Zap
Elite Member
- Oct 13, 1999
- 22,377
- 7
- 81
GAH! Tried to reply earlier, had stuff typed up and then forum stopped responding. :|
2) I recently got this Antec 92mm TriCool fan that I'm pleased with. I also have this Panaflo 92mm "M" fan that pushes lots of air with a lot of "whoosh" noise, could be slowed down plus there is a slower "L" model. If you believe their ratings, this Thermaltake 92mm Silent Cat fan gives 52CFM at 21dba. I also have one of these, came with my Thermaltake Silent Tower. It was anything but silent at full speed, but is fairly quiet at lower speeds. The thing is that fan manufacturers seem to each have their own methods of measuring performance so it is really difficult to compare fans by just using their published numbers. The thing is... how quiet is quiet enough for you? I'm personally a fan of quiet computers. Compared to all my RL friends my systems are completely inaudible if their systems are running in the same room. In any room with background noises (for instance desk fan or A/C, or people talking) I do not notice my systems running. However, I do not dare put myself in the same category as those frea^H^H^H^H enthusiasts at SPCR because while they seem to not tolerate any noise whatsoever I'm content with quiet enough to not hear under normal conditions. BTW, a friend of mine had some 80mm Vantec Stealth fans and though silent, they didn't feel like they were moving any air. The other thing about noise is that even fans which do produce noise can produce a different quality of noise. For instance, when SPCR tests a fan undervolted they will report on bearing "clicking" noises because some fans produce it and some don't, and it bothers some people but not others. Some fan motors make a high pitched whine. Some fans just sound like the "whoosh" of air. I've personally concluded that for my own purposes any decent fan will be quiet enough with a speed controller, and larger diameter fans will always push more air than smaller diameter fans - all else being equal.
4) If you're confident that you only applied a thin layer of compound, then I think paying for AS5 and reapplying compound will be a waste of time and money. If you want to spend time and money and really want low noise with better cooling, may as well sell the Zalman 7000 cooler and get something better. You'd probably get a much better gain off swapping to an XP-90 heatsink with decent fan with the same Zalman heatsink paste over keeping the Zalman heatsink/fan and getting AS5.
5) If it gets way too hot, power cuts and you'd experience a system lock. I've seen Intel systems (P3 and P4) lock due to heatsinks falling off. Don't recall ever seeing a CPU die from that. The P4 at 3GHz+ have a version of SpeedStep which cuts multiplier down to 14X, making an unoverclocked CPU into a 2.8GHz chip. There is another feature of the P4 which is throttling - been available since... maybe since the first P4 Willamette. This feature cools down the CPU by switching it on/off. People think it reduces clock - effectively it makes your system slower but clock stays same. What happens is that in tiny increments probabaly measured in nanoseconds, the CPU switches from FULL SPEED to COMPLETELY STOPPED and back to FULL SPEED. The percentage of time CPU is stopped is usually somewhere from 20% to 80% and I think can be controlled by BIOS. This is kind of like a fan speed controller that uses PWM - instead of a fan being UNDERvolted, the fan runs at FULL voltage, but not all the time, instead full voltage is pulsed in fractions of a second coupled with no voltage. End result is fan runs slower. You can't tell, unless you have an LED fan (as I recently found out
) and you get a nice strobe light effect. Oh yeah, plus the PWM controller doesn't get as hot. Back to the CPU throttling - typical benchmark and measuring programs wouldn't report the throttling because they can only make measurements while the CPU is in an ON state and not OFF state, and since when in ON state CPU is running full speed, software will report full speed. I think that's why throttlewatch software was made to be able to report throttling. While your Deleron (word I coined last year on these forums to mean Celeron D) doesn't do SpeedStep, it should support throttling.
6) Lots of complaints when the IS7/IC7 boards came out and reported much MUCH higher temperatures than competing brands. A bunch of complaints and finger pointing later, Abit released a BIOS update that lowered reported temperatures, though temperatures still reported higher than other brands. Abit still claims they are generating "proper" temperatures based on Intel guidelines. Users have found that regardless of temperatures, the same CPU feels the same temperature to the finger whether on an Abit board at 50º or another brand at 40º.
