Ivy Bridge CPUs on Z77 Benchmarks & Review - i7-3770K & i5-3570K

I assume someone isn't going to be getting Intel products for a while, and I could argue TweakTown breached their NDA. I made a prior post, but I am assuming folks don't know the Ivy Bridge CPU models (i7-3770K & i5-3570K). I was kinda surprised with the Gaming benchmarks being hampered with the Z77 Chipset, gives one a good argument to stick with the matured Z68 or not be an early adopter of Z77 until there's a couple BIOS updates.

TweakTown
i7-3770K - http://www.tweaktown.com/articles/4619/ivy_bridge_preview_with_gigabyte_z77x_ud5h_intel_z77_and_core_i7_3770k/index.html or http://tinyurl.com/88c9274
i5-3570K - http://www.tweaktown.com/articles/4618/ivy_bridge_preview_with_gigabyte_z77x_ud5h_intel_z77_and_core_i5_3570k/index.html or http://tinyurl.com/6ro5d9m

Shanghai seems to be selling the i7-3770K -> http://item.taobao.com/item.htm?id=14810392935

Gaming:
IB_AVP.png

3DMark11 - http://i1013.photobucket.com/albums/af254/Jaquith/IB_3DMark11.png
Cinebench 11.5 - http://i1013.photobucket.com/albums/af254/Jaquith/IB_CB115.png
MediaExpresso 6.5 - http://i1013.photobucket.com/albums/af254/Jaquith/IB_MediaEspresso65.png
Adobe LightRoom3.4 - http://i1013.photobucket.com/albums/af254/Jaquith/IB_AdobeLM34.png
PCMark7 - http://i1013.photobucket.com/albums/af254/Jaquith/IB_PCMark7.png

Looking at other benchmarks that I've seen their i7-3930K must be degraded; the numbers don't jive.

Prior post - http://www.tomshardware.com/forum/329765-28-full-review-3770k-3570k
 
Both the SB/IB work on the Z77 chipset. Frankly, I am glad they tested the i5-2500K on both the Z68 and Z77 otherwise my thoughts would be something was wrong with the Ivy Bridge CPU and not IMO (hopefully) the BIOS.

The only 'typo' that I noticed is the X59 vs the correct X58 for the i7-990X (LGA 1366).
 

hotthree

Distinguished
Dec 19, 2011
203
0
18,710
After reading the article, I think I am more interested in upgrading to z77 from my p67 board more then anything. I may upgrade to i5 3570k if i can sell my cpu off.

The gains seem to be ok, but not worth the money/upgrade imo unless you can sell off parts etc.
 


I actually noticed that too. The memory bandwidth is very low on their tests, compared to THG who got 45GB/s on the 3820, they are getting like 17GB/s.

Thats even slower than my 2500K which is pushing 22GB/s.
 
Since anything Intel 4-core and above, though I know some games do benefit slightly from Hyper-Threading, none of the SB/SB-E or soon IB or even the forthcoming Q1-13 IB-E are a bottleneck to the current GTX 500 or HD 7000 series or foreseeable i.e. GTX 600 series GPUs.

So if 'Gaming' is the primary concern IMO you're tossing money (file-13/eBay P67) for what is for all [strike]intensive purposes[/strike] intents and purposes a side-grade.

Again, my assumption for the Z77's performance loss is the BIOS...hopefully.
 

Yep, see -> http://media.bestofmicro.com/I/V/318199/original/sandra%20bandwidth.png
ref - http://www.tomshardware.com/reviews/core-i7-3930k-3820-test-benchmark,3090-6.html

Something is screwy, my assumption is past OC's with too much vCore i.e. >1.50v or ditto VCCSA/VTT > 1.35v. I could embed all the images in this thread, but I'm trying to reduce the scrolling :D
 

laircouk

Distinguished
Jan 28, 2012
170
0
18,710
WTH? Why is technology going back to Socket 1155? I thought that technology is thriving beyond x79 socket (lga2011).

dam nature, you're scary...
 

Blandge

Distinguished
Aug 25, 2011
316
0
18,810


You actually WANT a new socket for each new processor generation? The LGA 2011 socket is enormous. It doesn't make any sense to take up so much real estate on the motherboard for a < 200mm^2 die with signifcantly fewer PCIE lines coming off the silicon.
 
I kinda like my i7-3930K ;) I do agree it's a biga$$ CPU and socket, and the DIMM arrangement makes it difficult to manage: HSF's, VRM cooling, etc. However, 'for' me and what I do, SQL coding/testing, the 6-cores come in handy though I'm miffed 2-cores are just sitting there disabled (8-core die).
 

josejones

Distinguished
Oct 27, 2010
901
0
18,990
I was pretty disappointed with the performance of the Ivy Bridge tests there at Tweaktown.

