View Full Version : x58 chipset
KaptCrunch
01-05-2011, 10:59 PM
i've noticed that intel say 34c is optium rage, intel 320840.pdf (http://www.google.ca/url?sa=t&source=web&cd=1&ved=0CBcQFjAA&url=http%3A%2F%2Fwww.intel.com%2FAssets%2FPDF%2Fde signguide%2F320840.pdf&ei=HUklTaufB5G5nAfFmsntAQ&usg=AFQjCNGlkFisdR42nHTjl90Rq7S2Ym2UPA)
was hitting 64c on stock cooler. so added a pelt 5v@2.5amps = 12.5 watts
now noticed differance seems more responscive speed of doing things now then b4
also fresh format of w7 64 due to worm
ablatman
01-06-2011, 10:28 AM
Shit, my other 920 hit ~100C on stock cooler.
Snyxxx
01-06-2011, 12:03 PM
my 950 also hit 100C on stock HSF. It started throttling exactly at 100c.
He is talking about the X58 chip set not the cpu. I think. At least that's what the .pdf is talking about.
DiS3MBoW3LMeNt
01-06-2011, 12:34 PM
The Chipset on my three way sli EVGA ran regularly at 60c plus till it died after about a year and a half. Got it rma'ed and the replacement board runs in the 50c range. If I could find a Chipset block on the cheap to get it wet, I would. When I get the 920 up to 4.5ghz it stays above 65c.
Mine sits at about 23c but i have cold air pumped in to the case 24/7. Asus P6X58D-E
Snyxxx
01-06-2011, 01:17 PM
Table 3-2 shows the range to be 5°C to 100°C. I could not see where it said 34°C was optimal.
My Rampage III Formula on a test bench in 23°C ambient hovers around the 60°C~65°C range when running Prime95. About the same for both Northbridge and Southbridge. No air flow at all since watercooling CPU and GPUs.
When using the POS Intel stock HSF, these temperatures were much better due to the air flow from the HSF.
I cant get the link to load but I think it was on the 4th page or section in a table.
KaptCrunch
01-06-2011, 03:29 PM
yes was talking chipset, courious if you gays/gals can get same resaults
just that x58 is the gateway of traffic for MB, for heat = resistance = choke
synxx i thought i read that , 34 is were i'm @ now and sure beats stock HS
funny now i see the ich10r chip is climbing now, gonna toss pelt on it to
ablatman
01-06-2011, 04:52 PM
my 950 also hit 100C on stock HSF. It started throttling exactly at 100c.
I never noticed my 920 throttling when it hit 100C, and thus don't believe that it did...
Snyxxx
01-06-2011, 06:14 PM
I never noticed my 920 throttling when it hit 100C, and thus don't believe that it did...
I watched it happen in real time with CPU-Z. As soon as it went over 100°C, the multiplier would start dropping to reduce the frequency. I did not believe what I was seeing, but it was indeed happening and repeatable.
scraps
01-06-2011, 08:45 PM
Snyxxx, same shit happened with my 920 C0 when I first got it, on stock HSF. Load them cores up, watch temp rise to 100C the multiplier drop drastically, I guess cause it knew it was only gonna get hotter. This was on a DFI micro board around when the chip sets hit. I believe that their was an option in the BIOS to disable the thermal throttling, but I wouldn't have turned it off, and I can't check as the board has been dead for quite some time now. Replaced stock HSF solved that problem, and was only solution as Intel wouldn't RMA or send me another hockey puck. When I called them they gave me the "Oh your case must not be constructed for the thermal requirements" or whatever the fuck they said, In which my reply was that it was on my fucking table with a house box fan blowing across it. Then Big Daddy show'd me this place and got me addicted to :crack: And I've never looked back.
Now as far as the x58 chip set cooling, I am with Kapt that cooling it should help with alot of the things your computer does. But the article says you should plan to be cooling for 12-26 watts min-max on the chipsets, but idk wtf it really means because they say it is not the maximum watts that the chip can consume or put out. I like the article, and find it neat as I didn't know Intel publicly posts articles like that.
