The Sandy Bridge Preview
by Anand Lal Shimpi on August 27, 2010 2:38 PM ESTOverclocking Controversy
It wasn’t until the Pentium II that Intel started shipping multiplier locked CPUs. Before then you could set the multiplier on your CPU to anything that was supported by the line, and if you had a good chip and good enough cooling you just overclocked your processor. Intel’s policies changed once remarking, the process of relabeling and reselling a lower spec CPU as a higher one, started to take off.
While multipliers were locked, Intel left FSB overclocking open. That would be an end user or system integrator decision and not something that could be done when selling an individual CPU. However, ever since before the Pentium III Intel had aspirations of shipping fully locked CPUs. The power of the enthusiast community generally kept Intel from exploring such avenues, but we live in different times today.
Two things have changed Intel’s feelings on the topic. First and foremost is the advent of Turbo Boost. So long as Intel doesn’t artificially limit turbo modes, we now have the ability to run CPUs at whatever clock speed they can run at without exceeding thermal or current limits. We saw the first really exciting Turbo with Lynnfield, and Sandy Bridge is going to expand on that as well. On the flip side, Intel has used Turbo as a marketing differentiator between parts so there’s still a need to overclock.
The second major change within Intel is the willingness to directly address the enthusiast community with unlocked K-series SKUs. We saw this recently with the Core i7 875K and Core i5 655K parts that ship fully unlocked for the overclocking community.
The K-series SKUs, these will be more important with Sandy Bridge
With Sandy Bridge, Intel integrated the clock generator, usually present on the motherboard, onto the 6-series chipset die. While BCLK is adjustable on current Core iX processors, with Sandy Bridge it’s mostly locked at 100MHz. There will be some wiggle room as far as I can tell, but it’s not going to be much. Overclocking, as we know it, is dead.
Well, not exactly.
Intel makes three concessions.
First and foremost we have the K-series parts. These will be fully unlocked, supporting multipliers up to 57x. Sandy Bridge should have more attractive K SKUs than what we’ve seen to date. The Core i7 2600 and 2500 will both be available as a K-edition. The former should be priced around $562 and the latter at $205 if we go off of current pricing.
Secondly, some regular Sandy Bridge processors will have partially unlocked multipliers. The idea is that you take your highest turbo multiplier, add a few more bins on top of that, and that’ll be your maximum multiplier. It gives some overclocking headroom, but not limitless. Intel is still working out the details for how far you can go with these partially unlocked parts, but I’ve chimed in with my opinion and hopefully we’ll see something reasonable come from the company. I am hopeful that these partially unlocked parts will have enough multipliers available to make for decent overclocks.
Finally, if you focus on multiplier-only overclocking you lose the ability to increase memory bandwidth as you increase CPU clock speed. The faster your CPU, the more data it needs and thus the faster your memory subsystem needs to be in order to scale well. As a result, on P67 motherboards you’ll be able to adjust your memory ratios to support up to DDR3-2133.
Personally, I’d love nothing more than for everything to ship unlocked. The realities of Intel’s business apparently prevent that, so we’re left with something that could either be a non-issue or just horrible.
If the K-series parts are priced appropriately, which at first indication it seems they will be, then this will be a non-issue for a portion of the enthusiast market. You’ll pay the same amount for your Core i7 2500K as you would for a Core i5 750 and you’ll have the same overclocking potential.
Regardless of how they’re priced, what this is sure to hurt is the ability to buy a low end part like the Core i3 530 and overclock the crap out of it. What Intel decides to do with the available multiplier headroom on parts further down the stack is unknown at this point. If Intel wanted to, it could pick exciting parts at lower price points, give them a few more bins of overclocking headroom and compete in a more targeted way with AMD offerings at similar price points. A benevolent Intel would allow enough headroom as the parts can reliably hit with air cooling.
The potential for this to all go very wrong is there. I’m going to reserve final judgment until I get a better idea for what the Sandy Bridge family is going to look like.
Facebook
Twitter
RSS
200 Comments
View All Comments
overzealot - Saturday, August 28, 2010 - link
Now, that's a name I've not heard in a long time. A long time.mapesdhs - Saturday, August 28, 2010 - link
Seems to be Intel is slowly locking up the overclocking scene because it has no
competition. If so, and Intel continues in that direction, then it would be a great
chance for AMD to win back overclocking fans with something that just isn't
locked out in the same way.
Looking at the performance numbers, I see nothing which suggests a product that
would beat my current 4GHz i7 860, except for the expensive top-end unlocked
option which I wouldn't consider anyway given the price.
Oh well, perhaps my next system will be a 6-core AMD.
Ian.
