplied fan is adequate for me here, im in england so temps arent much of a worry, good old sunny england, the size isnt an issue with my soprano case, it wont even interfere with the acoustipack fitted in my case.

The fans are quiet even at 1200 rpm, mates got a couple on his pc as case fans, and they are suprisingly good, still push 48 cfm which will be good enough for my needs.

 

ck5134,
Modesty requires that I tell you that My Silverstone 120mm case fan moves about 70-75 CFM at 1600 RPM. 110 CFM at 2400! That's the Oh Sh17, it's overheating emergency speed, because it's "LOUD"! Very quiet at 1600 which is what mine is set at. Because I have the baby brother of sam's Toughpower, and it vents about 30% of it's cooling fans air, inside the case. My PS used to blow very hot air out the back. Now it's just warm and the case is much cooler. It even barely gets warm directly above the PS! Great fan that comes with a controller with a 3 1/2" mount. I drilled one of my spare 5.25 bay panels and mounted it there. Black knob goes good with the silver! If you should happen to buy one (highly recommended) Please don't laugh at the tiny little speed control. All the electronics are in the motor so it's just this tiny little Pot with a knob on it. It does work well and will maintain pretty much whatever speed you set it at!

Clock On,
theone :}

 

Go with what you like, but the Noctua fans aren't all they've cracked up to be. The actual fan speed is quiet, but most of them suffer from a defect where they make a whine that doesn't change pitch if you increase/decrease the fan speed. That's annoying. Nexus or Scythe fans are typically very silent. For the record I currently use the Silverstone fans, and whilst they push incredible air at 2400rpm, that's at the cost of them being pleasant to hear at the midrange setting. They only sound like a good fan up to about 800rpm, at which point a resonant whine kicks in. In a very quiet case, you'd notice this, and not like it. From what I've read the Thermalright Ultra outperforms the Noctua NH-U12F and is smaller, and you get to choose your own fan.

 

not that much tbh, it will run a bit hotter but as they aren't the hottest of chips to begin with...

i know of a semp 3300+ that hit 2.62 on 1.5v with stock cooling and it idled in the low 40s and hit mid to high 50s under load. that is between 10/15c higher than at stock.

my old 3400 had an evo33 blower on it and it ran cooler with it when it was oced than it did at stock speeds with stock cooling(not really a shocker but...).

i think the best thing to do is watch your temps, if it starts to hit 60c under load you need to get better cooling.

 

marsey99

ok so i can oc with the stock heatsink if i'm only going to go up that much. also would i really notice a different from the stock speed?

 

totaly honest, it depends what you do.

the link above shows what i maxxed my amd @ but i found the sweet spot to be 2.41ghz (241x10 on 1.45v with ram set 2 166 but oced to 201/ddr402 with tight timings) for most things but i got better benchmarks scores at 2.25ghz with my ram on a 1:1 divider. @ 2.4ghz windoze felt snappier, everything seemed to just work faster,more instant. only if was doing something cpu intensive would i notice the extra mhz.


just a modest +10% oc would give you a slightly more responsive system. i dont think many others in this thread will tell you a stock system is as good as it could be.

 

ok thanks. would you be willing to help me oc it because this is my first time doing it?

 

sammorris,

I don't get any whine from mine at all till it gets above 1600. I had the side cover off the other day when I changed the fan in the cover and was amazed at how quiet the Silverstone was at 1600 compared to the Arctic at 1000. If it was the only fan in the case, you wouldn't even hear it. Even with 5 other fans running, when I put my ear up to it, it's very quiet. All you hear is a gentle rush of air.

For anybody the bottom line is how well it all cools. With the old 37 CFM fan that came with the case I had to run the Arctic at top speed and the case fan at top speed. I didn't appreciate just how quiet my computer could get until I removed the Cooler Master and replaced it with the Silverstone. Now I let the MB control the Arctic and set the Silverstone to 1600. Right now the Arctic is running at 800 rpm and all the temps are very good and the noise level is not annoying, and especially since the GPU cooler runs at full speed all the time with my 7600GT overclocked to 653/1600. Yes, it's not up to your standards for quiet but for me it makes so much less noise than before. Just the fact that the PS runs so much cooler makes using the Silverstone worthwhile. Considering that I have 2x80mm @2400, 1x40mm @3000 and whatever the PS and vid card fans spin at all running together for a total of 7 fans, it's very quiet indeed! The higher cfm cover fan makes it a little bit louder, but not annoying and it removes heat from the case, which the old one didn't do.

marsey99,
I checked out your CPU-Z and concluded you have a very rare find in your CPU. It must be an exceptionally well binned chip as it's almost unheard of for the FSB to even get to 400 let alone higher with a 4300. The older B stepping would do it, but not most Ls! Mine will only run at multipliers of 6 or 9 and 374MHz is max. I've tried both the 7 and 8 multi and it won't even post if I try to exceed 374GHz. It just sits there rapidly beeping at me and turning the PS on and off until I hit the reset button while it's beeping. Then it posts but at 1.8GHz as it turns the Auto off! I would be curious to see your Sandra 2007 benchmarks to compare to mine. They should be reasonably close as for all intents and purposes they both run about the same speed but your FSB is higher. I'm curious to see whether the higher FSB makes very much difference one way or the other. I'm also curious about your temps!