Snyxxx
01-06-2011, 10:53 PM
Snyxxx, same shit happened with my 920 C0 when I first got it, on stock HSF. Load them cores up, watch temp rise to 100C the multiplier drop drastically, I guess cause it knew it was only gonna get hotter. This was on a DFI micro board around when the chip sets hit. I believe that their was an option in the BIOS to disable the thermal throttling, but I wouldn't have turned it off, and I can't check as the board has been dead for quite some time now. Replaced stock HSF solved that problem, and was only solution as Intel wouldn't RMA or send me another hockey puck. When I called them they gave me the "Oh your case must not be constructed for the thermal requirements" or whatever the fuck they said, In which my reply was that it was on my fucking table with a house box fan blowing across it. Then Big Daddy show'd me this place and got me addicted to :crack: And I've never looked back.
Now as far as the x58 chip set cooling, I am with Kapt that cooling it should help with alot of the things your computer does. But the article says you should plan to be cooling for 12-26 watts min-max on the chipsets, but idk wtf it really means because they say it is not the maximum watts that the chip can consume or put out. I like the article, and find it neat as I didn't know Intel publicly posts articles like that.
Yes, I only used the stock HSF to confirm the chip was working. Of course, the same day I put it under water. Much better.
In regrads to the X58 chipset temperatures, our family computer has a 780i chipset and Everest shows 98°C and 96°C for NB/SB temps. It has ran like this for 12 hours a day for a few years without missing a beat. Not overclocked BTW.
I do believe you can go up to 100°C without issue. I think the article did mention you do shorten life with higher temps (maybe ambient) though.
KaptCrunch
01-07-2011, 07:20 AM
i noticed that too with cpu 920 do and DFI for the volts were bouncing too
also concider that DFI uses digital and rest use anolog vreg's on MB's
yes cooling hardware helps the life like SIS, ALi died out of the blue for they never used any HS's
Floyd
01-07-2011, 07:24 AM
our family computer has a 780i chipset and Everest shows 98°C and 96°C for NB/SB temps. It has ran like this for 12 hours a day for a few years without missing a beat. Not overclocked BTW.
Wow! the NB on my 630i micro board gets so hot you cant touch it, After I install a new NB cooler ill see what happens.
ballz0r
01-07-2011, 07:26 AM
Does the cooler chipset let you hit higher CPU or ram clocks kapt?
KaptCrunch
01-07-2011, 07:50 AM
good question Ballz0r
haven't checked for not loaded cpuz or benchmarkers software yet
but with same settings in bios(148mgz), things are faster/smoother than before on stock HS cooling
like loading games were jittery now seems steady flow and fast
just want to confirm my findings with x58 cooling thats all, for i hate a slow machine
KaptCrunch
01-08-2011, 12:06 AM
yip still jumps 1.29 - 1.24v about 200mgz in cpuid very so often
guess its intel green mode working
KaptCrunch
03-12-2011, 09:06 PM
whats with intel with RST drivers and windows .... its the third time starting from ground up
fricken system frooze while play a game then reset to find all data off hdd and blank
wonder if x58 chipset is a sandy bridge in cammo
cyriene
03-12-2011, 10:10 PM
Were you running raid with Intel chipset?
FaRKle!
03-13-2011, 12:19 AM
Were you running raid with Intel chipset?
The X58 northbridge shouldn't really affect this as RAID is run off of the ICH10R southbridge.
Does the cooler chipset let you hit higher CPU or ram clocks kapt?
Technically it can since the PLL of the X58 system is on the northbridge. This in turn should allow you to bump up the BCLK a bit higher, of course there are other factors such as the CPU itself and how well it'll clock natively. Looking at people's OC's, more often than not the CPU is limiting the OC rather than the northbridge not being able to push a high enough BCLK.
This doesn't apply to Sandy Bridge since the PLL was moved onto the CPU die. Cooling the H/P67 and the Z68 chipsets shouldn't improve OCing at all.
As for RAM clocks the memory controller for the X58 system is on the CPU itself, so the northbridge has no effect on it.
KaptCrunch
03-13-2011, 12:39 AM
i tried both raid and ahci same resaults after 2-3 months mounted w7 pro 64bit
KaptCrunch
07-23-2011, 09:54 PM
Peltier update with summer temps and humidity/dew points has caused condenation under chip, gettimg weird errors and BSOD.
been down revising chiller putting summer winter/switch and armourlite
pics coming
http://www.hardwarecanucks.com/forum/attachments/water-cooling/14910d1319662371-dfi-jr-x58-caps-interfearence-img_3650cd.jpg
http://www.hardwarecanucks.com/forum/attachments/water-cooling/14909d1319662165-dfi-jr-x58-caps-interfearence-inyt.jpg
vBulletin® v3.7.6, Copyright ©2000-2012, Jelsoft Enterprises Ltd.