Sign in with
Sign up | Sign in
Your question

Official Intel Ivy Bridge Discussion

Last response: in CPUs
Share
April 23, 2012 12:06:26 PM

Hello,

So today is April 23rd... the release of Ivy Bridge! I'm making this thread to hopefully centralize the actual information for everybody who wants to know. Hopefully people who got ahold of an IB chip today can post benchmarks and overall opinions. I'm separating this from the "news" thread because that thread seems to focus on pre-release information.

When you post (and this is edited in the original thread as well), please tell us what cooler you're using (H100, custom liquid, Hyper 212 Evo, Noctua, etc.) so that we can evaluate its temperatures (which has been the main concern). Also, please use the format malmental (apparently now "verbalizer") used below, but include the cooler you're using (of course).

So for anyone with the chip--post your findings, overall opinions, benchmarks, etc. here!

EDIT: Benchmarks have been released on hwbot.org

i5-3570k: http://hwbot.org/hardware/processor/core_i5_3570k/
i7-3770k: http://hwbot.org/hardware/processor/core_i7_3770k/

Other Various Reviews on the i7-3770K: put together by user Jaquith from source @Hms1193
Intel Core i7 3770K @ PCPerspective
Intel Core i7 3770K @ Tomshardware
Intel Core i7 3770K @ Anandtech
Intel Core i7 3770K @ Hexus
Intel Core i7 3770K @ VR-Zone
Intel Core i7 3770K @ Wccftech
Intel Core i7 3770K and 3570K @ Sweclockers
Intel Core i7 3770K @ Guru3D
Intel Core i7 3770K @ HardOCP
Intel Core i7 3770K @ Expreview
Intel Core i7 3770K @ TweakTown
Intel Core i7 3770K @ Maximum-tech
Intel Core i7 3770K @ Hardware Info
Intel Core i7 3770K @ HardwareCanucks
Intel Core i7 3770K @ Overclockers
Intel Core i7 3770K @ Bit-tech
Intel Core i7 3770K @ Hardware Heaven
Intel Core i7 3770K @ Xbit Labs (Best review @Hms1193 opinion)
Intel Core i7 3770K and i5 3570K Review @ Vortez
Undervolting and Overclocking Core i7 3770K @ Anandtech
Intel Core i7 3770K Review @ Overclockers Club
a b à CPUs
April 23, 2012 12:25:00 PM

We have a stack of benchmarks, they will go up on HWBot when it gets the clear.
April 23, 2012 12:27:15 PM

Sweet. Good point on setting up protocol. Just use malmental's ;) 
Related resources
April 23, 2012 12:44:32 PM

Ah, last thing. When you post (and this is edited in the original thread as well), please tell us what cooler you're using (H100, custom liquid, Hyper 212 Evo, Noctua, etc.) so that we can evaluate its temperatures (which has been the main concern).
a b à CPUs
April 23, 2012 1:30:28 PM

I should post my thuban benches, but can only do stocks as my Cooler is being RMA'd

3Dmark 11
3Dmark Vantage
SuperPi
MaxXmem
Aquamark HWBOT
Heaven
Cinebench
April 23, 2012 7:00:58 PM

Hahaha. Anyway, what do you mean by making something like the i5-2500k... do you mean an i7 version?

Also, I can't seem to figure out if IB runs hotter speed for speed than SB. Meaning... if I overclock an i5-2500k to 4.5GHz on air cooling and an i5-3570k to 4.5GHz on air cooling, will the i5-3570k be hotter?

Oh... and AMD... c'mon...
a b à CPUs
April 23, 2012 7:44:42 PM

yes it runs hotter clock for clock, perhaps only a little hotter perf for perf.
I will be getting it, from a Q9550 the 3570K will be a nice update, it'll get me off PCI-E 1.1 and so I'd may as well jump to 3.0, I also don't want to be updating platform for another 3 years +

i'm after a mild OC, 4.3-4.4 (still 25%+) thats 24/7 and on quiet air (Xigmatek Achillies + Noctua NF F12)
April 23, 2012 8:22:59 PM

Man... I will probably go with SB. It's cheaper, and I'm not that interested in PCI-E 3.0... it's not like I upgrade all that much. I can't believe they have that problem when it uses less energy...
a c 188 à CPUs
a b å Intel
April 23, 2012 8:30:07 PM

Well I have a new system that is in the works. I haven't had the chance to overclock it yet but I am building around the Intel® Core™ i5-3570K, the Intel DZ77GA-70K and using the Corsair H100. Hopefully I should be able to report on my results next week but I am looking forward to seeing what I can do with this systems.

