-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
FX8350 and Crossfire setup
- Thread starter tdog2005
- Start date
- Joined
- Apr 20, 2004
- Location
- JAX, Mississauna
Since you already have the board and cpu no need to hear any replies about an Intel rig doing better. Nuff said.
I see responses across the forums on the net that say will not bottleneck and only a few that think it would be a bottleneck. I do believe you should make sure the cpu is to at least 4.0Ghz and then you should do fine. After all I have seen others writing.
I see responses across the forums on the net that say will not bottleneck and only a few that think it would be a bottleneck. I do believe you should make sure the cpu is to at least 4.0Ghz and then you should do fine. After all I have seen others writing.
- Thread Starter
- #3
Well i installed the 2nd card , instant jump of 45 fps in the unigine heaven benchmark. Case is now crowded and i did notice temps of everything has went up, there is no way i will be overclocking this cpu unless its on water. Does a push pull setup on the air cooler help out much?
- Joined
- Feb 17, 2010
- Location
- Manassas, VA
Only a few (2-3) degrees in most cases.
- Joined
- Jan 12, 2012
You may be better served now by adding fans or getting higher flowing case fans, then adding a fan to the cooler.
- Joined
- Dec 22, 2006
You could probably tinker around with lowering the voltage on the cpu and see what he lowest you can go stable is, a lot of mother boards tend to give the cpu's more voltage than necessary.
- Joined
- Jan 12, 2012
tdog try taking the side panel off the case and see if the temps drop at all. I'm thinking those 2 cards are throwing a substantial amount of heat into the case. With the standard fans installed you may not be getting enough air moving through the case. Therefore you're cooling the cpu with all that hot air.
- Joined
- Jan 12, 2012
I would first try pulling one of the GPU's put and stress test it at stock clocks to see how hot it gets.
- Thread Starter
- #12
I pulled the 2nd card and got the same results, I then decided to put the fan on the opposite side of the heatsink facing the ram and it dropped to about 58 degrees, I managed to undervolt a little bit , to much resulted in errors in prime and I am sitting at 55degree's under load, a lot better then 64 degrees, I think I am going to add another fan at the top of the case since there is room and maybe this might help a little bit. Under load what should this thing sit at?
- Joined
- Jan 12, 2012
So what I gathered is that you now have the heatsink fan on the side by the ram pushing, correct? Is it overclocked, I'm assuming it's not? I couldn't tell you what your temps should be with that cooler. My temps are a lot lower but I have a full H2O setup. I don't think they should be that high at stock clocks with that cooler, but that's just an assumption.
- Thread Starter
- #14
I wanted to update to the noctua d14 which is supposed to be really good but its just to big for my case so I was thinking about a closed loop system what would be my best choice, I hear the h100 is good but I cant go with this one as I was told that wouldn't fit, the options I see at the store are the Corsair H80I, Antec 920 and Zalman Lq 320, would any of these be worth it?
- Thread Starter
- #15
yep its stock speed not overclocked and the fan is now pushing on the ram side, motherboard temp also dropped when the fan was moved over to this side
- Joined
- Jan 12, 2012
I think the Evo should be doing a better job of cooling that chip then 55c at stock clocks. Maybe try putting another fan on it in push pull, though it probably only drop the temps like 3-4c more. I still think that's high at those clocks.
- Joined
- Jan 20, 2002
i was torn between getting the 920 or the H100, but i didnt like the idea if the pump going out and that i might cook my cpu so i ended up getting the Phanteks PH-TC14CS_RD. As far as my research, they were really close to same proformance...more of a extra feature difference then anything to me. the zalman and TT ones i dont know as much about so no thoughts. if you havent been here check out http://www.frostytech.com for some decent research database info.
the only thing im not a big fan of for my new setup is with my new cooler all the extra heat is dispersed into the case instead of outside as i would have had if i chose a closed loop water cooler setup. Now since you have taken off the side panel and are still getting high temps i'd say check you TIM and re-seat you cooler before swapping out to a new one, i agree with Mandrake 4565 that it should be doing ALOT better job then that.
Last edited:
- Joined
- Jan 12, 2012
This was exactly what I was going to say also. Tdog check your TIM application and make sure the evo is mounted correctly. Too much TIM can cause poor heat transfer.
- Joined
- Oct 29, 2012
This.
I was able to get my 8350 up to 4.75GHz on an H80, and it had about the same temps yours did at the start of this thread, if not a couple degrees cooler (Topped at 62). Was only able to push that to 4.82 on an H100, so I wouldn't consider that a worthwhile investment over an H80 haha
I'd definitly make sure you got a good paste, and a small application. I always reccommend the pea method, sometimes I attempt to enhance the pea method with near-corner tiny dabs of paste, but this typically either nets me no gain or 1-2c at most.. Also can decrease efficiency if you don't do it right (Can create air pockets)
You have no idea how many people have come to my shop with thermal issues just for me to find out they tried to reapply their TIM-- They typically do not clean the old TIM off, or they put a half gallon of TIM on their CPU, so it looks like a snow storm or a T1000 on their processor =_=
- Thread Starter
- #20
Re applying the TIM was the first thing i tried, i cleaned the cpu and heat sink with rubbing alcohol, the first time i applied a pea sized dot amount center of chip, 2nd time i cut it back to half pea sized center of chip, when tightening down heat sink i did the cross method, i seem to get same results both times.
I am guessing i should be 50 under load ? I will pickup one more case fan and see if it helps
I am guessing i should be 50 under load ? I will pickup one more case fan and see if it helps
Similar threads
