Sign in with
Sign up | Sign in
Your question

E8400 now or wait for revision?

Last response: in CPUs
Share
February 11, 2008 2:46:14 PM


Hi,
I was going to pick up an E8400 but then i heard people mentioning irregular temp readings when overclocking.
My question is should i wait for the next batch of E8400's with this detail fixed (supposing they are going to fix it) or is it too minor to worry about and just get one tomorrow?

Thanks for any input and advice in advance.

Panface.

More about : e8400 wait revision

a c 123 à CPUs
February 11, 2008 5:35:06 PM

Its not a major problem but if it were me, I would wait for a G0 type stepping or until the quad corres are out as a G0 type stepping will reduce power usage and heat output to even lower levels.
a b à CPUs
February 12, 2008 10:49:52 AM

I'd wait too ... looks like that batch are like the Phenom B2's ... only AMD were honest and disclosed the issue and offered patches.

Intel tried to cover the whole story up.

Related resources
a b à CPUs
February 12, 2008 10:51:19 AM

Just my purely unbiased comments ... LOL
February 12, 2008 11:23:27 AM

Yeah, ignore all the other users who are OCing these things to 4.2+Ghz on air....
a b à CPUs
February 12, 2008 11:31:06 AM

If the thermal sensers don't work and the system overheats what is to stop it causing a fire and burning down your house?

Pray tell ??

I'd imagine you could take it back under RMA ... Intel are good like that.

Dealers will get a bit grumpy for sure.

Not spreading FUD or anything ...

Spose you can flash the bios but Joe Public isn't that interested so lets wait till it happens and see what the payout is in the courts.

Safety comes first.

Wonder if they will issue a recall??

Responsible corporate citizen and all that...
February 12, 2008 11:43:18 AM

Waiting sounds like the wise move so i shelved my shopping trip but any ideas as to when revisions might be out.
Sorry i'm clueless on this but do most or all CPU's get revision releases?
If so are they at a fairly standard time frame, like 2 or 3 months or is it random depending on issues that need addressing etc?
February 12, 2008 11:47:58 AM

Reynod said:
If the thermal sensers don't work and the system overheats what is to stop it causing a fire and burning down your house?

Pray tell ??

I'd imagine you could take it back under RMA ... Intel are good like that.

Dealers will get a bit grumpy for sure.

Not spreading FUD or anything ...

Spose you can flash the bios but Joe Public isn't that interested so lets wait till it happens and see what the payout is in the courts.

Safety comes first.

Wonder if they will issue a recall??

Responsible corporate citizen and all that...


I think you are really exagerrating on this. I agree you should wait until after a revision that solves the issue is release, but I don't think an overheating computer / processor is going to burn someone's house down.

My question? Is it just E8400s? Will it affect the Q series as well? Made from the same dies. Either way, relax.. Unless someone is storing a gas tank inside their computer I doubt it's going to be a huge deal. I've blown up power supplies and had no issues.
a b à CPUs
February 12, 2008 11:52:37 AM

Well I remember when AMD had no internal sensor for thermal shutdown so the issue isn't one Red and Blue aren't guilty of.

I roasted a few cores.

The socker 462 clips were mongrels sometimes - had one or two that popped off ... disasterous result.

Little red square glowing for 30 seconds or so ... then darkness !!
February 12, 2008 11:58:03 AM

Reynod said:
Well I remember when AMD had no internal sensor for thermal shutdown so the issue isn't one Red and Blue aren't guilty of.

I roasted a few cores.

The socker 462 clips were mongrels sometimes - had one or two that popped off ... disasterous result.

Little red square glowing for 30 seconds or so ... then darkness !!


Did your house burn down? :kaola: 
February 12, 2008 12:04:34 PM

Reynod said:
Well I remember when AMD had no internal sensor for thermal shutdown so the issue isn't one Red and Blue aren't guilty of.

I roasted a few cores.

The socker 462 clips were mongrels sometimes - had one or two that popped off ... disasterous result.

Little red square glowing for 30 seconds or so ... then darkness !!


That was the night the lights went out in Georgia...or wherever he is from
February 12, 2008 12:09:05 PM

onestar said:
That was the night the lights went out in Georgia...or wherever he is from


Don't trust your soul to no backwoods southern lawyer.
February 12, 2008 12:21:50 PM

Reynod said:
If the thermal sensers don't work and the system overheats what is to stop it causing a fire and burning down your house?