Christian Wood
Intel Enthusiast Team
April 23, 2012 8:31:46 PM

Yeah, i5-2500k and ASRock Pro3 Gen3 (don't need extra SLI/Crossfire slot since I'm not that hardcore of a gamer). I'm on a really tight budget... might even have to forego overclocking and bite the bullet on an i5-2400... but that's for a different thread at a later time.

Overall, I think we've established that this chip is not really a disappointment because we were expecting this, but it's not really a surge forward for the tomshardware community. Tablets, laptops, smartphones, and IGP users will benefit from this, though.

Although... why would you want your smartphone to get hotter?
a b à CPUs
April 23, 2012 8:46:04 PM

Well I don't know if its still the record but the world HWBot record for memory performance was set on a Llano so, AMD is not as bad as made out, not by a long shot.

When our private forum releases the benches I will post results that I submitted, it is better but now wow, apart from memory performance which can do a SUPERPi 8m sample in like 1m 17 odd with DDR3 2133, I get 1m 23 with DDR3 1866 on my SB.
a b à CPUs
April 23, 2012 9:13:40 PM

There are a few instances where AMD gains wins, mostly out of IPC benches, but Piledriver is not going to change that, but it may close it with far better IPC's and IMC's, the good part is that AMD had time to produce PD after the entire Hector Ruiz debacle with a reshuffle of the business model I will expect significant performance yields over Zambezi.

Another thing for the AMD faithful is that we have to accept that the handicapped resources make it difficult to compete and beat Intel in the modern climate, it is about re-evaluating priorities and making them more realistic. Fact PD has to be significantly improved no questions asked, that is where perspective is needed. I would take somewhere between 10-15% which will show on all synthetics as hefty gains.

If we are expecting to see AMD chips above intels mainstream and enthusiast chips, then perhaps we are expecting to much and will likely always be disappointed.
a c 80 à CPUs
April 23, 2012 9:53:51 PM

Quote:
dude, I'm not even going to look because EYE already know....

(still got the water-cooling that leads all the way into the kitchen..)
I've seen it..


And it works really good too!

You'll look, you can't resist! :lol: 
a c 80 à CPUs
April 23, 2012 10:05:00 PM

:lol: 
April 24, 2012 2:48:49 AM

i've seen everywhere that it was released today, but why can't we buy it? is it not in the United States?
a b à CPUs
April 24, 2012 5:43:20 AM

The GTX 580 SLI scaling is rather weak, 12k for cards that score 8500P alone is rather low, seen 7870's beat that effortlessly.
a c 184 à CPUs
a b å Intel
April 24, 2012 6:54:50 AM

Well honestly, there is NO point in buying it! For a 3% more performance than sandybridge and it runs much hotter? Doesn't seem like it is worth it.
April 24, 2012 7:49:41 AM

Is it worth upgrading from an i7 1366? Im not sure the performance increase would be worth the money.
a c 184 à CPUs
a b å Intel
April 24, 2012 8:23:06 AM

Quote:
Is it worth upgrading from an i7 1366? Im not sure the performance increase would be worth the money.

Just overclock.
a c 80 à CPUs
April 24, 2012 9:18:28 AM

That's a good point sarinaide,

More than likely driver related, even though nvidia keeps it's driver data base updated pretty good, some of the older drivers just flat perform better in SLI with the older cards, and as time goes on the 580s are already older cards.

It wasn't the cards by themselves that produced that score, improving the scaling, improves the score and pulls further away from Toms tests, I probably need to run some tests and find which drivers yield the best SLI performance with the 580GTX cards.

Still no need to upgrade yet.
a b à CPUs
April 24, 2012 9:25:20 AM

The one positive about Nvidia is the website and interface, other than that Nvidia have become a disappointment with delays and lacklustre performance, if you take the 580 out which was a bit of a one trick pony, Fermi was a massive let down.
a c 80 à CPUs
April 24, 2012 9:41:01 AM

Well we can't all enjoy perfect driver releases like ATI/AMD releases can we. :) 
a b à CPUs
April 24, 2012 10:23:28 AM

Quote:
AMD/Nvidia have as much hassle as each other.
My view:
AMD drivers cause errors with windows 7 (my experience)
Nvidia drivers either gain performance or lose it.