Of course I could be totally wrong... spending too much time on forums will do that to a brain...
2) I recently got this Antec 92mm TriCool fan that I'm pleased with. I also have this Panaflo 92mm "M" fan that pushes lots of air with a lot of "whoosh" noise, could be slowed down plus there is a slower "L" model. If you believe their ratings, this Thermaltake 92mm Silent Cat fan gives 52CFM at 21dba. I also have one of these, came with my Thermaltake Silent Tower. It was anything but silent at full speed, but is fairly quiet at lower speeds. The thing is that fan manufacturers seem to each have their own methods of measuring performance so it is really difficult to compare fans by just using their published numbers. The thing is... how quiet is quiet enough for you? I'm personally a fan of quiet computers. Compared to all my RL friends my systems are completely inaudible if their systems are running in the same room. In any room with background noises (for instance desk fan or A/C, or people talking) I do not notice my systems running. However, I do not dare put myself in the same category as those frea^H^H^H^H enthusiasts at SPCR because while they seem to not tolerate any noise whatsoever I'm content with quiet enough to not hear under normal conditions. BTW, a friend of mine had some 80mm Vantec Stealth fans and though silent, they didn't feel like they were moving any air. The other thing about noise is that even fans which do produce noise can produce a different quality of noise. For instance, when SPCR tests a fan undervolted they will report on bearing "clicking" noises because some fans produce it and some don't, and it bothers some people but not others. Some fan motors make a high pitched whine. Some fans just sound like the "whoosh" of air. I've personally concluded that for my own purposes any decent fan will be quiet enough with a speed controller, and larger diameter fans will always push more air than smaller diameter fans - all else being equal.
4) If you're confident that you only applied a thin layer of compound, then I think paying for AS5 and reapplying compound will be a waste of time and money. If you want to spend time and money and really want low noise with better cooling, may as well sell the Zalman 7000 cooler and get something better. You'd probably get a much better gain off swapping to an XP-90 heatsink with decent fan with the same Zalman heatsink paste over keeping the Zalman heatsink/fan and getting AS5.
5) If it gets way too hot, power cuts and you'd experience a system lock. I've seen Intel systems (P3 and P4) lock due to heatsinks falling off. Don't recall ever seeing a CPU die from that. The P4 at 3GHz+ have a version of SpeedStep which cuts multiplier down to 14X, making an unoverclocked CPU into a 2.8GHz chip. There is another feature of the P4 which is throttling - been available since... maybe since the first P4 Willamette. This feature cools down the CPU by switching it on/off. People think it reduces clock - effectively it makes your system slower but clock stays same. What happens is that in tiny increments probabaly measured in nanoseconds, the CPU switches from FULL SPEED to COMPLETELY STOPPED and back to FULL SPEED. The percentage of time CPU is stopped is usually somewhere from 20% to 80% and I think can be controlled by BIOS. This is kind of like a fan speed controller that uses PWM - instead of a fan being UNDERvolted, the fan runs at FULL voltage, but not all the time, instead full voltage is pulsed in fractions of a second coupled with no voltage. End result is fan runs slower. You can't tell, unless you have an LED fan (as I recently found out
) and you get a nice strobe light effect. Oh yeah, plus the PWM controller doesn't get as hot. Back to the CPU throttling - typical benchmark and measuring programs wouldn't report the throttling because they can only make measurements while the CPU is in an ON state and not OFF state, and since when in ON state CPU is running full speed, software will report full speed. I think that's why throttlewatch software was made to be able to report throttling. While your Deleron (word I coined last year on these forums to mean Celeron D) doesn't do SpeedStep, it should support throttling.6) Lots of complaints when the IS7/IC7 boards came out and reported much MUCH higher temperatures than competing brands. A bunch of complaints and finger pointing later, Abit released a BIOS update that lowered reported temperatures, though temperatures still reported higher than other brands. Abit still claims they are generating "proper" temperatures based on Intel guidelines. Users have found that regardless of temperatures, the same CPU feels the same temperature to the finger whether on an Abit board at 50º or another brand at 40º.
Of course I could be totally wrong... spending too much time on forums will do that to a brain...
Facebook
Twitter