Sandy Bridge Reviews Released!

Look how a Bulldozer handles Sand....


VVysJ.jpg
 
Give it up AMD... your stock will be fine.

However, who knows anything about Bulldozer, no one so stop pimping it in every thread, no one cares :)
 
Not impressed at all with Intel only supporting dual channel memory.

I'm also not sure what kind of RAM and timings they are using with the 980x setup as the system in my sig easily blows the memory bandwidth and latency away. Probably something like DDR3-1066 with loose timings.

And only 1 x16 or 2 x8 PCI-e support? That is another regression compared to LGA 1366 but is kind of expected as the current gimped LGA 1156 setup is the same.

Looks like I will be holding on to my current setup for quite a while as I wait for Bulldozer to be released.
 
I just found this... which makes me happy. One of my main uses of my machine is running VMs for testing/lab, etc...


VT-d is used for pass-through I/O devices. ESX(i) is the only VMware product that uses VT-d.


VT-x is used for instruction set virtualization. All VMware virtualization products are capable of using VT-x.


Looks like the lack of VT-d shouldn't have any effect on me as I don't plan on running ESXi just VMWare workstation.

I think I'm still leaning towards the 2600k
 
nope. from what i see in the F@H benchmarks
980x is still faster

Barely, def not worth the 3-4x the cost.... you gotta admit, you aren't all that happy. Don't get me wrong, it's an awesome chip. But if you could PRETTY much have the same thing for 1/4 of the price I think you would jump on it.
 
Barely, def not worth the 3-4x the cost.... you gotta admit, you aren't all that happy. Don't get me wrong, it's an awesome chip. But if you could PRETTY much have the same thing for 1/4 of the price I think you would jump on it.

i still am for another big adv folder.

but from the benchmarks Ive seen. in the DC forum at 2600k 4.9ghz and im at 4.4ghz im still getting near 5mins better per frame. And still have more room to o/c.

so yes im still happy with my 980x.
 
As far as gaming it looks like intel made the 2500k an easy choice for gamers and it looks like most other things.
 
FWIW, the Techspot review says Intel moved the launch day up to the 4th.
 
I think I'm going to go with the i5 2500k.

I'm a bit sad that I can't use the integrated HD 3000 graphics along with a discrete GPU on current motherboards, though. I would've liked to run an Eyefinity display setup along with a separate display for television/movies and whatnot like this. Also it would have been nice to be able to take advantage of the Quick Sync stuff and have a built-in backup GPU in case an RMA was ever necessary. These aren't worth giving up the ability to overclock, though, so H67 is out of the question, and I doubt it'll be worth the extra cash (for me) to go with a new Z68 motherboard when they come out.
 
I think I'm going to go with the i5 2500k.

I'm a bit sad that I can't use the integrated HD 3000 graphics along with a discrete GPU on current motherboards, though. I would've liked to run an Eyefinity display setup along with a separate display for television/movies and whatnot like this. Also it would have been nice to be able to take advantage of the Quick Sync stuff and have a built-in backup GPU in case an RMA was ever necessary. These aren't worth giving up the ability to overclock, though, so H67 is out of the question, and I doubt it'll be worth the extra cash (for me) to go with a new Z68 motherboard when they come out.

Have fun with your DX10 graphics.
 
Insane 2600K performance! I never expected it to be this good! I thought it will only edge out the i7 950 a little bit. But based on the reviews, it actually does trade blows even with the i7 980X! Simply amazing! Now for those who already have an i5/i7, I hope you won't be tempted to upgrade ;p
 
Insane 2600K performance! I never expected it to be this good! I thought it will only edge out the i7 950 a little bit. But based on the reviews, it actually does trade blows even with the i7 980X! Simply amazing! Now for those who already have an i5/i7, I hope you won't be tempted to upgrade ;p

When you compare SB to Lynnfield at equal clocks, it's not compelling at all for us to upgrade. I had a feeling they would pull this card, and it seems almost no review site (currently) is recognizing this. I.e. some are calling for a landslide victory when you compare the i5 760 vs. the i5 2500K, but then fail to realize the clock advantage at hand.

The i7 970 (3.2Ghz) is the only one close to the clock rates of the 2400 and 2500 (3.1 and 3.3Ghz respectively). Obviously we at H aren't running our Lynnfields at stock, so all those reviews are kind of missing critical information in that respect.
 
When you compare SB to Lynnfield at equal clocks, it's not compelling at all for us to upgrade. I had a feeling they would pull this card, and it seems almost no review site (currently) is recognizing this. I.e. some are calling for a landslide victory when you compare the i5 760 vs. the i5 2500K, but then fail to realize the clock advantage at hand.

The i7 970 (3.2Ghz) is the only one close to the clock rates of the 2400 and 2500 (3.1 and 3.3Ghz respectively). Obviously we at H aren't running our Lynnfields at stock, so all those reviews are kind of missing critical information in that respect.