This, AMD plagued with driver hassels, Nvidia plagued by delays. But we digress, on the benches still rather shocked at you SLI score. I was expecting more to be honest, considering the GTX 580 was undoubtedly the king of the previous generation.
a c 80 à CPUs
April 24, 2012 10:28:36 AM

@sarinaide Thanks for bringing the scaling to my attention from my research I've found the best tested scaling drivers for the 580GTX are the 262.99 release and from Nvidias driver data base they were the very last available!

And I have never run those exact drivers! Thank You! Ry
a b à CPUs
April 24, 2012 10:43:29 AM

Quote:
was not my score...


Hence why it seemed wrong a GTX 580 is a monster and that just didn't seem right for a SLI bench.

I am trying to see results from our local OC community that have tested SLI 580's, though we don't run SLI/CF or dual GPU's for our benches so it will be private stash. I would say a GTX 580 SLI should hit 14.5~15k with ease.
a b à CPUs
April 24, 2012 1:26:36 PM

This next topics has been merged by Amdfangirl
  • Ivybridge released
    April 24, 2012 4:10:35 PM

    so when will us regular plebs be able to get oour hands on an IB? i want a 3570 really badly!
    April 24, 2012 4:51:34 PM

    Crap, I have my new rig assembly halted waiting for the 3570k to be available. Upgrading from a Q6600, I was initially going for the 2500K, but here is no sense in buying a SB cpu when IB is almost out.
    April 24, 2012 6:24:21 PM

    Ivy Bridge: i7 or i5 ???

    I've never had Intel before so, how do I determine if I need an Ivy Bridge i7 3770 or if an IB i5 3570 will due just fine?

    I have no intention of overclocking because;

    1. I'm no gamer (occasional online stuff)
    2. I want my system to have a very long lifespan.

    Is there a noticeable difference between the i7 3770 and the i5 3570? Is it worth the extra $100?

    The only differences I see between the i7 and i5 are:

    - $100

    - i7 has 4 cores/8 threads while the i5 has 4/4.

    How much of a difference will this make and where?

    - i7 has HD4000 graphics while i5 has HD2500. I'd prefer the 4k graphics for obvious reasons.

    According to this review the i5 3570k was a larger improvement over the Sandy i5 2500k (8%). While the i7 3770k is only a 5% improvement over the Sandy i7 2700:

    Desktop Ivy Bridge. Intel Core i7-3770K and Core i5-3570K Processors Review
    http://www.xbitlabs.com/articles/cpu/display/core-i7-37...

    * However, the Ivy Bridge CPU's consume about 20% less power:

    Quote:
    "There were rumors that Intel increased the TDP of the senior 22nm CPUs to 95 watts although had planned to set it at 77 watts. Intel representatives told us that it was just a rumor, though. The actual TDP of the new CPUs, including the Core i7-3770K, is indeed limited to 77 watts but 95 watts is written on their packaging in order to maintain the standard scale of 35/65/95 watts that many Intel partners have got used to. Thus, the Ivy Bridge series can be expected to consume about 20% less power compared to the 95-watt CPUs with the previous microarchitecture."

    http://www.xbitlabs.com/articles/cpu/display/core-i7-37...
    a b à CPUs
    April 24, 2012 7:15:47 PM

    for those uses an I5-2300 would be fine
    April 24, 2012 8:16:38 PM



    Sweet! Sticky!

    Anyway, would the handbrake advantage be due to the 100MHz higher clock?

    EDIT: Bah... right. 100MHz isn't much, is it, malmental. Yeah... funny how the forum member that started the stickied thread doesn't know much about CPUs :lol: 
    April 24, 2012 8:22:20 PM

    Yeah, malmental. Read the edit above for my reply... I need to post another question, though.

