• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Will OCing an A64 show a significant gain over HSF?

ScottFern

ScottFern

Diamond Member
Ok, I went from stock cooling to Zalman HSF to a Koolance watercooling setup. Now I know my Koolance case is not exactly what we would call top end but its nice. However, I talked to a friend with a dangerden setup with almos the exact same specs and he reported the same temps as me.

Now I am just posing the question....Is watercooling show a significant improvement over a nice zalman or thermalright HSF on a A64?
 
Your title is a little misleading...

Air cooling has come a very long way. You really need to put some funds into your WC setup to see the bigger temperature deltas. The big difference with water is that your idle and load temperatures will not fluctuate as much. Idle temps will be mostly the same, given a moderate WC system vs high end air.
 
I changed the topic hopes its a little easier to figure out. Anyway I am fully aware my watercooling is nowhere near top end but for what i paid for its on par with a nice Zalman cooler....which still is a little disappointing. My idle is around 35C and my load under prime is 52C which is a very big swing in my opinion. I dunno what to do anymore when it comes to buying better cooling.......If dangerden setup is on par with this, then I have to ask why? I should have just stayed with my zalman and left it at that.......

Need opinions!
 
Honestly, unless you have a hardcore cooling set up and are going for extreme overclocks, water cooling isn't worth it. As beatle said, air cooling has had some major developments, and there are many good cool and quiet solutions out there thanks to thermalright, Zalman, and the likes. Your overclock isn't that extreme, so I'd sell your water cooling and go back to a Zalman. For $40, the zalman is just as good as the $150+ Koolance kit.
 
Actually, no heatsink can touch a watercooled setup such as the koolance. Scottfern, I don't know know what to say but except the fact that your readings are way off. I'm telling you this because my setup temperture changes everything I reboot with my h2o setup. you can't trust the neo2 temperture sensor. My philosphy is that if you can overclock and it doesn't break with your h2o you're doing a fine job. I can jpretty much guarantee that for all the people with high end heatsink that their numbers are higher than report. Just because a your software monitor says you're at 35 load doesn't really mean it.
 
Your right, the mobo sensor is wayyyy off. However, how can we ever expect to get an accurate reading? I just want a base line to compare to. What cooling are you using and what method of monitoring are you using?

Right now I am using Speedfan......which is using the Winbond W83627THF sensor. So I dunno? Let me know what the best option is to use for monitoring?
 
I compared all 3 with my neo2, including ntune by nvidia, msi corecenter, and speedfan, all of them report the same reading as all the other software.
 
Don't people with watercooling worry about leakage? Tiny drops will kill your thousnads of dollars worth of investment.... Doesn't that register with anyone??? Is couple degrees drop of temperature or 10% or whatever % increase you get from your oc's worth the risk? Why not spend the $ buy a better cpu and be done with it? I am lost on the whole h2o thing.....

 
Originally posted by: Idleuser
Actually, no heatsink can touch a watercooled setup such as the koolance. Scottfern, I don't know know what to say but except the fact that your readings are way off. I'm telling you this because my setup temperture changes everything I reboot with my h2o setup. you can't trust the neo2 temperture sensor. My philosphy is that if you can overclock and it doesn't break with your h2o you're doing a fine job. I can jpretty much guarantee that for all the people with high end heatsink that their numbers are higher than report. Just because a your software monitor says you're at 35 load doesn't really mean it.
Click to expand...
If that's the case, doesn't that also mean that the temperatures with watercooling is higher than the software/hardware reports? If not, please tell us what godly hardware/software you use? I use two hardware monitors and two software monitors and they are all consistent.

You come across as a watercooling fanboy trying to justify $200+ watercooling setups for single-digit deltas. In a few weeks when I post the results of my XP-120 with one of the new 200+ CFM 120mm Deltas, we'll see how my < $100 cooling setup compares with $200+ watercooling kits.
 
Originally posted by: mrscintilla
Don't people with watercooling worry about leakage? Tiny drops will kill your thousnads of dollars worth of investment.... Doesn't that register with anyone??? Is couple degrees drop of temperature or 10% or whatever % increase you get from your oc's worth the risk? Why not spend the $ buy a better cpu and be done with it? I am lost on the whole h2o thing.....
Click to expand...
The same could be asked of your decision to build your own PCs. You can get a comparable system for a comparable amount, backed by tech support and a warranty.

Nobody watercools because they have to. They watercool because they get a kick out of it; because it's fun and challenging as a hobby in and of itself.

And some of us watercool because we move into tiny Manhattan apartments with our fiances, who go batsh1t when they hear a Delta strapped to an SLK-900....

So don't troll!

-HC-
 
If leaking was that much of a problem then people wouldn't be buying these types of systems. I think the theory goes......If you know what you are doing and assemble it correctly the margain of it breaking is slim to none.....
 
Originally posted by: ScottFern
If leaking was that much of a problem then people wouldn't be buying these types of systems. I think the theory goes......If you know what you are doing and assemble it correctly the margain of it breaking is slim to none.....
Click to expand...

If that were tru, I'd agree that watercooling looks pretty good. But if the repair record of my car speaks to anything, the margin of failure is guaranteed to be there. If an expensive engine can leak oil, ask yourself will a $200 plastic watercooling kit have a better failure rate?
 
Originally posted by: mrscintilla
Originally posted by: ScottFern
If leaking was that much of a problem then people wouldn't be buying these types of systems. I think the theory goes......If you know what you are doing and assemble it correctly the margain of it breaking is slim to none.....
Click to expand...

If that were tru, I'd agree that watercooling looks pretty good. But if the repair record of my car speaks to anything, the margin of failure is guaranteed to be there. If an expensive engine can leak oil, ask yourself will a $200 plastic watercooling kit have a better failure rate?
Click to expand...

