Editor's Corner: Overclocking Core i7

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Status
Not open for further replies.
I like it when hardware review sites test hardware. Sounds pretty simple. Good work. It is even better when hardware review sites test hardware before making conclusions about new possibly revolutionary hardware. Now that you have found the 920 to easily hit a 43% overclock on an engineering sample, you might want to go back and change the title and half the conclusions of your first core i7 review. Why don't you try changing your title to something meaning the complete opposite of the old title, like: "Intel's Core i7: Blazing Fast, But even more amazing O/C". Sounds a lot more accurate than "Intel's Core i7: Blazing Fast, But Crippled O/C" Never knew 43% O/C could be considered crippled
LOL
 
[citation][nom]heoaheoa[/nom]I like it when hardware review sites test hardware. Sounds pretty simple. Good work. It is even better when hardware review sites test hardware before making conclusions about new possibly revolutionary hardware. Now that you have found the 920 to easily hit a 43% overclock on an engineering sample, you might want to go back and change the title and half the conclusions of your first core i7 review. Why don't you try changing your title to something meaning the complete opposite of the old title, like: "Intel's Core i7: Blazing Fast, But even more amazing O/C". Sounds a lot more accurate than "Intel's Core i7: Blazing Fast, But Crippled O/C" Never knew 43% O/C could be considered crippled LOL[/citation]

And I hear that next week, some of us will evidently walk on water and then turn it into wine for the Veteran's day celebration. lol

that's one lake i'd like to swim in. MD 20/20 anyone? lol

seriously...didn't a manufacturer once make chips that were underclocked...and then enthusiasts got like 60% O/C out of them with oversized air cooling? i think that "crippled" is all relative. I consider my Athlon64x2-5600+ OC crippled cause i only got it up to 2.96GHz instead of over 3.0 GHz. Of course, I don't know what the heck i'm doing either. so, i didn't worry too much. And after all...hasn't Tom's oc'ed 60% increases before? that shadows 43%.

the guys at tom's probably don't have all the time in the world like we do to sit around and comment on everyone else's work. they probably keep busy with multiple projects and multiple deadlines like most real working stiffs in the tech world. plus, they probably rely on some information from sources. not all sources can be 100% accurate. if so, news would be a LOT different in this world.

i liked the article, and don't mind so much a bit of misinformation...the fact it was corrected is what counts to me...and it's even better that it was done with efforts by both users and editors.

kudos to everyone.
 
jcknouse, Well that is true at least they corrected their previous article. I just think it is pretty poor of them to make such a big deal out of some overclock limit, when they were so wrong. They even used bold face type 3 times saying how the o/c as we know it is finished for non extreme i7 chips. Talking about the negative effects on the after market cooling solutions....... Going from 2.66 GHz to 3.8 with an engineering sample quad core is a superb o/c.
 
How can you hold someone accountable for reporting what they know to be true at the time? In the previous i7 article, TH was only reporting what they knew to be true. Now, they know a couple of the assumptions and implications were wrong, so they printed a correction. Thanks TH! You guys rock. You could throw a little more humor around, though, like the folks over at theregister.co.uk :) There are too many folks taking this crap too seriously...
 
hurbt: You can hold someone accountable if they report something as fact, and it is their job to test it, and they don't bother to test it and pretend they know what they are talking about. Tom's isn't a hardware reporting site, they are a hardware review site. They are supposed to test hardware and report the results of repeatable, verifiable, systematic tests so that we can pick the best hardware to buy. They made a huge deal out of some "crippled" o/c and one day later they decide to actually test the chip. This wasn't just some run of the mill article, it was an architectual change being evaluated by Tom's, of which only occurs on average once every year (assuming amd and intel change once every 2 years). This type of error brings into question the integrity of the webpage, but at least they tried to fix it a day late. gg
 
[citation][nom]kami3k[/nom]you would be lucky if the CPU didn't fry instantly no matter what if that were to happen. [/citation]
Quite the contrary. You can run a Core 2 CPU without a heatsink just fine, it just won't be performing it's best once PROCHOT kicks in (almost straight away). With TM disabled you can go well over 100C but then you're hitting the real danger zone.
 
Heo,

Bert's assertion remains correct. If you don't have a platform with an override for the Overspeed Protection (we know ASUS and Intel do, but will Dell, HP, etc enable this?) then the overclocking headroom of an i7 is in fact crippled. There's a hard-wired limit in place there that requires a specific BIOS option to work around it.

After reading Bert's piece, however, and factoring in reader feedback, I went back and did my own exploration into what was going on, published above.

BTW, check out Fugger's screenshots--CPUZ reports 5.2 GHz, while Everest reports 3.8 GHz. Clearly all of the tools to make overclocking on i7 a smooth experience are not in place yet.

