Anyone using the Zalman Flower P4 CNPS6500-ALCU?

[arman68]

I am thinking of changing my existing heatsink which is a Coolermaster IAC-002. It provides reasonable performance (48 celsius on CPU at full load with P95 on a 2.0a @ 2.5GHz), but I have reached a point where my RDRAM is getting hotter than CPU.

If I want to increase the FSB I would have to drop down the RDRAM to 3x.

An attractive solution would be to use the Zalman Flower P4 CNPS6500-ALCU with 2 fans: 1 over the CPU + Nothbridge, and another one above the RDRAM.

I anyone using it on the TH7-II? How does it fit, and what is the performance?

Thanks.

.aR

 

[Cooler666]

I'm using it and its pretty good, i have two fans blowing into the Zalman flower, i can reach up to 140FSB without a problem.
At 133FSB i have the RDRAM set at 3x, not stable at 4x as i don't have a fan blowing on the RAM.

 

[batboy]

Actually, this is the first favorable report I've heard about the Zalman. Most reviewers don't think much of it. I wonder if it's because you have two fans? Maybe they need more air flow?

 

[Cooler666]

I have a Zalman 120mm and a YS-Tech 80mm fan blowing on it, with not much noise at all. All i can say is that the CPU rans at least 10 Celcius colder than with the stock heatsink. The Zalman 120mm fan is not even running at full speed - its on quiet mode.
The Swiftech 478 is better i'm sure but it's noisier and much more expensive, although its for the more serious overclocker for FSB's of 140+.

How it compares to the Sunflower or Volcano i'm not sure.

 

[arman68]

Cooler666,

Are both fans fitted at the end of the Zalman bracket? And if so, does the heatsink comes with enough kit to attach 2 fans?

One last thing, would you mind doing an experiment for me:


1. Turn one fan off, and direct the remaining one on the CPU

2. Place the thermal sensor that comes with the motherboard between 2 RIMMS (I find it provides a fairly accurate temperature report for the RAM)

3. Stress test your system (Prime95 generates the most heat)

4. Take a temperature reading for your RAM and CPU

5. Reconnect the second fan and position it over the RAM (does it reach?)

6. Repeat the stress test and take a temperature reading.


I will be very interested in seeing how efficient it is. FYI with vCore=1.725, FSB=125, RDRAM=4x on a P4 2.0a, under P95 torture test I get CPU=48 celsius and RMA=52 celsius

Regards

.aR

 

[arman68]

^bump^

Cooler666, have you had a chance to try my experiment?

I intend to buy one of those soon and replace all the fans in my case with quiet ones.

Wish me luck in making my case 'silent' but still overclockable

 

[Cooler666]

i've tried this a long time ago, i don't won't to ruin my overclock now, but i'm going to try this again anyway, i'll check if the fan reaches the RAM

 

[LarryJoe]

arman68,

I have the same set-up as you and have the same limitations. It is definitely your RAM and it may not just be heat, but this is where you should start. 48C/52C full load is not bad at all given 1.725 volts. I don't think this contraption is going to help you. Take an 80MM fan and point it at your RAM and see what happens before you go through the effort and expense of changing heatsinks.

I am at 2.5/500 at 1.625 actual volts. This is the highest my ram can go at 4X. I can run at 2.667/400 (3X), but it requires almost 1.9 volts to become stable, causing extreme temps that air cooling can't handle, at least my sir cooling (AVC) anyway.

Overclocking is fun, but take my advice here, be happy with 2.5/500, going higher is a lot of work and you will end up with XP errors and reinstallations. I just finally gave up trying and started using my new system constructively. Getting to 2.5 was effortless, going higher has been a headache.

LJ