• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Intel Core i7 Previews/Reviews Posted

Well they talk about the lower end models that draw 60wat stock, then you've got a theoretical 70w over overhead to reach the 130w limit.

...and like I said "Let's hope some clever engineers at Gigabyte or ASUS, or whereever find a way to circumvent the "hard-wired" limit."

If they can't then gone are the days where you can take a $200 CPU and get $1000+ worth of power out of it :mad:


Ok after poking around and using some common sense which I wasn't at first, after looking at the power the extreme pulls at 3.2GHZ stock "100Watts" I am going to guess you will be able to pull between 3.2GHZ- 3.5GHZ with the lower end models at the most. I am guessamating but, I think I will be close to the mark..
 
this is a sad day for overclockers. I'm not buying a processor with any OCing restrictions. I would rather go AMD. (and im a *former* Intel fan).
 
Exciting stuff.

I have to say the s1366 i7 has outperformed my expectations. Clock for clock it is faster than Core 2, but only by about 10-20%. It also manages to get between 0% and 20% extra zoom from HT depending on application. Net improvement 10-40% for the same clocks. And for a great price (for the 920 at least).

Lets take an average, 20%, and then remember OCing is pretty much out, and then factor in the cost of a new platform and DDR3, and IMO, upgrading can WAIT until there are further gains, either in performance, or in performance/power. And currently, the i7 is not more power efficient in any material way. If Intel can manage a die shrink, then it will be a winner. Before that time, IMO, a power workstation is better built on LV Xeon Core 2 system.

And it is a shame there will be market confusion with the s1160. And it's a shame that, unlike s775, there isnt a nice (and simple) upgrade path for Joe Consumer. It's a much cleverer strategy for Intel to "lock" a customer in to a platform, like s775, and then sell them better CPUs in the future. If you lock them OUT of an upgrade path, then they are just as likely to SWITCH.

Did anyone find any reviews comparing i7 with, and WITHOUT, HT? The cinebench results look very mediocre for HT.
 
Last edited:
If I hit the lotery I am gonna get a i7 extreme and a ASUS X58 with 6x4gb sticks. I cant beleive how exspensive this stuff is tho.. When are they releasing the lower-end 100$ chip/100$ mobo stuff?
 
Ok after poking around and using some common sense which I wasn't at first, after looking at the power the extreme pulls at 3.2GHZ stock "100Watts" I am going to guess you will be able to pull between 3.2GHZ- 3.5GHZ with the lower end models at the most. I am guessamating but, I think I will be close to the mark..

LOL, no more bragging about Ghz. "I'm running my 70w proccy @ 128w!!!"

Sad.
 
like you said ... hopefully MSI, gigabyte, or Asus figure something out to overcome this.
 
They probably won't unless they want to risk Intel refusing to supply them with i7 platform chips.
 
Good find, and yes, I think we do need to be concerned, fortunatly mine will run at 1.5V although at lower speeds and looser timings, at 1.65v I could probably only get about 1400mhz @ 7-7-7-20 :cry:

Yeah mine would probably run lower to but still it is just a shame that they are holding us back on overclocking the CPU and forcing us to run our memory at slower speeds than what it is rated for :mad:
 
So is Tom's Hardware right about the chips throttling them selves back after they reach the 130Watt mark if it isn't an extreme model?? If this is true then I say F intel and I shall not buy one until they get that issue resolved or some one figures out how to get around the drawback because I have never spent $1000 on a chip and I never will.....

Yea, just reading that here:
http://www.tomshardware.com/reviews/Intel-Core-i7-Nehalem,2057-7.html

"Apparently, Intel has realized that the good overclocking potential of its processors has led ambitious users to choose less expensive models, relegating the faster (and more expensive) CPUs to the sidelines."

Let's hope some clever engineers at Gigabyte or ASUS, or whereever find a way to circumvent the "hard-wired" limit.

Thank you some one paying attention lol. Right, this IMO will make or break the deal for me because I am not buying a piece o shiat that is limited by the power it consumes. Talk about a joke... I will not spend $1000 to be able to go over 130Watts "or get a good overclock". I am also wondering if this is true,, then what would that option limit the over clock to ??

Yeah mine would probably run lower to but still it is just a shame that they are holding us back on overclocking the CPU and forcing us to run our memory at slower speeds than what it is rated for :mad:

I would not listen to any BS statements like this from Tom's until it is confirmed by other sources. Tom's has ran pretty much the same BS type statements every major Intel processor release that I can remember.
 
Even if they overcome the power rating it would seem the limited capabilities of the bus to run above it's rated speed will cause problems oc'ing the lower chips. Although, I thought I saw somewheres that someone had got the bus up to 200, which for the 920 would be 4ghz, but then further complications w/ voltages and memory could prevent something like that from working well. Still, they may yet oc decently (not like core 2 probably, but if I could get 3.5ghz or so I would be happy), we should find out soon.......
 