I agree with cnumartyr, on that. Your not going to burn down your house or your MB. The thermal trip prevents it from over heating. Most of the claims that the temps are too high at stock speeds. But in order to OC at a desired speed (4Ghz), those people will be hindered from the system shutting down when it's not actually as hot as the sensors reading.

So whether or not if your claim not to spread FUD, you are. :kaola: 
February 12, 2008 12:37:27 PM

tricky trees said:
Hi,
I was going to pick up an E8400 but then i heard people mentioning irregular temp readings when overclocking.
My question is should i wait for the next batch of E8400's with this detail fixed (supposing they are going to fix it) or is it too minor to worry about and just get one tomorrow?

Thanks for any input and advice in advance.

Panface.


Google search shows the only people experiencing temperature problems with the E8400 are overcockers.
There are no news reports of temperature errata from computer tech news soruces.
All reports about temperature problems are posted in computer hardware forums.


February 12, 2008 12:39:42 PM

Grimmy said:
I agree with cnumartyr, on that. Your not going to burn down your house or your MB. The thermal trip prevents it from over heating. Most of the claims that the temps are too high at stock speeds. But in order to OC at a desired speed (4Ghz), those people will be hindered from the system shutting down when it's not actually as hot as the sensors reading.

So whether or not if your claim not to spread FUD, you are. :kaola: 

True it may not burn your house down, but if the temp sensors are under-reading by say 30C, you could quite easily fry your mobo & the CPU.
I hardly think you'd be defending Intel then...
Although I must admit I've had my 6000+ hit about 110C when the heatsink came away (it was my fault! :D  ) and it still works, so maybe they'd survive!
February 12, 2008 1:08:52 PM

LukeBird said:
True it may not burn your house down, but if the temp sensors are under-reading by say 30C, you could quite easily fry your mobo & the CPU.
I hardly think you'd be defending Intel then...
Although I must admit I've had my 6000+ hit about 110C when the heatsink came away (it was my fault! :D  ) and it still works, so maybe they'd survive!


I'm not defending Intel at all. I'm just saying from what I've read. Usually if you see a temp of 126C, your going to shut it down and feel the HS, normally. :heink:  . o O (although I suppose most people would just take the side panel off and touch the HS while its still running)

I've even read people with wild temps (as high as 200C), and touch the HS, and it's basically cool. But then I kinda question if the software is getting the right sensor with the correct reading. Like for example on speed fan when you see -120C stuck all the time. Also hard to say if the thermal trip will actually kick in, even though the software with-in windows reports high temps.

I've also seen a 775 socket pretty much ruined when the thermal trip didn't work when his water pump failed (can't remember if the guy turned something off that he shouldn't have in the bios, when OC'ing), so I know it can do damage to components.

Your lucky AMD ended up putting some kind of thermal protection as well. I still remember those videos of Tom's, where the HS was taken off while a game was running. :sweat:  . o O (still a scary thought)
February 12, 2008 1:21:43 PM

Reynod said:
If the thermal sensers don't work and the system overheats what is to stop it causing a fire and burning down your house?

Pray tell ??

I'd imagine you could take it back under RMA ... Intel are good like that.

Dealers will get a bit grumpy for sure.

Not spreading FUD or anything ...

Spose you can flash the bios but Joe Public isn't that interested so lets wait till it happens and see what the payout is in the courts.

Safety comes first.

Wonder if they will issue a recall??

Responsible corporate citizen and all that...


If that's not spreading FUD, then what is it? Selective memory?

It is an errata that has been documented in the manual before release, with fixes (microcode update).

Plus, if you're not overclocking that CPU, chances are you won't even overheat them.

P.S.: Before CPU catch on fire, it will simply overheat and shut down. Make sure you get the simple fact right next time.
February 12, 2008 1:25:32 PM

I currently own an E8400 running at 3.6GHz 100% stable with no Vcore increase with an NV120, and have been running it for a number of weeks now. I have noticed the temp reading errata that you are referring to, one of the cores reads at a constant 33 degrees, where the other core fluctuates at least +/- 3-4 degrees when placed under load. I can only guess that once core is reading correctly while the other isn't... but what is the greatest difference in temp between cores that usually is seen? For the most part, one cores temp can be assumed from the other core's temp with maybe a 10% margian of error at the most... I guess this is important when you are doing some extreme OCing, but for standard operation to moderate OCing, i think it would be safe to say that these cores are worth the purchase, as long as you arent planning on pushing the chip to its limits on air.
February 12, 2008 2:48:18 PM

This topic has me interested. I am planning an E8400 build and 1st heard of the temp issues by reading turpit's cpu sticky. So I started googling and ran across lots of "discussions" in forums primarily (as zpyrd mentioned). Several things jumped out at me:

1. The faulty readings seemed to be mentioned more with Abit P35 boards than with others, IIRC - some said it was a non-issue with Gigabyte P35 boards ...