Uhh..because you're not DRIVING your computer case around the city?

Sorry..seemed like an obvious answer. It's the car's hoses that leak, not the engine, and the hoses are not worth $200. Plus your computer case is not being subject to the same conditions your car's hoses are (extreme heat, extreme cold, humidity, dryness, all types of weather).
 
Originally posted by: mrscintilla
Don't people with watercooling worry about leakage? Tiny drops will kill your thousnads of dollars worth of investment.... Doesn't that register with anyone??? Is couple degrees drop of temperature or 10% or whatever % increase you get from your oc's worth the risk? Why not spend the $ buy a better cpu and be done with it? I am lost on the whole h2o thing.....
Click to expand...


True. Also if oil leaks out of your car engine it could ruin your thousands of dollars worth of investment in your car.
 
Originally posted by: mrscintilla
Originally posted by: ScottFern
If leaking was that much of a problem then people wouldn't be buying these types of systems. I think the theory goes......If you know what you are doing and assemble it correctly the margain of it breaking is slim to none.....
Click to expand...

If that were tru, I'd agree that watercooling looks pretty good. But if the repair record of my car speaks to anything, the margin of failure is guaranteed to be there. If an expensive engine can leak oil, ask yourself will a $200 plastic watercooling kit have a better failure rate?
Click to expand...

Thanks but I only paid 150 for mines with high quality part. your xp 120 and fans will never be able to touch my watercooling and low noise. You can blast those delta's away while i'm running dual 120mm fans @ 25dba 🙂 It's a given fact that water 30x-40x more efficient than air what part of that concept don't you understand? Btw, for the price you paid for your heatsink and fans it'll probably cost 1/2 of my setup 🙂 Btw, your GPU temps can never touch my maze 4 dangerden. You can run loads at 60+ while i'll keep my 46°C load temps and 40°C with my 6800 OC unlocked to 16x6.


Like I say monitor software will never come close to didoe reading. That means you got to put the diode down to the built in diode of the cpu. PH over at procooling uses this and found out that it's relatively close to a few C degree close to actual temp using a diode on his AXP. Your software monitor means nothing it's only use for a measure of a base line and nothing more. So please before you come bashing me with those comments you should know that software base temps means nothing.

The DFI Nforce 4 reads any where from mid 20C on idle and 35C on load. now tell me why does the temperture varies so much from a msi neo2 to a dfi nforce ultra? That's almost a 10 degree in temperture drop. So are you saying software base is correct? No I didn't think so. I can guarantee that every chipset that needs cooling (ie: northrbidge, cpu) will result in a higher temp when you get under it with real device testing it.
 
How did we get off track? lol. Anyway, whats everyones opinion on the statement: Watercooling an A64 does little to bring down the temps at any significant number but instead barely brings it down 3-5C at max. Yes my theory is flawed, but I am basing it off of my Koolance case (which I bought for ONLY $120 shipped) and a friends dangerden setup and then back to my old Zalman which from what I have seen have been all relatively close in temps.

And for the record the only reason I wanted to try this was because of the smoking hot deal I got on the Koolance case.
 
Didn't mean to get the threat off track. But I thought persuading you away from watercooling may help your choice a bit. What I can offer is that my zalman 7700-alcu can already give me 36 C idle temp at inaudible noise. W/ fan controllers, hardware or software, you do not need watercooling to get a low noise pc.
 
You might get a slightly better overclock out of an A64 with a watercooled rig, but overall you're right -- the A64 simply doesn't pump out enough heat to benefit all that much from cooling above a certain level. It all depends on the platform. If you had a Prescott it would be a different story. Myself, I've been able to bump the voltage of my AXP Barton and push the clock to 2.5GHz, virtually doubling heat dissipation, with a DD TDX setup.

-HC-
 
Originally posted by: HeaterCore
You might get a slightly better overclock out of an A64 with a watercooled rig, but overall you're right -- the A64 simply doesn't pump out enough heat to benefit all that much from cooling above a certain level. It all depends on the platform. If you had a Prescott it would be a different story. Myself, I've been able to bump the voltage of my AXP Barton and push the clock to 2.5GHz, virtually doubling heat dissipation, with a DD TDX setup.

-HC-
Click to expand...

Yes! A quality response! I have Oc'd my 3500+ to 2.6Ghz (400Mhz difference) and I am starting to come to the realization that watercooling was foolish for my A64 and I should be focusing on air ventilation in the case instead!
 
The thread got off track because we had fan boys going 'round and 'round trying to justify their expenses. Back on topic, you have come to the correct conclusion ScottFern, heat is not the problem in A64 overclocking. Just using a SilentBoost HS with a 92mm panaflo my overclocked 130nm 3000+ is loading at 35C @ 2.35GHz; that's nothing compared to the northwoods or even the prescotts. Again, I suggest finding someone with a P4 and selling your Koolance to them, and go back to your Zalman; depends if you wanna make some cash back or not.

Also to clarify a few things in this thread:
-Using 100 different temp programs means jack because they're all reading from the same hardware sensor. If you want to find a different way of getting temps, use a thermal laser or a different diode and take them from that.
-Watercooling is a little more "dangerous" because there is water near the electric parts. However, when one carefully builds a watercooling set up and throughly researches his design and then tests for leaks BEFORE putting it in his system, the chance for any mishaps plummets.
 
for now I am going to stick with the koolance for the *BLING* factor.....and yeah it frustrates me that I can't get a real read of what the temp is.......so even if I got 30C idle and 40C load everyone could say, that sensor is off!!!! whats the point!?!?
 
They have thermal lasers that are very accurate that you could use, but they're about $100, which I find to be a realy waste just to see how far below safe temps you are.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top