I returned my 8400 today!!!!

Page 9 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
I am a big AMD diehard, so I'm going to lay it on the line.
Nobody makes a sensor, and reader combo that works, without having some type of calibration.
No desktop has a method to calibrate it's sensors.
If you think your computer is giving accurate temps, you are sadly mistaken.
On a single core chip, there will be huge variations in diferent parts of the chip. Dual cores just multiply those variations.
If you want to know how hot your chip is, run F@H. If it cant complete a work order, it's too hot. If it always completes, it's running nice and cool.
 
Endyn is right. Hot spot are in different areas. Later steppings help those hotspots along with other issues such as errata.

F@H is a good way to test it. Sois Prime95. If you can run Prime95 for 8+ hours(mine was about 11 hours then I stopped) can be assured your CPU is not over heating. I use 2 programs, a thermal sensor on my heatsink and the BIOS. And so far my CPU stays at 50-55c when under load with Prime95 even after 10 hours.

But in comparison to single cores having a dual/quad run as cool or cooler is amazing. But thats the amazing thing about technology. It gets smaller allowing you to add more and do the same thing only faster and better.
 
From a consumer point of view the chip doesn't have issues.

However, from a overclocker's view, the malfunctioning temperature sensor is an issue.

It's not a show-stopper, but it is annoying and lower the value of the chip to OCers, than and that there is less overhead in the higher clocked chips.
 




This sounds like the rant of an AMD mole. More than 1/2? What a joke. I can't believe people actually lend credibility to posts like this by responding with logic.
 


TC prove that the Wolfdale sensor is the one at fault here? I can just as easily say it is the software that is messed up.

Nothing I have seen in any of the multitude of threads has indicated it is the problem of the processor thermal sensor. I think you are spreading FUD. My opinion.
 
I've read most of this thread and would like to add my experience with the 8400.

I had an E6600 @ 3.7ghz with 1.575v. Under watercooling with several other items in the loop, temps were 37'C idle and 59'C under load. These are the temps in Core Temp. Intel TAT reported similar and Asus probe was a few degrees higher as per usual.

NOW installing the E8400 again under water with the E6600 having been in 30 seconds prior, my IDLE temps are HIGHER than the load temps above.... Without even touching the bios leaving all settings at default and the CPU at 3Ghz, it's idling at 44'C and loads in the high 60'C. Reseated, re-applied TIM and still the same. Put E6600 back in and temps again are WAY lower than the 8400. Put 8400 back in, temps sky high again. Insane. At default voltage the E6600 would idle at 24'C and load at around 40'c!!!!!

There IS an issue. No other chip has had this kind of press for high temps. What some of you appear to have forgotten is that the 8400 has only just been released so there are hardly any out in the wild and to have this many reports at this stage is WORRYING.

I have the chip at 4.4Ghz and it's perfectly stable at 1.575v using Orthos for 8 hours. Idles at exactly the same temp as stock.....weird. Loads at 75'C. the E6600 never went above 60'C at this voltage.
 



Que? The supposed problem isnt high temps, it is temps being read incorrectly. I thought you said you read the whole thread?

And BTW, these beauties are going for $300+ on ebay, and people are buying them. Crazy to think I got mine for $189. The bad news is, that shows Intel that people are willing to pay this much for this chip.
 



I have no proof of this condition.
 


i quit this site since the banned me to teletubies for 2 weeks so this is not me!


first off the cpu works fine - there is nothing in intels product info that says temp sensors are required

second - the stock hsf is totally crap - so warning to anyone toss it, the all aluminum design is to save shipping money and eliinate the costly copper core


finally, you can estimate the cpu temperature form the mobo temp

when first boot you see it says 10 or 15c then both the mobo and cpu heat up to 20c or so. if you know anything about mobo temps you know your off by 10-15c

when you heat the cpu at 100% load you see mobo at 20-25c and cpu at 40c this equals 40-45c mobo and 60c cpu

this post sucks

this is not me since i quit the site!
 
So is my e8400 doing it aswell? my idel temps (with a Zalman 9500led) are about 38 on a cool day with windows open and reaches 48-50 full load. $h!tty sensors?
 


I never said the issue was high temps in general, although I don't see many people with reading of -60'C and if you read my post I didn't say I'd read the whole thread, only most of it and simply wanted to share my experience especially as I'm coming from using an E6600 which is a similar chip that is supposedly running hotter. I only wanted to address comments I had read that are quashing both notions as crap ie there isn't an issue with temp sensor programs or the chip sensor itself it's simply people installing things incorrectly.
 
How exactly do you install Core Temp incorrectly? The only thing that seems to fix the erratic temperature sensor behavior is a BIOS flash if the motherboard maker has addressed the issue.
 
Well my 8400 stock ran idle at 30c and up to 42c under load with hours of orthos and now have it at 4.0Ghz running 35c idle and 52c load all on water btw. I have checked this using core temp and hw and they seem to be within 2c when checking it with infrared thermometer and water loop never gets over 30c. I will say speed fan is off by 10c or so. Just my 2 cents
 
Just like AMD is trying to fix the Phenom so it works 100% out of the box Intel will fix this new chip just the same. I fail to see the big debate. Its not like these things never happen. What I realy can not understand is what would lead people to think these new chips would not degrade faster when OCed. Its the natural way things work with CPUs. I mean I could go put some high end racing fuel in my van and make it go pretty fast just like I can push these new Intels to the max on air. I highly doubt my van would last long enough to make it worth doing.

OCers need to realise that you have little to no place in this market. I am down with having a hobby but I do not and never have considered OCing as a selling point and honestly neither does the industry as a whole.
 
Racing fuel with its higher octane won't increase the HP of your van unless you also increase the compression ratio, but the analogy is valid. If you add a nitrous oxide system with a 110HP increase it will affect the longevitiy and reliability of your van.
 



My thoughts exactly! That chip wont last very long...and then you are going to make a post about how terrible the Wolfies are, even though you were running it at 1.8v.


Do some reasearch on the stock Vcore on these, and how these smaller chips cant handle that much power. You shouldnt need that much anyway for 4ghz.
 
E8400 stock voltage (VID) is likely in the range of 1.125V~1.15V. By putting 1.8V into that chip, you're like injecting 5 gallons of steroids into your body.
 
You guys read the post about the Anandtech review of the E8400? From what I gather it is actually the mobos that are reading the temps wrong and not a bad sensor. Or at least that is what I got from it.

But DAM 4GHz on 1.18v? I wish my Q6600 would do that. That would be one hell of a beast right there. 😀
 


It doesn't matter if its the mobos or CPUs fault, people will still bitch about the temps because 'OMGZ ITZ GUNNA BLOW UP MY CPU MAN!!!' :lol:
 
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom