AnandTech reviews the 4960X

[Skrips]

Here it is, fresh off the press:

On the surface, Sandy Bridge E was a very good gesture on Intel's part. Unfortunately, the fact that it's been nearly two years since we first met LGA-2011 without a single architecture update, despite seeing the arrival of both Ivy Bridge and Haswell, doesn't send a great message to the users willing to part with hard earned money to buy into the platform.

Today we see that long awaited update. LGA-2011 remains unchanged, but the processor you plug into the socket moves to 22nm. This is Ivy Bridge Extreme.

Read the article HERE.

[hokiealumnus]

I'm surprised there wasn't a thread yet; NDA lifted at (the odd time of) 12:01am PST. Here are some more:

Hardware Canucks
Overclockersclub
TechReport
Overclockers
PCPer
Guru3D
VR Zone

There we go. Tired of copying/pasting now.

(I refuse to link to the two that published before NDA.)

[lutjens]

These reviews confirm what we already knew...Ivy-E provides only the barest of an incremental performance increase, overclocks worse than Sandy-E, has a severely obsolete chipset and should've have been an 8 core.

Nothing new at all...

[Skrips]

These reviews confirm what we already knew...Ivy-E provides only the barest of an incremental performance increase, overclocks worse than Sandy-E, has a severely obsolete chipset and should've have been an 8 core.

Nothing new at all...

Be that as it may, it still does not change the fact that I have to buy one. I cannot limp along with 4 cores anymore. My development workload has surpassed regular Sandy Bridge parts.

Do I want to lay down $4k+ for a new system? Not really, but there doesn't seem to be any other option, and I sure as hell cannot wait for Haswel-E late next year.


--Skrips

[jogshy]

Another 3-5% performance improvement that does not worth upgrade.

I'm waiting for real new things like DDR4/Hypercubemem/reRAM, PCI4, 512bits AVX, Sata Express, desktop 16-core ARM A57 16nm fin-FET APUs, etc..
Thanks Intel, but nope

[AliG]

Another 3-5% performance improvement that does not worth upgrade.

I'm waiting for real new things like DDR4/Hypercubemem/reRAM, PCI4, 512bits AVX, Sata Express, desktop 16-core ARM A57 16nm fin-FET APUs, etc..
Thanks Intel, but nope

I agree, Intel is completely milking their high end performance lead. However one thing that's gone really under the radar are the new monitors.

I just got one of the Korean 1440p matte PLS monitors this thing blows every other screen I've used out of the water, even the workstation HPs at my office. Right now you still have to overclock it to get 120 hz (seriously that's a thing now) but I'm guessing it's only a matter of time before we get cheap 120 Hz panels as well. So if you're going to spend money on an upgrade this generation, why not just hit the peripherals? Corsair also recently launched a baller mechanical keyboard (the K95) so I'm thinking about getting one of those.

[lutjens]

Be that as it may, it still does not change the fact that I have to buy one. I cannot limp along with 4 cores anymore. My development workload has surpassed regular Sandy Bridge parts.

Do I want to lay down $4k+ for a new system? Not really, but there doesn't seem to be any other option, and I sure as hell cannot wait for Haswel-E late next year.


--Skrips

Pick up an i7-3930K or Xeon E5-1650...these'll do the job and save a couple of bucks at the same time. You'll end up with performance that's about the same or a bit more, as the SB-E chips should overclock more. I ended up buying an E5-1660 a couple of months ago, instead of waiting for Ivy-E. The reviews show me I made the right call. It'll also not help Intel to sell one of these new turds...a small protest but a symbolic one.

[mAJORD]

Ha, tech report review has been sabotaged. Better have a read for lols before someone fixes it!

Sent from my HTC_PN071 using Tapatalk 2

[LordEC911]

Ha, tech report review has been sabotaged. Better have a read for lols before someone fixes it!

Sent from my HTC_PN071 using Tapatalk 2

It is them trying to be a unique snowflake. Was like that last night when they first posted it.

[mAJORD]

It is them trying to be a unique snowflake. Was like that last night when they first posted it.

Yeah, so it seems..

Have they gone nuts?

[DefStar]

Be that as it may, it still does not change the fact that I have to buy one. I cannot limp along with 4 cores anymore. My development workload has surpassed regular Sandy Bridge parts.

Do I want to lay down $4k+ for a new system? Not really, but there doesn't seem to be any other option, and I sure as hell cannot wait for Haswel-E late next year.


--Skrips

Yeah Dude, you have every reason to get one...!

[SKYMTL]

Yeah, so it seems..

Have they gone nuts?

No. It's Scott's way of telling Intel: you're wasting everyone's damn time.

[lutjens]

(I refuse to link to the two that published before NDA.)