LuckyKnight - Saturday, August 28, 2010 - link
Do we have something more precise about the release date? Q1 is what - Jan/Feb/March/Apri?Looking to upgrade a core 2 duo at the moment - not sure whether to wait
mino - Saturday, August 28, 2010 - link
Q1 (in this case) means tricle amounts in Jan/Feb, mainstream availability Mar/April and worth-buying mature mobos in May/June timeframe.tatertot - Saturday, August 28, 2010 - link
Intel has already announced that shipments for revenue will occur in Q4 of this year. So, January launch.They've also commented that Sandy Bridge OEM demand is very strong, and they are adjusting the 32nm ramp up to increase supply. So January should be a decent launch.
Not surprising-- these parts have been in silicon since LAST summer.
chrsjav - Saturday, August 28, 2010 - link
Do modern clock generators use a quartz resonator? How would that be put on-die?iwodo - Saturday, August 28, 2010 - link
Since you didn't get this chip directly from Intel , i suspect there were no reviews guideline for you to follow, like which test to run and what test not to run etc.Therefore those benchmark from Games were not a results of special optimization in drivers. Which is great, because drivers matter much more then Hardware in GPU. If these are only early indication of what Intel new GPU can do, i expect there are more to extract from drivers.
You mention 2 Core GPU ( 12 EU ) verus 1 GPU ( 6 EU ), Any Guess as to what "E" stand for? And it seems like a SLI like tech rather then actually having more EU in one chip. The different being SLI or crossfire does not get any advantage unless drivers and games are working together. Which greatly reduces the chances of it working at full performance.
It also seems every one fail to realize one of the greatest performance will be coming from AVX. AVX will be like MMX again when we had the Pentium. I cant think of any other SSE having as great important to performance as AVX. Once software are specially optimize for AVX we should get another major lift in performance.
I also heard about rumors that 64bit in Sandy Bridge will work much better. But i dont know if there are anything we could test this.
The OpenCL sounds like a Intel management decision rather then a technical decision. May be Intel will provide or work with Apple to provide OpenCL on these GPU?
You also mention that Intel somehow support PCI -Express 2.0 with 1.0 performance. I dont get that bit there. Could you elaborate? 2.5GT/s for G45 Chipset??
If Intel ever decide to finally work on their drivers, then their GPU will be great for entry levels.
Are Dual Channel DDR3 1333 enough for Quad Core CPU + GPU? or even Dual core CPU.
Is GPU memory bandwidth limited?
Any update on Hardware Decoder? And what about transcoding part?
Would there be ways to lock the GPU to run at Turbo Clock all the time? Or GPU gets higher priority in Turbo etc..
How big is the Die?
P.S - ( Any news on Intel G3 SSD? i am getting worried that next Gen Sandforce is too good for intel. )
ssj4Gogeta - Saturday, August 28, 2010 - link
I believe EU means execution units.DanNeely - Sunday, August 29, 2010 - link
"You also mention that Intel somehow support PCI -Express 2.0 with 1.0 performance. I dont get that bit there. Could you elaborate? 2.5GT/s for G45 Chipset??"PCIE 2.0 included other low level protocol improvements in addition to the doubled clock speed. Intel only implemented the former; probably because the latter would have strangled the DMI bus.
"Are Dual Channel DDR3 1333 enough for Quad Core CPU + GPU? or even Dual core CPU."
Probably. The performance gains vs the previous generation isn't that large and it was enough for anything except pathological test cases (eg memory benchmarks). If it wasn't there'd be no reason why Intel couldn't officially support DDR3-1600 in their locked chipsets to give a bit of extra bandwidth.
chizow - Saturday, August 28, 2010 - link
@AnandCould you please clarify and expand on this comment please? Is this true for all Intel chipsets that claim support for PCIe 2.0?
[q]The other major (and welcome) change is the move to PCIe 2.0 lanes running at 5GT/s. Currently, Intel chipsets support PCIe 2.0 but they only run at 2.5GT/s, which limits them to a maximum of 250MB/s per direction per lane. This is a problem with high bandwidth USB 3.0 and 6Gbps SATA interfaces connected over PCIe x1 slots. With the move to 5GT/s, Intel is at feature parity with AMD’s chipsets and more importantly the bandwidth limits are a lot higher. A single PCIe x1 slot on a P67 motherboard can support up to 500MB/s of bandwidth in each direction (1GB/s bidirectional bandwidth).[/q]
If this is true, current Intel chipsets do not support PCIe 2.0 as 2.5GT/s and 250MB/s is actually the same effective bandwidth as PCIe 1.1. How did you come across this information? I was looking for ways to measure PCIe bandwidth but only found obscure proprietary tools not available publicly.
If Intel chipsets are only running at PCIe 1.1 regardless of what they're claiming externally, that would explain some of the complaints/concerns about bandwidth on older Intel chipsets.