    How much longer is SB going to be available? 'Cause I don't want to get stuck (as of now, yes, stuck) with an IB chip.
    a b à CPUs
    April 24, 2012 8:22:52 PM

    Quote:
    you have more to learn.
    depending on usage and overclocking you might have been better off going with 25xxK..

    ivy is generally better choice I would imagine. Unless you want to run at 4.5+ghz 24/7 I would think ivy is perfectly good and offers better single threaded performance and using less power sometimes. You also get quicksync 2.0 if you need it. I don't see the point of going sandy when you can get ivy unless sandy is much cheaper.
    April 24, 2012 8:29:53 PM

    ^Agreed. Check out some benchmarks and stuff... 4.5GHz is where the problems seem to really start, though.

    esrever... you get me with that profile picture EVERY SINGLE TIME! :o 
    April 24, 2012 8:59:16 PM

    So, summarizing that article, it seems that Intel has released a competitor for APUs (sort of) with better IGP.

    So it's worthless to us. It's just Intel covering their bases.
    April 24, 2012 9:07:24 PM

    ...Did you just copy and paste that?
    April 24, 2012 9:18:15 PM

    but +0.1GHz is +2.2%
    We praise other players for such gains, why not intel
    edit: I should read to bottom of thead before responding. I think we're past this lol ...
    April 24, 2012 9:32:04 PM

    And TDP
    April 24, 2012 10:31:24 PM

    Quote:
    you have more to learn.
    depending on usage and overclocking you might have been better off going with 25xxK..


    You've said it yourself, usage and overclocking.
    I realize that SB is still superior on someways but this CPU will go into a gaming rig with mild overclocking (since I'm sticking to air cooling). I will not be aiming for Über +4.7 GHz speeds so I guess Im not too far off and that IB fits my bill, but hey, I haven't put down the money yet and I'm happy to hear any suggestions and/or suggestion.
    a c 122 à CPUs
    April 25, 2012 3:45:53 AM

    Quote:
    Conclusions

    Intel's transformation from a graphics laggard to a viable competitor has been a long journey. The strategy has been one of consistent and measured improvement, rather than a dramatic revolution. With Sandy Bridge, Intel doubled the graphics performance of the previous generation Ironlake, and brought acceptable integrated graphics to the PC ecosystem in early 2010.
    Ivy Bridge tackles GPU programmability and will nearly double performance yet again.

    Competitively, the Sandy Bridge GPU was eclipsed 6 months after release by AMD's Llano GPU, which was both higher performance and programmable.
    The saving grace was Intel's industry leading media encoding and decoding. Based on reviews and benchmarks, the gap in graphics performance varies from 1.3× to 2×.
    AMD's next integrated offering in Trinity is reportedly 1.3× to 1.5× faster than the previous generation and should have competitive media processing.

    Putting this all together, Intel will substantially narrow the gap with AMD for integrated graphics capabilities in 2012. Actual product level performance depends on pricing, binning and the market. For instance, Intel has an edge for very low power designs due to process technology. The 22nm FinFETs are exceptionally efficient at low voltage and it is likely that Ivy Bridge will match Trinity for 17W designs. At 25-35W for conventional notebooks, Intel should trail by around 20%, which is close enough to be competitive.
    Looking to desktops though, AMD will have a substantial advantage and the performance gap may be much higher.

    Based on these estimates, the Ivy Bridge GPU will be the first truly competitive integrated graphics solution and a significant milestone for Intel.
    AMD has the benefit of over a decade of experience with high performance graphics and can leverage tremendous investments in discrete GPUs.
    While Intel might lack the same experience, it appears that a full process node advantage (22nm vs. 32nm) makes up for this deficiency.
    Ivy Bridge should exceed Llano in most workloads, and compared to Trinity, narrow the performance gap to reasonable levels for most markets.

    Looking forward, there are some features that simply did not make it into Ivy Bridge and are obvious candidates for Haswell. In particular, the programmability is still somewhat nascent. Features such as DX11.1 and OpenCL 1.2 were omitted either due to project timing or risk and complexity reasons, and are expected in the next generation.
    System level integration was mostly unchanged in Ivy Bridge, but will probably be updated for Haswell. That would present an opportunity for more elegant sharing of the virtual address space and coherent communication between the CPU and GPU and greater efficiency.
    The performance should increase, particularly if the rumors are correct and there are 3 different variants of Haswell.
    That would give Intel the freedom to aggressively push performance for an expensive, high-end version, without compromising cost for the 'free' or mainstream flavors.


    Should have. nVidias solution should also be competitive but still falls very short of QS.