2. Others suggested that the issue was more pronounced with modular PSU's - and not a real problem with single-rail P/S's (though I have no idea why this would be).

So here is why I'm "more" interested in these issues now:

1. I just bought the Abit IP35 Pro board ...

2. My PSU is the CORSAIR 620HX

I can report this. I had ordered an E8400 from a etailer last week (was on B/O) for a nice price of $209. The salesman told me his suppliers had said that stock would be available in about 7 days. Now the suppliers are saying that E8400 stock won't be available until the end of the month, at the earliest, and probably not until sometime in early March ... (I canceled the order).

I have no idea if this extended delay is the result of Intel's "fix", or just some market dynamic. I hope Intel is working on a new stepping.

Bottom line, I'm holding off on buying the E8400 until I know more than rumor and speculation. I sure wish Intel would issue some tech updates (if necessary).
February 12, 2008 3:36:22 PM

I suppose its usually better to wait for a new stepping.

I have that exact same PSU, but no E8400, and I use a MSI MB. :lol: 

I know I can run one, but I'm planning on the Q6600.

It just seems to be a bigger upgrade, and to pay less per core to me. :whistle: 
February 12, 2008 10:01:33 PM

Somewhat OT (or belongs in Motherboards) - but I noticed a lot of posts around the net that mentioned that Abit was problematic with this E8400 chip and that Abit did not "officially" support it.

RE: http://www.uabit.com/index.php?option=com_content&task=...

I bought this board for the e8400 based on this:

Quote:
... FSB 1333 - Speed up to FSB 1333
Intel's best processor for multimedia applications with up to 80 percent faster performance for highly-threaded applications and four processing cores to handle the most massive throughput, the Intel Core 2 Extreme quad-core processor provides exceptional performance. With the IP35 Pro you are prepared for the latest 65nm and next generation 45 nm multi-core CPUs. With processors supporting Intel’s SpeedStep Technology (EIST) the IP35 Pro offers even more advanced energy efficiency....

So, noticing that temp problems were being reported with the Abit boards ... and that I owned one, I called Abit's tech support line. After a 40 minute hold, the tech support guy told me that "no", the IP35 Pro does not officially support a Core 2 Duo, Socket 775, 1333MHz FSB CPU.

Can someone tell me where I misread the Abit spec's ... dam, the guy said a bios update would be needed but that he couldn't give me any estimate as to when that new bios would come out. Now I've got a board to eat or RMA :(  [/rant]
February 12, 2008 10:17:23 PM

This seems like a really minor reason to wait for an E8400.

The only way I could see this being an issue is for extreme overclockers, but for the rest of you, the time you wait is time the long term value of the CPU declines, for no real gain.

I upgraded from an E6400 to an E8400 and I didn't even bother to overclock. My E6400 I upped from 266 to 300 fsb (2.4 ghz), with the E8400 I'm already at 3.0ghz, and with an extra 4mb L2 to boot.

There hardly seems reason to overclock unless you're going for the fastest system possible. And even still, the temp issue won't even be a factor unless you're overclocking past 3.6. Who really does that on a new CPU for every day use? I'm an overclocking gamer as much as the next guy, but why waste a chance at an amazing CPU like the E8400 just to overclock it an extra 200mhz later?
a b à CPUs
a b K Overclocking
February 12, 2008 10:23:54 PM

I had an E6850 running in my IP35 for a couple of months whilst waiting for the Q6600 to arrive so I know it can run a 1333FSB CPU, but like Jake_Barnes I was/still am under the impression that I could drop a 45nm CPU straight in. I may have to get an E8400 just to confirm or disprove, but my fav shop is out of stock at the moment.
February 12, 2008 10:26:29 PM

My E8400 dropped right into my MSI P35 Neo2 (an amazing MB btw, no idea why it's not more popular)

At work I use a lot of Asus P5KSE's, and they require a Bios flash, but work fine afterwards.
February 12, 2008 10:39:14 PM

I would buy it now. It's a fantastic CPU. But dude, how needy are ye?
I myself i really need a new PC! my current PC is 3½ years old and it was a budget PC back then! (running 3000+ sempron, 7600 GS, 1gig ram. . .)

