Recommendations for AMD system needed
- Thread starter Duracell
- Start date
Looks great.....but where are the hd coolers??? 
Duracell
1
CPU while inactivity and normal usage: 39 °C (diode and socket)bionic said:
CPU while gaming: 48 °C (diode and socket)
Case: 28 °C
external clock (processor/RAM synchronous) is 196 MHz @ 1,675 V (Prime95 stable (24h torture mode))
resulting CPU clock is 2254 MHz (genuine Athlon XP 2600+, 1916MHz, 166MHz FSB, 512kB Cache)
couldn't reach 200 stable even with 1,725 V;
1046 (220V) sitting on two foam blocks; you have to go nearer than 5 inches with your ear to the pump to hear a peaceful humming;
the only sound you can hear outside the case is the metallic whizzing from the HDDs;
in front in each HDD cage, slow 80mm fans, rpm controlled by the power supply;roadworker said:Looks great.....but where are the hd coolers???
Not bad considering no fans
Im running my barton 2500 at same speed and 1.75v. I only clock the multiplier(13.5) not the fsb though.
Temps on load 38cpu x 45 board(sensor somwhere between nb and mosfets) idle 33x40 (usually with fans running at 5\7v)
Strangely enough my old xp2400 would run with lower temps @ xp3200(166fsbx14mpx1.725)
Ok, sounds good! Bought an mcp600 a while back and were not pleased with the noise, guess ill go for a 1046 when the one i use (hydorl20) croaks.
Anyway! congrats on the new rig bud, looks great and i hope your happy with the result
Btw:
Here you can find a little disconnectbit tool that has support for nforce, might lower idle temps some.
Duracell
1
thanks for the laudation;bionic said:Anyway! congrats on the new rig bud, looks great and i hope your happy with the result
Btw: Here you can find a little disconnectbit tool that has support for nforce, might lower idle temps some.
But what is the benefit of such tools? The temps aren't problematic, or?
Not at all
They "per definition" extends cpu life time and saves a little power.
By now the disconnectbit should be enabled on all athlon compatible boards(by amd recomendation), but ive noticed that asus sometimes dont comply.
Dont know, but there might\should be an bios option to turn it on.
After i swiched to water it doesnt make that much of a difference but when using air i found the difference quite remarkable(im a little sensitive to noise) when idle or "normal" use, fans could be run unhearable(4-7v)
Also the psu will be get less load, which makes a difference with automatic temp\fan controll.
Also its nice with low temps, me thinks
They "per definition" extends cpu life time and saves a little power.
By now the disconnectbit should be enabled on all athlon compatible boards(by amd recomendation), but ive noticed that asus sometimes dont comply.
Dont know, but there might\should be an bios option to turn it on.
After i swiched to water it doesnt make that much of a difference but when using air i found the difference quite remarkable(im a little sensitive to noise
) when idle or "normal" use, fans could be run unhearable(4-7v)Also the psu will be get less load, which makes a difference with automatic temp\fan controll.
Also its nice with low temps, me thinks
The disconnect feature is no free lunch though, the price of that low power sleep, is that the CPU cannot snoop for DMA durin sleep (unlike Socket 7 Intel, AMD, Cyrix and all current Intels) - so every time the HALT engages, the writeback is flushed, and the whole of the CPU cache is voided, so there is a small burst of RAM write, at HALT, and a heavy RAM read on unhalt - this can freeze out the soundcard, so that it clicks, and would probably be bad for video playback and capture
Found it!
http://www.grassomusic.de/english/frameset.htm?http://www.grassomusic.de/english/acpi.htm
Rather heavy going at times
http://www.grassomusic.de/english/frameset.htm?http://www.grassomusic.de/english/acpi.htm
Rather heavy going at times
Mmmm, tasty
must say though, it escapes me as to why amd now instruct mobo manufacturers to implement it and turn it on by default(to recieve amd certification). Is this a mere(and incomplete) workaround for the high temps.
Hmm, Also it seems the hlt command can cause voltage fluctuations. So for oced systems this can cause instabilities too.
But then again, these tweakers can be easily paused(if the disconnect function is turned off in bios). And the drawbacks should be eliminated.
must say though, it escapes me as to why amd now instruct mobo manufacturers to implement it and turn it on by default(to recieve amd certification). Is this a mere(and incomplete) workaround for the high temps.
Hmm, Also it seems the hlt command can cause voltage fluctuations. So for oced systems this can cause instabilities too.
But then again, these tweakers can be easily paused(if the disconnect function is turned off in bios). And the drawbacks should be eliminated.
Probably need it to qualify for "energy star" - though given the issues with it, it would be better implemented ONLY in full sleep mode, rather that as idle cooling.
I used to use "RAIN" on my old Cyrix - and that certainly cooled the old linear regulator on the motherboard!
Stability and noise problems were noted even then.
I also found that, while none of the "software coolers" supported my chipset, a settings bit could be changed with WPCREDIT to activate auto-disconnect mode.
Great cooling when idle, but stability took a dive.
Soft cooling, also does NOT help overclockers - you overclock, because you're pushing for performance, and if you're using that performance, there is NO CPU IDLE TIME!
I used to use "RAIN" on my old Cyrix - and that certainly cooled the old linear regulator on the motherboard!
Stability and noise problems were noted even then.
I also found that, while none of the "software coolers" supported my chipset, a settings bit could be changed with WPCREDIT to activate auto-disconnect mode.
Great cooling when idle, but stability took a dive.
Soft cooling, also does NOT help overclockers - you overclock, because you're pushing for performance, and if you're using that performance, there is NO CPU IDLE TIME!