Clock On,
theone :}

 

Hi again everyone and thanks as always for helping me to get up to speed.
I've been all over trying to find out as much as I could about C2D temperatures. I came to a few conclusions.

The core temps are a number calculated inside the CPU. Core temp = the Tjunction constant (85°C for these Conroes) - the value from the sensor. This is read from the CPU by the software.
The CPU case temperature is done differently. There is a table in the bios that is used to come up with the core temp.
Both the Nvidia 680i and Intel 965 bios can show core/case inversion. That's not good.

I figured swapping the CPUs should shed some light on the matter:



So now I'm running them both with mild overclocks:

Overclock: ~14%, 2.7 Ghz for the E6600 3.0 Ghz for the E6700, Vcore set so both display 1.30V in Everest.



This is what I'm getting with Orthos running:
E6700 on the Asus P5W DH
CPU 41°C
Core 1 40°C
Core 2 41°C

E6600 on the EVGA 680i
CPU 36°C
Core 1 36°C
Core 2 35°C

All I can say that is that when the E6600 was on the Asus board, the core temps were a couple of degrees higher than the case temp.

All this seems inconclusive. but AFAIK, both these have damaged DTSs.
What I've done is ordered another E6700 from the egg and got an RMA number for the one I already have. I'll try both at stock bios settings and see what the Everest numbers look like.
I also bought another psu for the SLI computer, It's an Enermax Galaxy 850 Watt Unit. It's SLI certified and gets pretty good reviews. I hope that helps with the .4-.5V difference I'm getting in Vcore. This is not dropping under load. When I set the Vcore in bios, it's getting reported .4-.5V lower. Even at idle. It drops another .2-.3 volts under load from what I selected in bios. That's why I kept upping the Vcore and probably cooked the DTS's then.
If that doesn't help that then I'm RMAing the EVGA board.

 

Nice post, with a few show off pics as well! Do I spy a Thermalright heatsink and scythe S-flex fan? :O

I think I may have got to the bottom of why I rate the Silverstone fans worse than theone does. What is your case made from Russ? Mine's aluminium and will therefore pick up more vibrations, if yours is steel, I think perhaps we know what the problem is!

 

fasfrank

been busy w/ work lately, haven't had time to reply to your eVGA 680i inquiries as of late, my apologies for that

good choice w/ the PSU, i was using the big brother to yours, the Enermax Galaxy 1000w a few months ago, Enermax PSUs, (at least for ME) have always been very stable and well built...i was running the older Enermax Liberty version on my AMD rig. when i had my 7900gtx video cards in SLI, i only had to replace it with a bigger PSU once i went single 8800gtx there~

as to your current temp issues, while the core temps should be higher than the cpu temp, i too have also found that on the 680i mobo, they have been reversed at times; but that didn't bother me that much only because you have to consider WHICH specific version of Everest you are using, i've always been the type to experiment and test out the newest betas the day they are released, i've rarely ever just stuck it out with the "final release" versions for very long.

sometimes in the beta versions, temp readings are corrected and sometimes they are not...it's hard to tell by simply loading up the app. and having it monitor your various temps...you also have to remember that the 680i chipset is "fairly" new and Everest hasn't had nearly as much time tweaking/testing/revising the software to work perfectly with it.

also, with the eVGA 680i mobo, i've gone through AT LEAST 6 of them at this point and have kept only 2 of them, sometimes it actually is more headache and trouble than it's worth to be an "early adopter" of the newest tech at one point, i've had 3 concurrent RMAs w/ eVGA setup just to replace: 1. BIOS chip issue, 2. chipset issue, 3. faulty RAM slots issue on 3 different boards. i have confidence in eVGA's customer support dept. though and have had nothing but smooth sailing in getting an RMA quickly and painlessly (i'm setup with eVGA's "advanced RMA" process on all my mobos)

anyways, are you using the latest BIOS for the evga mobo? it's P28 now and SO FAR it's proven to be pretty decent at least for me. i RARELY ever use the Nvidia Ntune Console to report temps/readings only because i've found it to be highly inaccurate in the earlier revisions of the software itself. i don't know firsthand if it's improved much recently, and if some of the bugs were corrected, but i've since abandoned it long ago while messing with it.

whether you fried the sensors on the cpu(s) themselves or not, i couldn't tell ya, but i was just concerned about your vcore volts being so high from a few weeks back, even if your cpu required that much to maintain stable OC's you achieved, it was a bit extreme in my eyes despite having very HIGH END air cooling i suppose the newer cpus DO require more volts to achieve the higher overclocks that were possible with lower voltages in months past, can't win em' all~

docTY

p.s. i wouldn't go and RMA your eVGA mobo if everything works on it and you don't see/experience a REAL issue with it, that point CAN be argued as it's a very personal decision and truth of the matter is, IT's YOUR MONEY and mobo; i just wanted to let you know that at least from MY past experience, when i had a SPECIFIC issue with a particular component on the board itself, that's when i RMA'ed...otherwise, if it runs fine and stable and all your other components seem to work well together, i see no real reason to get it replaced (unless you just want a new mobo for the fun of it, but know the replacement CAN be a refurbished model and they won't necessarily send you out a brand new, retail sealed mobo in your RMA...i've gotten 2 "refurbished" boards to replace my retail purchased boards and those DID NOT fix the initial problem either)