Those results from Tweaktown were completely different than what Anandtech posted here: The Ivy Bridge Preview: Core i7 3770K Tested

It makes me wonder how much longer will I have to wait after Ivy CPU's do finally come out at the end of April before I buy? Will 2 months be enough time for Intel and mobo makers to fix all the bugs and the bios? That Tweaktown review makes me think they'll need about 6 months or more.
 

josejones

Distinguished
Oct 27, 2010
901
0
18,990
Intel's Ivy Bridge Hotter Than Sandy Bridge When Overclocked
http://www.tomshardware.com/news/ivy-bridge-overclocking-high-temp,15512.html

Unfortunately, this article doesn't address the i7 3770 without overclocking at all. I'd like to see reviews on that chip w/o overclocking to see if heat is still an issue. Will that chip run consistently hotter too? I'd like to know if the stock cooler from Intel is fine or not for the i7 3770 too.

Or, I'd like to know if Intel plans on making some changes on Ivy to address this heat issue. I'm not buying the Ivy i7 3770 if it's just going to die on me after just 2 or 3 years due to heat exhaustion.
 

josejones

Distinguished
Oct 27, 2010
901
0
18,990
i5-3550 Ivy Bridge review:

"The thing is i got this cpu today and i installed it onto my case. Then when i boot it up its fine and everything. Then after like 20 minutes it starts to have a black screen, i taught it was my power problem so when i got on, i check the asrock temp sensor in the bios. The temp was 85 and going up by the second by fives. Later did i know it became 100oC then my computer shutted down to keep my proccesor from over heating"

i5-3570K Ivy Bridge review:

"Wasn't expecting the heat

Pros: Fast, big upgrade speed wise from older core 2 duo I had before. Newest tech for reasonable price. Was expecting it to run cooler and more efficient due to hype about new transistors. Waited through Sandy Bridge for this release.

Cons: I'm a mild overclocker and was expecting to get a small over clock on the stock cooler without seeing 90C temps using stock voltages. That didn't happen, 4.0ghz = 90C for me. That's not even a 5% overclock.

They should price this unlocked boxed chip lower and not force users to buy the HSF if it can barely keep the chip cool at stock speeds. I hate paying for something just to have to throw it away.

Took an egg off for this heat issue and pathetic stock cooler.

Other Thoughts: I'm sure spending more money on a serious cooler I can get the temps down. I'm sure for a serious overclocker my complaint isn't a valid one. But I am slightly disappointed in this release given all the hype about being more efficient, then having to deal with this heat issue which seems pretty warm to me even at stock speeds. I was expecting it to run cooler than SB, not hotter."

"... with CM Hyper212 evo, after installing drivers immediately went to 43x100mhx for 4.3ghz, no voltage change..hit 70c and started to worry about the heat issues."

http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&N=100007671%20600315409&IsNodeId=1&name=Ivy%20Bridge
 

Azok

Honorable
Apr 11, 2012
246
0
10,690


Of course they run hotter, they purposely engineered them to be hotter.
 

computernewb

Distinguished
Aug 9, 2010
1,025
0
19,360


its a conspiracy. intel has an under-the-table deal with some of the heatsink manufacturers. 2012 will have the highest heatsink sales ever
 

Azok

Honorable
Apr 11, 2012
246
0
10,690
Why should I be banned? It's true. they changed the way the constructed them. There is a certain part that they are like, thermal pasting, that used to be soldered. I read up on it.
 
^Nobody's going to ban you for what you said.

IMO - Intel Zigged when they should have zagged. Tri-Gate looked dynamite on paper and by the time Intel figured-out that it's consequences of horrible thermal properties it was too late. I only 'hope' that the IB-E doesn't carry-over Tri-Gate's problems, but if I were a betting man I'd say it will - misery loves company.

The end result is everyone with a SB and more than likely those with the SB-E will be happy with their current CPU's.

Conversely, IF Intel got it right and the IB would OC well into the mid-5GHz or higher range and significantly outperformed SB (SB-E) then 'many' folks would eBay their CPU's for the new IB (IB-E).

Therefore, Intel again IMO lost significant revenue potential. Intel isn't rebating the IB just for kicks...
 
I like FPS e.g. Crysis series, BF3 and a few others - though I really don't have the time and my wife hates it if I play. My daughter likes any game IF it's decent. Correct, I have a SB-E (i7-3930K).

/edit - I do SQL so yes more cores = less time, and the differences are indeed significant.
 
The consensus isn't the fluxless solder vs TIM, it's how the Tri-Gate locks in the heat. If it were that simple of fluxless solder then Intel would have used it on the IB --- ESPECIALLY the (K) series.

In other words Design Flaw vs TIM.

Intel is losing significant money on IB to the enthusiasts, the SIMPLE solution rather than ALL the bad press would be for Intel to utilize fluxless solder, but instead IMO either Intel and or via the rumor mill is pulling attention away from bad design.

Bottom-line, IF I deliberately pooched one of my products over something so simple then everyone would lose faith in my abilities and think me a fool. However, you slice it Intel screwed the pooch.