FX8120 and his temp

[Bubba-Hotepp]

When the liquid temp reach equilibrium, doesn't matter if there is 1, 10 or 100 litres of liquid.
It only takes longer to reach this state and, when talking about a couple of feet of tubing, it's a matter of seconds.

You've made the assumption that a system with x amount of liquid will reach the same temperature "equilibrium" as a system with x + y amount of liquid.

So let's summarize some of the things you've said. According to you, "seasoned overclockers" and "benchers" know more about the inner design and workings of AMD CPU's than AMD Tech Support and the Embedded Dept (i.e. the engineers), and it doesn't matter how much cooling liquid is in a system the temperatures and cooling capacity are going to be the same. Still want to stand by those two statements?

 

[manu2b]

- when reaching the mid 50s on the cores, the cpu get unstable.
- the volume of liquid doesn't matter once the equilibrium is reached.

Discussion closed on my side.

OP, sorry for hijacking your thread

 

[Bubba-Hotepp]

- when reaching the mid 50s on the cores, the cpu get unstable.
- the volume of liquid doesn't matter once the equilibrium is reached.

Discussion closed on my side.

OP, sorry for hijacking your thread

I've posed both of those "assumptions" to AMD for the first, and Swiftech for the second. Let's see what they have to say.

 

[EarthDog]

Lower temps with more water? No. Just no. Not remotely right especially with the meager difference. As manu said it will just take longer ti reach an equilibrium. Its basic thermodynamics.

As far as instability, 55c you tend to get unstable. I thought fx may handle a little more but its in the ballpark.

 

[Bubba-Hotepp]

Lower temps with more water? No. Just no. Not remotely right especially with the meager difference. As manu said it will just take longer ti reach an equilibrium. Its basic thermodynamics.

As far as instability, 55c you tend to get unstable. I thought fx may handle a little more but its in the ballpark.

So your saying that with a system with the same cooling efficiency, same thermal energy applied, that it doesn't matter whether there's 1 cup of water or 1 gallon, temperatures are going to remain the same. Am I understanding what you're saying?

 

[manu2b]

Yes, equilibrium, basic thermodynamics:
Thermal equilibrium
Main article: Thermal equilibrium

Thermal equilibrium is achieved when two systems in thermal contact with each other cease to have a net exchange of energy. It follows that if two systems are in thermal equilibrium, then their temperatures are the same.[2]

Thermal equilibrium occurs when a system's macroscopic thermal observables have ceased to change with time. For example, an ideal gas whose distribution function has stabilised to a specific Maxwell-Boltzmann distribution would be in thermal equilibrium. This outcome allows a single temperature and pressure to be attributed to the whole system. Thermal equilibrium of a system does not imply absolute uniformity within a system; for example, a river system can be in thermal equilibrium when the macroscopic temperature distribution is stable and not changing in time, even though the spatial temperature distribution reflects thermal pollution inputs.

Source Wikipedia

 

[EarthDog]

Yes. If I have .5 gallons or 1 gallon of water in a loop at 23c, with the same sized radiator and same block, same flow rates, the temps will be he same. The loop with more water will take longer to reach an an equilibrium.

 

[manu2b]

Yes. If I have .5 gallons or 1 gallon of water in a loop at 23c, with the same sized radiator and same block, same flow rates, the temps will be he same. The loop with more water will take longer to reach an an equilibrium.

I did the experiment myself with a 20l res... It took 3 hours to reach the equilibrium, but finaly, the water temp was the same as with a 0.2l res after 10 min...

 

[EarthDog]

Let's keep it in the one thread.

 

[manu2b]

Lol, yes.

 

[Bubba-Hotepp]

For the coolant part let's keep it in the other thread and clear this one of that issue.

Edit - (That's this thread btw - http://www.overclockers.com/forums/showthread.php?t=710464 )

 

[caddi daddi]

cloudheaven,

What clock speed are you running?

 

[Cloudheaven]

cloudheaven,

What clock speed are you running?