The fact of the matter is that we're following up quickly and in as much detail as possible to convey the whole story.
 
Chris you missed your calling, should have been in PR :) Nice CYA though, "overclocker in the know", thats an instant classic. A sign of good journalism is knowing when to make a proper retraction...
 
[citation][nom]Porksmuggler[/nom]Chris you missed your calling, should have been in PR Nice CYA though, "overclocker in the know", thats an instant classic. A sign of good journalism is knowing when to make a proper retraction...[/citation]

Are you kidding? I'd be a horrible PR person.
 
1.5 Volts ... gee ... thats a bit high.

I'd like to know how much of a bump you got out of it with just a slight shot of voltage ... before major juice needed to be applied.

I'd guess quite a bit.

Good intro article Bert.
 
[citation][nom]reynod[/nom]1.5 Volts ... gee ... thats a bit high.I'd like to know how much of a bump you got out of it with just a slight shot of voltage ... before major juice needed to be applied.I'd guess quite a bit.Good intro article Bert.[/citation]

Rey, I was able to boot 3.4 without adding anything. Wasn't stable though. You can get a good overclock on minimal voltage.
 
So a 920 chip at 3,8ghz would be 66% faster than my e6600 @ 3,4ghz in apps that support only single or dualcore mashines? And only if you can afford to buy ddr3, an enthusiast motherboard supporting disabling the voltage limiter and the cpu itself?

While it may be good news to those who else would've bought an extreme part, I don't think it's any real option for people like me yet. It still appears too expensive. Only if you use it for productivity, you may make use of all cores, and if that's the case you're probably using a brandname pc with stock speed, gaining less than 50% performance boost at full price.

Guess we'll have to wait till a new lower-midrange chip comes along (the equvivalent of an e6400 2 years ago or an e2180 earlier this year)

On another note - is the 18x multiplier on the quickpath link static, or can it be lowered like the ht on an amd mashine?
 
cangelini,


"If you don't have a platform with an override for the Overspeed Protection (we know ASUS and Intel do, but will Dell, HP, etc enable this?) then the overclocking headroom of an i7 is in fact crippled."-cang

How much is it crippled by? Did anyone test this? Crippled implies it is really bad. Maybe you guys tested it and just forgot to put the results in the article. Like whoops where did that missing page go, oh here it is....... Maybe you didn't test it at all and just assumed the o/c would be lower. Then you assumed it would be real bad and decided to call it crippled. That seems the most likely scenario, but who knows!
Wait!! You said it is "in fact crippled" so i'm sure you tested it!!!

HAHA
 
Sure did! If you scroll up and read the story to which you're replying you'll get a fairly good idea, based on testing, where the ceiling is =)
 
Bert's assertion remains correct. If you don't have a platform with an override for the Overspeed Protection (we know ASUS and Intel do, but will Dell, HP, etc enable this?) then the overclocking headroom of an i7 is in fact crippled. There's a hard-wired limit in place there that requires a specific BIOS option to work around it.
Click to expand...
Sorry, you need to go back to the original article and cross out the misinformation and replace it with the correct information that you now know to be the truth. No one is perfect, but you must correct the original article. People will read it and may not read the correction. By the way, that whole Dell, HP etc. comment is really beneath you. I can 99.99% guaranty you that they will not have the switch in the BIOS, or any other settings of any value. The last Dell, that I had to dirty my hands by even touching, didn't even have any means to increase the CPU fan speed, and it was running a presc-hot. I was forced to use SpeedFan. I would also be willing to bet that all other non-EOM mobos will have the switch.

Just go back and clean up the other article and move on.
 
thank you for clearing this up somewhat I was hoping there was a way around the limiter as I did plan on getting the i7 sometime after xmas season. But from what I seen in the last review about not being able to overclock I was very worried. I have been a over clocker since the 80286 days back when it was a challenge to overclock anything & you had to make your own heatsinks & buy the right clock crystal chips to make your smoking machine go even faster. So when I read the last review I was pretty much not gonna buy a i7 if you could not over clock it as every chip I have ever owned has been drivin to much greater speeds so it is a relief to know that there is a work around for this limiter issue I just hope the work around does not fry the CPU if it gets to hot
but hey those are the chances you take when you want to over clock I guess.
 
Wait 2 or 3 months after the lunch of the Core i7 when they add new Stepping to it, then you should get 4 GHz easy with air. But for now I'm sticking with my ASUS P5Q PRO and my E8400 then upgrading to a Q9550.
 
I hear from many people through forums that the CPU TM FUNCTION does not override Intel's Overspeed Protection. Is your testing really legit? Most people are having their processors throttle back as if the overspeed protection is engaging even with the CPU TM FUNCTION turned off.
 
The results presented above are indeed legit. More info on "most people" (link?) would help us and the other folks reading your post dig further into any potential issue, though.
 
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom