Wednesday, April 18th 2012

Core i7-3770K Retail Boxes Pictured, TDP 95W, Overclocks Worse Than Sandy Bridge?

Here are the first pictures of retail boxes of Intel's Core i7-3770K "Ivy Bridge" processors in the LGA1155 package. Pictured below are boxes sourced from a Chinese distributor. Regional branding aside, the box-art hasn't changed from that of the 2nd Generation Core processor family, even the die-shot CGI in the center hasn't changed, which is a missed opportunity. Intel could have used art inspired by the Ivy Bridge silicon, which could have helped identify the new chips easier. The box simply marks the model number "3770K" and socket type "LGA1155" on the key sticker.

The side sticker is where the action is. We know from countless earlier reports, including Intel's RetailEdge marketing material that the TDP rating of "Ivy Bridge" quad-core parts, including the i7-3770K, was rated to be 77W. The sticker on retail i7-3770K, however, tells a different story. The TDP is rated at 95W, on par with previous-generation parts such as i7-2700K. The S-spec number is revealed to be "SR0PL". Before such an important CPU launch as "Ivy Bridge", it's hard to control pre-launch proliferation of retail parts to people who are not NDA signatories. Such people have put the i7-3770K through overclocking, and voices are getting louder that the i7-3770K is a worse overclocker than previous-generation "Sandy Bridge". The chip was found to get too hot, too soon, when overclocking.

Sources: Semi Accurate (forums), NordicHardware
Add your own comment

80 Comments on Core i7-3770K Retail Boxes Pictured, TDP 95W, Overclocks Worse Than Sandy Bridge?

#1
btarunr
Editor & Senior Moderator
Many Thanks to NHKS for the tip.
Posted on Reply
#2
sc
What does this mean? :twitch: :eek:
The IB we all been waiting for is a failure?!?!
Posted on Reply
#3
btarunr
Editor & Senior Moderator
by: sc
What does this mean? :twitch: :eek:
It means I owe a kiss to Jared's ass :shadedshu
Posted on Reply
#5
Aquinus
Resident Wat-man
It's only a die shrink. I mean, how much better can you really expect it to be? It access memory faster and it gets more something like 13% more done per clock, so not all is bad. I'm still glad I got my SB-E though.
Posted on Reply
#6
repman244
by: Aquinus
It's only a die shrink. I mean, how much better can you really expect it to be? It access memory faster and it gets more something like 13% more done per clock, so not all is bad. I'm still glad I got my SB-E though.
It's not just a die shrink this time, it includes some new technology which might be causing the high temperatures (3D gating).
Posted on Reply
#8
badtaylorx
ive had a 2700k in my cart for a month now figuring this would happen.....hey AMD,,, heres your chance
Posted on Reply
#9
BigMack70
All the info pointing to hot temps on IVB led me to just get a 2600k and put it at 5.1 GHz... no regrets from me - only took like 20 minutes to get 5.1 GHz stable and fairly cool on my Noctua.

IVB is looking like a big win for the IGP side of things (which is irrelevant to most desktop users), but the CPU side of things doesn't look so good for overclockers... of course once official reviews go live that could be different, but we'll see.

Part of me wonders if some of the problem is no pressure from AMD to make Intel do better... at its current pace AMD won't have Sandy Bridge performance until like 2014. :shadedshu
Posted on Reply
#10
renz496
i've been hearing about this a few days ago. personally i'm gunning for 3570K build but if the rumor about ivy being worse overclocker than sandy might swing me to 2500K instead :roll:

but i still believe ivy will be a great processor despite that. stock vs stock (performance) ivy should be the same or a bit better than sandy. also from nordic hardware:

http://www.nordichardware.com/news/69-cpu-chipset/45738-ivy-bridge-sells-with-95w-tdp-but-uses-a-maximum-of-77w.html
Posted on Reply
#11
FreedomEclipse
~Technological Technocrat~
by: Dj-ElectriC
I cannot confirm the same results ATM but all i can say is that my 2500K will stay in it's socket safe and sound.
+1

other then a new GPU, i dont see the need for me to go IB
Posted on Reply
#12
Sasqui
We'll see some revisions and refinement to IB chips (like watching Conroe vs. Wolfdale), and these problems will dissapear with time.
Posted on Reply
#13
Aquinus
Resident Wat-man
by: repman244
It's not just a die shrink this time, it includes some new technology which might be causing the high temperatures (3D gating).
3D gating is used to reduce temperatures and make it use less power. I doubt this is why it runs hot. It's because it's a die shrink and the resistance of a smaller wire is higher. I had an entire rant about this along with how voltage scales with current and such. To make a long story short, it just runs hot at voltages required for similar clock speeds as its 32nm counterpart.