It's a bad error if you overclock (like i intend) nevertheless it's an awsome CPU and performance wise it dosn't lack anything!

Go for it!
February 17, 2008 10:18:59 PM

Jake_Barnes said:
...

I can report this. I had ordered an E8400 from a etailer last week (was on B/O) for a nice price of $209. The salesman told me his suppliers had said that stock would be available in about 7 days. Now the suppliers are saying that E8400 stock won't be available until the end of the month, at the earliest, and probably not until sometime in early March ... (I canceled the order).

I have no idea if this extended delay is the result of Intel's "fix", or just some market dynamic. I hope Intel is working on a new stepping.

Bottom line, I'm holding off on buying the E8400 until I know more than rumor and speculation. I sure wish Intel would issue some tech updates (if necessary).


I know how you feel Jake_Barnes

I've been given the same run around with my e8400 CPU. Before I ordered parts for my new build, I looked around at online retailers in my country and they were all saying that the e8400 would be in stock around 4-8th of Feb. So i decided to order one with my build... only to be placed on backorder with its stock ETA due on the 26th of Feb.

I decided to do some hunting around the other online retailers; keeping an eye on the ETA for when the e8400 would be due in their stock and this is what i got:

one store: e8400 due on the 4th of feb,
Then changed to the 25th of feb

another store: e8400 due on the 8th of feb,
Then changed to have to no ETA date

yet another store: e8400 due on the 8th of feb,
then ETA changed to 11th of feb (reports "no stock" on the ETA day)
then ETA changed to 14th of feb (reports "no stock" on the ETA day)
I email them, they say it will be in stock on the 19th of feb
then ETA changed to 17th of feb (reports "no stock" on the ETA day)
I email them again and they tell me its not due until the end of the month, or mid march

And now with you also saying that the suppliers wont have the e8400 until around march (which is exactly what one store said), gets me worried that my new build is gonna have to wait possibly another month (i orderd my parts on the 30th of Jan, i have everything bar the CPU :(  )

I'll have to light a fire under the retailers ass I ordered my parts from and ask them to give me a "truthful and realistic" ETA for my e8400. I only ordered the e8400 because it was $30 more than the e6750 and at the time it was due between 4-8th of feb
February 17, 2008 11:19:50 PM

i hope one of those stores isnt newegg. im waiting on them just like you.
February 17, 2008 11:36:35 PM

Aw man. Just got an email back from my retailer saying that their supplier is the same as the store that has pushed back their ETA to mid march. So that means i may have to wait till mid march then.

On the other hand, i have talked to another retailer with 3 left, but is selling them at $30 more than the one i have back-ordered. I paid NZ $350 ($277 USD) for it. the other retailer is selling them for NZ $380 ($301 USD)

What do you guys think? get the more expensive one? or wait? (and keep my heartless new monster comp collecting dust under the stairs lmao)
February 17, 2008 11:43:36 PM

O whoops! Sorry for hijacking the post with my own questions Tricky Trees
February 18, 2008 1:49:50 AM

Dude I just bought my 8400 and I will tell you Im not returning this chip this chip is just plane awesome. The sensor issue does not affect the chip its only an issue for us overclockers, not for regular consumers. I doubt anyone that owns this chip would return it those who have it know what Im talking about. People are overclocking this chip to 4.0 on air even with the sensor issue. This is by far the best chip I have ever owned and I would return it if it was faulty, but this chip is far too great to return. I doubt there will even be a revision for this chip, because it functions perfectly as far as most consumers are concerned.
February 18, 2008 2:11:44 AM

I have a three 8400's two of which give proper temperature readings. The other one with a earlier pack date doesn't give me any good temp reading no matter what I'm using, but it does 4.15ghz with just under 1.3 volts! Tomorrow I'm borrowing an infrared thermometer from work to see what this chip is really doing.
February 18, 2008 3:27:42 AM

No worries Nuke.
Availability hadn't crossed my mind. I had the chance to buy an E8400 but didn't cos i wanted to check out the temp OC'ing issue. Now that i'm convinced it's not a big deal i bet i'll find i can't get one, LOL!
Oh well!
!