Yeah, how dare they violate Intel's almighty directive. Don't they know that the word of Intel is to obeyed as you would that of your higher power? We should never question Intel, speak poorly of Intel or do anything that Intel might disagree with and joyously accept whatever half-baked product that Intel feels like foisting upon us, thanking them profusely for their benign benevolence and the opportunity and privilege to be their customer...

/sarcasm


There...I've done it for you...

http://www.tomshardware.com/reviews/...mark,3557.html

http://www.hardocp.com/article/2013/...c_ocing_review

[Chrono Detector]

I would have been more interested if they released a new chipset because X79 is old but Intel still insists on using X79 while the mainstream platform gets a new chipset and socket which doesn't make any sense. Other than that I'm not even interested in IB-E because it offers absolutely nothing if you already have a SB-E CPU unless you want a lower power draw but still not worth spending that extra money.

[tajoh111]

In general, 22nm has been a failure, particularly looking from an overclocking point of view. If anytime was good for AMD to start making a big move against Intel this is it. The only thing 22nm managed to bring was some power improvement but overclocking went down the toilette in general.

AMD needs a more aggressive release schedule. If they didn't slow down so much and bulldozer wasn't a failure, they might have caught up to intel in the desktop arena.

[mdzcpa]

No. It's Scott's way of telling Intel: you're wasting everyone's damn time.

Exactly.

And I concur. Outdated chipset, outdated architecture, and crappy overclocking to boot.

I'm disappointed that the IB-E overclocks so poorly. Some of the reviews tried to put a good face on it, but it's clear IB-E on average falls well below SB-E overclock averages. There is no way I'm laying down my SB-E that benches stable at 5Ghz and runs 24/7 at 4.8Ghz at 1.4v for one of these IB-E chips at 4.6...if I'm lucky. What I'd gain in IPC will be lost in mhz. I was hoping IB-E would at least match SB-E but that's not the case. And I don't feel liking binning 200 CPUs to find the golden chip.

Oh well. I've been running my RIVE/3960X since November 2011. I guess I can hold on until Haswell-E.

[[XC] hipno650]

It's a real let down that so far IVB-E seems to overclock so poorly... I was really hoping to ditch my dog of a 3930K (any higher than 4.5GHz is just futile) and throw in a new IVB-E CPU that would hopefully clock to 4.8+ or so... but now it seems that I would be looking at the same level of clock speed in the end if I upgraded... not really worth my money.

The chipset problem is a whole other story. I am just frustrated by the fact that Intels "Enthusiast" platform plays second horn to the mainstream in so many aspects. Only 2 native SATA 6G ports is utterly brutal. I would love to go out and by 6 Samsung 840 pros and put them in RAID... but without buying a third party RAID controller I am just wasting my time, USB 3.0 is the same story. Hell i would have bought a new board and probably a new CPU just for the hell of it if they had a proper chipset to play with.

I really wish Intel would think like they did when X58 was released and give the high end platform all the bells and whistles BEFORE the rest of the market. If you ask me SB-E should have come BEFORE SB, same with IB-E before IB and so on.

Really Intel I am not a hard sell. I WANT to spend my money on a bunch a new parts that I most likely don't need..... heck maybe if I can find a really good 4770K I will buy that just for a change of pace.

[STaRGaZeR]

Ha, tech report review has been sabotaged. Better have a read for lols before someone fixes it!

Sent from my HTC_PN071 using Tapatalk 2

Hilarious!

What's with the gif in the first page?

[lutjens]

It's a real let down that so far IVB-E seems to overclock so poorly... I was really hoping to ditch my dog of a 3930K (any higher than 4.5GHz is just futile) and throw in a new IVB-E CPU that would hopefully clock to 4.8+ or so... but now it seems that I would be looking at the same level of clock speed in the end if I upgraded... not really worth my money.

The chipset problem is a whole other story. I am just frustrated by the fact that Intels "Enthusiast" platform plays second horn to the mainstream in so many aspects. Only 2 native SATA 6G ports is utterly brutal. I would love to go out and by 6 Samsung 840 pros and put them in RAID... but without buying a third party RAID controller I am just wasting my time, USB 3.0 is the same story. Hell i would have bought a new board and probably a new CPU just for the hell of it if they had a proper chipset to play with.

I really wish Intel would think like they did when X58 was released and give the high end platform all the bells and whistles BEFORE the rest of the market. If you ask me SB-E should have come BEFORE SB, same with IB-E before IB and so on.

Really Intel I am not a hard sell. I WANT to spend my money on a bunch a new parts that I most likely don't need..... heck maybe if I can find a really good 4770K I will buy that just for a change of pace.

Intel seemingly doesn't want our money...what they want is for us to not complain, pay top dollar for obsolete, weak sauce technology and thank them for the opportunity to do so.

