That's probably because of the 256-bit divider.
But that's nothing compared to what Haswell will bring us. Twice the throughput for every SIMD instruction, and 8x parallel memory access!
]Not to be argumentative, but couldn't you just look at a SB @ ~4.5-4.7 ghz, subtract ~10% power[/B] and be pretty confident in the numbers?
It doesn't seem like it really changes the SB-E vs SB/IVB equation much, does it?
If there was a six core IVB, then we'd be talking about some excitement (on my part)
Yes, that does seem to be a rather intriguing finding... I'd be good with that for 24/7 clocks and I'd take the big power savings over the 2600k as well. I think I'll leave the 2600k in its box until I see the MC deal on the 3770k deal.
Dons IB i5 is right there with my 5.2GHz SB i5 and he isn't even pushing his chip.
I am happy with 4.5GHz as I didn't want a chip that would practically be dead in a matter of months and having high temps for the sake of higher speed is not my concern as the heat has to go somewhere and that somewhere is my room which will be a furnace and I am not comfortable with that. Unless you're willing to supply me with a dozen or so spare IB processors and an air conditioner, I'll just be happy with what I can get with it.
I don't think I ever tried to convince someone with SB to upgrade to IB before. I've always said to upgrade every other generation, not every generation.
I don't think it is that simple. If 8-10C parts for IB-E are released, getting that many cores at a decent OC, but keeping power/heat controllable will be key. For many parallel applications, 8-10C @ 4.2ghz would be >>>>>>4C @ 5.0ghz+.
If IB-E allows a good under-volt + overclock, it could be awesome. Predicting based on IB is not that straight-forward though IMHO. There were a lot of differences between SB and SB-E. Add into the fact that IB-E will have a much larger die (Cache especially) and it could get interesting.
Some reviews do have the 3570K.The reviews are for 3770k. I'm curious though as far as price to performance, how the 3570k stacks up against the 2600k and 2500k.
any point upgrading from an i7 950? Primary functions gaming and 1080p movie watching
What the hell does crapdozer have to do with anything? Yeah I get your stupid passive aggressive comment pal maybe you should go to some other website that is about conspiracy theories, anyway, I expected a better overclock considering its a smaller node and an improved process, this has been consistently happening for the past 5 years and has come to an end.
The IPC increase is miniscule and there is no reason for me to upgrade my 2600k.
That one even has this tidbit (for posters claming defeat to SB):
Take note that the Ivy Bridge CPUs look good at any load and seem to have no weak spots in comparison with the Sandy Bridge series. Well, where could such weak spots spring from? Their execution cores, memory controller and cache memory duplicate the Sandy Bridge microarchitecture with but minor optimizations which are reflected in the performance charts.
One of the reviews listed (not sure which it was) said IB ran better with 3DS Max and Maya which are multi-threaded
A 3 to 5 % performance difference is nothing to talk about and certainly doesn't create a 200-300mhz performance gap. Not speaking down on Ivy, just the false information Ivy owners are spreading.
Wow, showing 2-7% IPC improvement: http://www.hardocp.com/article/2012/04/23/intel_ivy_bridge_processor_ipc_overclocking_review/3
That's quite minimal, and probably not enough to make up for the lack of OC'ing headroom.
I'm passing on this one for now unless Microcenter has some crazy deals going.
I'm not saying it's anything to write home about if you're already running a SB chip. But, your statment saying that video encoding is 30% slower is false (see: http://www.guru3d.com/article/core-i7-3770k-and-3750-review-with-z77/17). And this shows that a 4.5 IB is pretty much on par with a 4.7 SB, so there's the 200 mhz difference you were talking about: http://www.tomshardware.com/reviews/ivy-bridge-benchmark-core-i7-3770k,3181-9.html and this is with multi-threaded applications. Here's another one running 3DS Max and Blender: http://www.xbitlabs.com/articles/cpu/display/core-i7-3770k-i5-3570k_7.html#sect0. More video manipulation and raytrace rendering: http://pcper.com/reviews/Processors...vy-Bridge-Processor-Review/OpenCL-Performance. Again, I'm not saying this is a major upgrade to SB, but to discredit altogether isn't accurate either. It's still a decent chip and for those of us w/o SB or even one generation behind that, this is good upgrade. I'll be happy with a 4.5 OC.
For the sake of advancement I do hope you get 5ghz. Don's Ivy gets outperformed by identical clocked Sandie's in some benches on this board, I'll take those results over a questionable 5.2ghz overclock. Lord only knows how many errors you're spitting out at those speeds.
A 4.5ghz Ivy is *NOT* equal to a 4.8ghz Sandy in multi-threaded tasks. Its 30% slower video encoding. Clearly in apps that count it's no faster.
Looks like all 5GHZ SBs are going to be under threat from 4.5GHZ IVBs.
5GHZ IVB vs 5.5GHZ SB. IVB wins all benchmarks; pretty much.
http://ud3.weebly.com/index.html
Looks like all 5GHZ SBs are going to be under threat from 4.5GHZ IVBs.
5GHZ IVB vs 5.5GHZ SB. IVB wins all benchmarks; pretty much.
http://ud3.weebly.com/index.html
Looks like all 5GHZ SBs are going to be under threat from 4.5GHZ IVBs.
5GHZ IVB vs 5.5GHZ SB. IVB wins all benchmarks; pretty much.
http://ud3.weebly.com/index.html
Updated link. I quoted the 'losing' bencher. Seems like not all results are up there. You're right though.Except it didn't? 5.5ghz SB = 10.8 cinebench, 3770k = 9.3
They didn't even run the same tests or testbed on both CPU's.
Fugger matched my or exceeded my speeds with less cooling. Its no joke, Ivy Bridge beat my Sandy Bridge at everything overclocked.