Even if they overcome the power rating it would seem the limited capabilities of the bus to run above it's rated speed will cause problems oc'ing the lower chips. Although, I thought I saw somewheres that someone had got the bus up to 200, which for the 920 would be 4ghz, but then further complications w/ voltages and memory could prevent something like that from working well. Still, they may yet oc decently, we should find out soon.......

that was an ES... ES tend to be unlocked. http://www.coolaler.com.tw/toppc/I7920/06-3.JPG - note the (ES) after the clock speed.

I would not listen to any BS statements like this from Tom's until it is confirmed by other sources. Tom's has ran pretty much the same BS type statements every major Intel processor release that I can remember.

that just made my day.
 
LOL, no more bragging about Ghz. "I'm running my 70w proccy @ 128w!!!"

Sad.


Unless we boycott and stay with 775. :p There will be people that will buy them any way not knowing what they are getting into though of course but, maybe we will raise a big enough stink for them to rethink this some how. I personally am appauled that they want to use us now to make more money but, that is just my oppinion I guess..
 
that was an ES... ES tend to be unlocked.



that just made my day.

Well the unlocked multi has nothing to do w/ what I was saying, the multi on the 920 is 20, so like I said if the bus can go up to 200 then that'd be 4ghz. The bus has been said to be resistant to increase, very much so so I don't know how stable that'd be.
 
i honestly think its all BS with the memory if you look at most 939 and AM2 boards when you go to adjust the memory it warns you that increasing the voltage above stock will damage the processor ram and the intigrated memory controller some boards go farther than warning you and display the jumps in green yellows reds and purples i just think this new intel 1.65v spec is exactly like AMD only instead of saying "anything above stock" they say 1.65 in all honesty this means nothing to me and i dont think running ram at 1.8v is going to hurt much their is a possibility obviously that intel with their brand new unused untested newborn memory controller their is more of a chance and less tolerance of burning out/higher voltages but in all honesty the message is no diffirent than the mesage us overclockers have been seeing for years...the thing that gets me and that i was NOT aware of is the 130w limit...perhaps they will find away around it...like if the readings come from the mobo their may be a way to trick the pro via an option or bios update that will shut off or give the proc false wattage reading....or someone someware might even figure out a way to bios mod and minipulate the readings so it detects wrong...however if the wattage reading code chip or ability in general is on the processor die i honestly do not see any way it can be minipulated through the bios or on the mobo in general and if that cannot be done..than 775 is were i will stay.
 
Well the unlocked multi has nothing to do w/ what I was saying, the multi on the 920 is 20, so like I said if the bus can go up to 200 then that'd be 4ghz. The bus has been said to be resistant to increase, very much so so I don't know how stable that'd be.


I'm just taking a stab here that, that is going to be way over the 130Watt limit so 4GHZ is unrealistic on a 920... Hell if I am wrong then 920 here I come but, I doubt it will go that far at all.
 
after reading all the articles i can find... i have come to the conclusion that unless you get the 965 chip you may be dissapointed. the 965 is the only one that can make it to 4ghz+.

i know this is not a big problem for alot of you guys but for someone like me its a big downer. my plans may have changed. i think i'll have to just save up for the 965.
 
I'm just taking a stab here that, that is going to be way over the 130Watt limit so 4GHZ is unrealistic on a 920... Hell if I am wrong then 920 here I come but, I doubt it will go that far at all.

Yeah we'll have to see, really that's what I've been waiting for these reviews to find out, sadly none of them seem to cover it, although some say they will soon. I almost don't even care what the 965 can do as I would never spend that much on a proc, I just wanna see what a 920 could do, but none of the reviewers seem to.....:cry:
 
I think there will be an awful lot of people saving for a 965 too fits.
 
early ES (RTM procs are heard to OC better)

06-3.JPG



I can't wait to get my hands on some nice third party overclocking boards, I think 4GHz ought to be quite reachable with an i7 920-neoseeker
 
Well the unlocked multi has nothing to do w/ what I was saying, the multi on the 920 is 20, so like I said if the bus can go up to 200 then that'd be 4ghz. The bus has been said to be resistant to increase, very much so so I don't know how stable that'd be.

FSB is locked by intel. If its anything like the way in which they lock multi's, ES samples might be unlocked. Of course i have absolutely no idea.

Lets hope newtekie is right and Tom's is full of it.

http://forums.bit-tech.net/showthread.php?p=1856653
 
Last edited:
I'm having mixed feelings about these new cpu's, i haven't decided if i like the new difficultities that they could pose for ocing.
 
I'm having mixed feelings about these new cpu's, i haven't decided if i like the new difficultities that they could pose for ocing.

We really haven't seen the oc'ing yet though. We just keep hearing you can't do this and you have to do that, but w/ no real attempts at oc'ing. Well, save for a few (like the one above from phanbeuy, or xtreme w/ 965) which actually seem to be indicating they may oc quite well after all......
 
Back
Top