by: Sasqui
We'll see some revisions and refinement to IB chips (like watching Conroe vs. Wolfdale), and these problems will dissapear with time.
You mean it will disappear when Haswell comes around, and even then I don't think that will be the case. You can't break the laws of physics unfortunately.
Posted on Reply
#14
brandonwh64
Addicted to Bacon and StarCrunches!!!
LOL damn! These things are heat monsters

Posted on Reply
#15
repman244
by: Aquinus
3D gating is used to reduce temperatures and make it use less power. I doubt this is why it runs hot. It's because it's a die shrink and the resistance of a smaller wire is higher. I had an entire rant about this along with how voltage scales with current and such. To make a long story short, it just runs hot at voltages required for similar clock speeds as its 32nm counterpart.
Isn't 3D gating supposed to only lower the required voltage? By mentioning 3D gating I didn't mean that it runs hotter because of the gating itself but because of the new technology incorporated in the process which still needs some tweaking (perhaps I didn't word it very well).
Also you need to be aware that the transistor count is supposedly increased (Intel claims the density COULD be increased by 2x) and along with smaller area you get more concentrated heat.
Posted on Reply
#16
function69
Disturbing news indeed, totally didn't expect this.
Posted on Reply
#18
flashcrew
Ivy Bridge sells with 95W TDP, but uses a maximum of 77W. This explains all regarding that 77W, 95W TDP confusion.

http://www.nordichardware.com/news/69-cpu-chipset/45738-ivy-bridge-sells-with-95w-tdp-but-uses-a-maximum-of-77w.html

Well most SB chips won't use more than 70 watts at stock, at full load, either.

People just don't know what TDP is, nor do they seem to realize stock power consumption and heat production can vary considerably from chip to chip.
Posted on Reply
#20
MeanBruce
by: flashcrew
Ivy Bridge sells with 95W TDP, but uses a maximum of 77W. This explains all regarding that 77W, 95W TDP confusion.

http://www.nordichardware.com/news/69-cpu-chipset/45738-ivy-bridge-sells-with-95w-tdp-but-uses-a-maximum-of-77w.html

Well most SB chips won't use more than 70 watts at stock, at full load, either.

People just don't know what TDP is, nor do they seem to realize stock power consumption and heat production can vary considerably from chip to chip.
I hope you are from Intel and know what you are talking about!

This article is just freaking people out, it should not have been posted until confirmed!:eek:
Posted on Reply
#21
ap4lifetn
by: flashcrew
Ivy Bridge sells with 95W TDP, but uses a maximum of 77W. This explains all regarding that 77W, 95W TDP confusion.

http://www.nordichardware.com/news/69-cpu-chipset/45738-ivy-bridge-sells-with-95w-tdp-but-uses-a-maximum-of-77w.html

Well most SB chips won't use more than 70 watts at stock, at full load, either.

People just don't know what TDP is, nor do they seem to realize stock power consumption and heat production can vary considerably from chip to chip.
intel gets a +1 in my book if this is true, this helps ensure the 1155 platform remain stable, and gives us higher quality z77 motherboards

i am sure hoping for that 6core ivy bridge
Posted on Reply
#22
brandonwh64
Addicted to Bacon and StarCrunches!!!
by: sc
Wait... WHAT??!?!
Are those temperatures? Because if that FX-8150 does "8" (calsius, fahrenheit, kelvin, whatever) I know what I'm going to buy tomorrow :cool:
They were using a H100 with push pull config....... I don't believe the bulldozer but if the IB is 98 deg full load on that kind of water then FUUUUUUUUUUU that.
Posted on Reply
#23
Aquinus
Resident Wat-man
Sounds like AMD has a chance to catch up with Piledriver. Will it happen? We shall see!
Posted on Reply
#24
Isenstaedt
by: Aquinus
It's only a die shrink.
I'd expect a greater increase in performance from a die shrink than from a new architecture.
Posted on Reply
#25
Initialised
It's not the die or the transistors.

A water block designer told me the IHS is different and this may be the reason for the higher temperatures.
Posted on Reply
Add your own comment