[vario]

Funny thing is, i can guarantee that there will be meeting with some managers and accountants , and they gonna come to a conclusion :
"Look, our enthusiast parts arent selling, people dont want them anymore, we must build some tablet chips ASAP"

[hokiealumnus]

Yeah, how dare they violate Intel's almighty directive. Don't they know that the word of Intel is to obeyed as you would that of your higher power? We should never question Intel, speak poorly of Intel or do anything that Intel might disagree with and joyously accept whatever half-baked product that Intel feels like foisting upon us, thanking them profusely for their benign benevolence and the opportunity and privilege to be their customer...

/sarcasm

Or, rather than acting like a child, you could look at it from our perspective. The rest of us respected the agreements we made. We gave our word and won't break it. [H] & Tom's intentionally set out to screw everyone else over for a few pageviews.

Grow up.

[SKYMTL]

Or, rather than acting like a child, you could look at it from our perspective. The rest of us respected the agreements we made. We gave our word and won't break it. [H] & Tom's intentionally set out to screw everyone else over for a few pageviews.

Grow up.

Yes and no. Anyone is free to source processors from anywhere they want, thereby bypassing the NDA. In this case, Tom's got their early ES from a mobo manufacturer while [H] did the same thing. I'm sure either of us could go the same route but we value our relationship with Intel and would rather not burn that bridge in case the CPU access through the mobo guys dries up.

[hokiealumnus]

I guess I'm a little more stringent than they are about NDAs. IMHO it's a cop out to go that route, then claim they weren't under NDA. Obviously, results from the chip itself are under NDA. They went the circuitous route to get it early in order to technically avoid agreeing to an NDA. They still broke it, even if they didn't technically agree to it. It's not about who they got the chip from to me, it's about respecting everyone else.

Maybe I'm too strict about that for my own good.

Heh, plus the entire tone of his post implying we are somehow beholden to Intel by simply agreeing to abide by their NDA hit a nerve, as you could probably tell.

[lutjens]

Or, rather than acting like a child, you could look at it from our perspective. The rest of us respected the agreements we made. We gave our word and won't break it. [H] & Tom's intentionally set out to screw everyone else over for a few pageviews.

Grow up.

Grow up? Acting like a child? Hardly. Given the way Intel has blatantly flipped the bird at the enthusiast community over the last couple of years, I applaud those who reciprocate, rather than roll over in meek subservient submission. If they made an NDA agreement and broke it, then Intel can pursue them and that's between them and Intel. They did, however, find a creative way around it, and disrespected Intel by doing so. When a company so throughly steeped in arrogance like Intel is right now gets disrespected and insulted, I personally have no problem with that, as Intel has little problem doing it to others. They're behaving like a neighbourhood bully, telling people through their actions that if they don't like it, tough.

So if anyone should "grow up" and stop "acting like a child", it's Intel.

[hokiealumnus]

Grow up? Acting like a child? Hardly. Given the way Intel has blatantly flipped the bird at the enthusiast community over the last couple of years, I applaud those who reciprocate, rather than roll over in meek subservient submission. If they made an NDA agreement and broke it, then Intel can pursue them and that's between them and Intel. They did, however, find a creative way around it, and disrespected Intel by doing so. When a company so throughly steeped in arrogance like Intel is right now gets disrespected and insulted, I personally have no problem with that, as Intel has little problem doing it to others. They're behaving like a neighbourhood bully, telling people through their actions that if they don't like it, tough.

So if anyone should "grow up" and stop "acting like a child", it's Intel.

Unfortunately, you're missing the point. You think by doing that they actually hurt Intel in any way, shape or form? No. Rather than stacking coverage up on one day when the NDA lifts, they gave Intel even more coverage, driving more people to be talking about the new CPUs for a longer time. If they had published when everyone else did, Intel would have gotten one day of coverage for the new CPUs.

No, my man, the point of their "reviews" is to get pageviews and that is it. If it actually hurt Intel, they would give more of a crap about it and pursue the people that let ES CPUs out of their hands when they shouldn't have. Aside from Francois (and possibly the PR team, who got to hear a whole bunch of grief from other reviewers I'm sure), Intel as a whole doesn't care that they broke the NDA, because it keeps their product in front of more people for a longer time period. So rather than even getting so much as a slap on the wrist, they continue on as though nothing ever happened.

You think it somehow hurts Intel? On the contrary, it helps them. But it hurts everyone else that works hard on these reviews by giving the pageview hunting pair an advantage none of the rest of us have (or, in some of our cases, want to try to get because we have respect for the rest of our fellow reviewers). It's not about respect or disrespect for Intel (who, again, benefits from extra coverage), it's about respect for the rest of the reviewers out there.