Now i'm running at stock (3.1 ghz), I've disabled the TurboBoost feature
in the BIOS that caused me the overheat at the cpu, and I changed the
cooler for one better suited to my fx8120, an Arctic Cooling Freezer 13,
previously I was on Arctic Cooling PRO64, now the temps are these:

Core Temps
Idle: 24/25C°
Load: 48/50C°

Socket Temps
Idle: 36/38C°
Load: 59/60C°

Unfortunately my problem is the heat of the room, which is constantly around 30C°.
The case is already completely open.

 

[manu2b]

Now i'm running at stock (3.1 ghz), I've disabled the TurboBoost feature
in the BIOS that caused me the overheat at the cpu, and I changed the
cooler for one better suited to my fx8120, an Arctic Cooling Freezer 13,
previously I was on Arctic Cooling PRO64, now the temps are these:

Core Temps
Idle: 24/25C°
Load: 48/50C°

Socket Temps
Idle: 36/38C°
Load: 59/60C°

Unfortunately my problem is the heat of the room, which is constantly around 30C°.
The case is already completely open.

Same problem here in Corsica: ambient 32°C...

First thing I would do is trying to find the lower vCore for the stock clock: usually, auto vCore gives more juice than necessary.

Go from there: load with Prime95, check your temps after 15/20 min and slowly increase Freq/vCore until you reach the temp limit.

You can do quick and dirty 20 mins tests until you reach max temp in order to find a "quasi" stable OC..

When done, run at least a 6 hours prime95 blend test. If it fails, decrease the clock by a 100Mhz, and run a 6 hours test again: it should be OK.

EDIT: when OCing, disable turbo and use multi instead.

 

[Bubba-Hotepp]

Now i'm running at stock (3.1 ghz), I've disabled the TurboBoost feature
in the BIOS that caused me the overheat at the cpu, and I changed the
cooler for one better suited to my fx8120, an Arctic Cooling Freezer 13,
previously I was on Arctic Cooling PRO64, now the temps are these:

Core Temps
Idle: 24/25C°
Load: 48/50C°

Socket Temps
Idle: 36/38C°
Load: 59/60C°

Unfortunately my problem is the heat of the room, which is constantly around 30C°.
The case is already completely open.

Can you post a screenshot of HWMonitor and Speedfan?

At stock settings what does your MB default the CPU/NB speed and voltage to?

 

[manu2b]

Speedfan?!?

EDIT: I apologize. I just learned it has been designed by a NASA thermodynamics engineer!

 

[manu2b]

On a more serious note, any update on the Prime tests?

 

[Bubba-Hotepp]

Uploaded with ImageShack.us

That should be your CPU Temp (TCase) if I remember correctly for Gigabyte boards but you'll have to verify that by comparing it to the "CPU Temp" listed in your bios as it's hard to verify without seeing results from other programs. The best way would be to use a utility from Gigabyte (if they make one) under windows that lists "CPU Temp" and directly compare it to HWMonitor. As you can see in the screenshot you've posted you've blown way past the max operating temp specified for the 8120 which is 60C (one of the dangers in listening to people tell you "core temps" are what you should be watching). With air cooling you'll need to dial back those settings a bit. One of the things to check is to see if your MB is defaulting the CPU/NB V too high. Stock is 2000MHz @ 1.175V but I've seen people post screenshots of their MB's (mainly asus but some others as well) defaulting to 2200MHz and 1.25V. The added voltage there will also add to the temps as well.

 

[EarthDog]

Thats the problem with applications like Speedfan and Hwmonitor as opposed to Coretemp. In Coretemp there is one temp which is the only one we need to care about frankly. We have all been fine using Coretemp for years with great success. I have seen the emails, and done my best to decipher them, but am coming up empty.

I will look at the emails again and contact the maker of Coretemp to see if he can help get to the bottom of this.

 

[manu2b]

Earthdog, you are wrong!

You must check CPU TEMP!!! Not Core TEMP!!!