    Of course I want to see image quality results but still its impressive that such little die space can take on monster die GPUs......
    April 25, 2012 4:20:45 AM

    Intel's graphics driver is also being re-architected, although independently from the release of Ivy Bridge. Historically, Intel's GPUs have been relatively low performance, so the CPU overhead was negligible on modern processors and the focus was on functionality rather than optimization. In comparison, AMD and Nvidia have products that are an order of magnitude faster and the driver stack must have similarly less CPU overhead. To prepare for future integrated GPUs which will be even faster, Intel is planning to reduce the driver overhead to comparable levels when measured in CPU cycles per draw call. To accomplish this, a new graphics driver architecture is expected later this year. The current driver has a layer that abstracts away the underlying GPU hardware from the OS. The newer architecture eliminates this abstraction layer, so that the driver is tailored to a specific OS and hardware combination, but lower overhead.

    From the same link
    April 25, 2012 5:35:47 AM

    ddan49 said:
    Yeah, i5-2500k and ASRock Pro3 Gen3 (don't need extra SLI/Crossfire slot since I'm not that hardcore of a gamer). I'm on a really tight budget... might even have to forego overclocking and bite the bullet on an i5-2400... but that's for a different thread at a later time.

    Overall, I think we've established that this chip is not really a disappointment because we were expecting this, but it's not really a surge forward for the tomshardware community. Tablets, laptops, smartphones, and IGP users will benefit from this, though.

    Although... why would you want your smartphone to get hotter?



    Battery life.
    a b à CPUs
    April 25, 2012 9:22:56 AM

    Quote:
    mild overclock will still run hotter than SB, but I don't see an issue either way.
    it seems Ivy is a little fickle still..

    heat isn't a issue when you are comparing different chips considering the transistors are different. For all we know, ivy could handle more heat than sandy.

    from what I seen a mild OC on ivy is no problem so I don't see the need to avoid it unless you want to push more than 4.5ghz, 24/7.

    Personally I would choose Ivy given the chance between the 2.
    a c 80 à CPUs
    April 25, 2012 10:25:44 AM

    Good Information regarding Overclocking IB

    Disappointing information to me, :ouch:  but what is to be expected when new technology is added to the SB basic package, you're going to have to pay for it, somewhere. :non: 

    The testing cooling he uses leaves much to be desired but not all overclockers strive for the same thing either. :heink: 

    I have to admit the SB 2500K has completely spoiled me! :bounce: 

    Being tested and solid enough to punch in the multiplier level and voltage to run that multiplier in the BIOS, save and exit, and Wallah! you're there! from any range between 3.3ghz to 5.1ghz, is enough to spoil anyone. :p 

    What surprises me more, is that Intel did not release this shrunken head version of a Sandy Bridge Frankenstein Monster in a new socket. :lol: 

    I apologize if that offends the IB flag wavers, but end result wise, it is, what it is! :pfff: 

    This is my opinion only, and FYI no contrary arguments will be changing my opinion! :whistle: 

    However not every overclocking possibility has been explored yet either. :o 

    When the Sandy Bridge Ks were first released, all we heard were limitations, limitations, limitations. :( 

    When there actually were ways around those limitations, that had not been discovered yet, so there could be a light at the end of the tunnel for IB too, we'll see! Ryan
    a b à CPUs
    April 25, 2012 11:19:32 AM

    http://www.dailytech.com/article.aspx?newsid=24523

    Poor overclocker but low TDP ... poor graphics.

    http://www.anandtech.com/show/5763/undervolting-and-ove...

    "The bad news is that it perhaps will not overclock as well as people think it should. Those wishing for 4.8GHz at 1.4 volts (similar to Sandy Bridge) will run into a lot of issues if they think that 1.4 volts is appropriate for Ivy Bridge. In comparison, you may end up with something more reasonable like 4.6GHz at 1.1 volts, or 4.8GHz at 1.2 volts (as per some boards I have tested). Then it will be a case of deciding whether the small IPC gains that Ivy brings will be worth 200 MHz less on your CPU compared to Sandy Bridge".

    Ian Cutress Anandtech.

    Keep your SB / SBE ... hell keep your Nehalem.

    It should make a mean low power laptop tho ... maybe the next few days will bring some surprises wrt overclocking ... who knows.

    1.3 volts looks scary ...
      • 1 / 15
      • 2
      • 3
      • 4
      • 5
      • More pages
      • Next
      • Newest
    !