That's why I said i5/i7 shouldn't be tempted to upgrade.
 
keep in mind its only trading blows with the 980x because some of the tests are thread limited. some are limited to 4 and 8 threads. So its at a disadvantage with its 12 threads. but compare it to programs that can use all threads it doesnt trade blows with it.
 
Have fun with your DX10 graphics.

What? I specifically said I I'm a bit disappointed that I won't be able to use the integrated GPU with a P67 board, and even then (on a Z68 board) it would be augmenting the discrete GPU, driving a fourth monitor or functioning as a backup GPU. Besides, I'm not sure why DX10 vs 11 matters at the (low) level of performance it provides.

For a mobile platform, on the other hand (especially something like Lenovo's tablets, which never have discrete GPUs), I'm very excited for the performance increases SB will provide on both the CPU and GPU sides of things.

I'm not sure what you mean to contribute with this post.
 
When you compare SB to Lynnfield at equal clocks, it's not compelling at all for us to upgrade. I had a feeling they would pull this card, and it seems almost no review site (currently) is recognizing this. I.e. some are calling for a landslide victory when you compare the i5 760 vs. the i5 2500K, but then fail to realize the clock advantage at hand.

The i7 970 (3.2Ghz) is the only one close to the clock rates of the 2400 and 2500 (3.1 and 3.3Ghz respectively). Obviously we at H aren't running our Lynnfields at stock, so all those reviews are kind of missing critical information in that respect.

But you wouldn't be comparing equal clocks in the real world because Lynnfield's can't reach SB clocks. Most of the reviews were clocking a full gigahertz over your sig proc.
 
You guys hear something? I do............
http://i.imgur.com/f316F.jpg[/I MG][/QUOTE]lawl. You mean the 2 year late CPU that still lacks a release date? The one that AMD won't release benchmarks or seed for previews, despite the hazy "Q2" release period?

Cards are on the table and AMD probably can't understand how they got blindsided again... on a mid-range socket vs its own high end. :D

BD 2 in 4 years might have a shot. Uh huh.
 
Wouldn't SB be faster when the F@H client is upgraded to AVX? And aren't GPU PPD much more than a CPU can get? Using less power per point? Kind of seems like a bad excuse to justify the 980x.
 
Wouldn't SB be faster when the F@H client is upgraded to AVX? And aren't GPU PPD much more than a CPU can get? Using less power per point? Kind of seems like a bad excuse to justify the 980x.
For the most part thats true, but cpu's with 8 threads or more can run bigadv SMP units where shorter completion times result in exponentially higher bonus points. The TPF difference he is talking about (~25min vs ~20min) is probably worth around 20k ppd.

Still, I expect that once we get some more test results with high speed ram (2133+) this TPF difference will shrink considerably and AVX would further close the gap. The 980x was a dubious value proposition to begin with, and with the availability of the 2600k I don't see how anyone could recommend it for a new build. About the only viable use for a hex core 1366 chip now is to extend the life of an existing 1366 setup. Even then however, the prices will need to drop considerably as its currently cheaper to get a new motherboard and 2600k than to put a hex in your x58.
 
It's 2 generations behind, what more did you expect? C4C it's already slower then lynnfields. AMD Is the cheapo budget area of like $100-130, not the $200+ CPU area, 6 cores or not.
 
ESX(i) is the only VMware product that uses VT-d.

VT-x is used for instruction set virtualization. All VMware virtualization products are capable of using VT-x.

Looks like the lack of VT-d shouldn't have any effect on me as I don't plan on running ESXi just VMWare workstation.

Don't forget KVM on Linux!
 
Isn't it nice building a OC top of the range PC less than a year ago and having it being obsolete already :(. You think we'd learn....
 
Anybody see any review of sb and i7 950 at the same clocks. I wouldnt mind seeing how much faster sandy vs an i7.
 
I must say, I am super impressed with mobile sandy bridges. Wow @ the power consumption. Finally some decent integrated gfx from intel as well
 
From the TR review:

One thing that you'll want to watch for in Sandy Bridge mobos: with the new Turbo algorithm, our understanding is that VRM design can potentially influence the amount of time the processor spends resident in Turbo mode clock speeds. It's possible motherboard designs may once again influence overall performance in way that they haven't for several years. Could make things interesting. Stay tuned for Geoff's take on four of the first such boards based on the P67, which we should be posting soon.

Looks like you may want to be careful about the choice of motherboard if you plan on using Turbo. I assume the comment either points to voltage sag or overvolting and the effects it has on Turbo.
 
Crackhead Charlie trashes SB in a "review": http://semiaccurate.com/2011/01/02/sandy-bridge-biggest-disapointment-year/

The sad thing is that people still take this fool seriously.

Intel's response is funny, as is the first comment:

Editor's note: Intel provided S|A with the following response.

Intel: 'Charlie tested unreleased Sandybridge hardware without drivers. No hardware can function properly without drivers on any operating system, and as such, these results are not valid. We encourage Semiaccurate to retest the platform with the appropriate drivers in place'

first comment said:
NotAGoodName January 2, 2011, 11:28 p.m.

Welcome to Linux.
 
Back
Top