• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Why is Sandy Bridge "not promising"?

I havn't seen that price list before. From what I have read so far was with an understanding there would be quite the premium on the K series, which made me think the vast majority of the market wouldn't have these and so therefore not to much of an interest for mobo makers to make the higher spec boards.

But looking at that, without the really high premium I can see the price difference isn't that much difference than between the difference in phenom ii's and the BE's. Don't see how this will be a put off for mobo makers like I thought.

Overall looking more interesting than before I read cmndr's two articles. I would just now really like to see some well done bench's against chips with the same clock speed, see how they fair up.
 
if intel do limit the overclocking this will be a time for amd to shine once again like the amd 64/opteron 144 days... and then will come the true overclock : price : performance ratio :). I bet amd are sitting back tightly waiting for this :D
 
I do hope that the figures quoted in the article are accurate it will make the choice rather easy when it comes time to fork out some money for a new Motherboard and CPU. K prefix for the win if you overclock.
 
It's fairly promising, certainly not a dud architecture. 20-30% clock for clock advantage over Westmere, just like Nehalem had 20-30% over Penryn.
Considering it'll be replacing the current i5/i7 price points, and brings EFI motherboards It's not too bad.

The only reason it might not be considered exciting is that a) The overclockable parts are going to carry a premium, and b) bulldozer hits in April and that could be even more interesting due to a fairly radical new architechture.

Honestly though, I think it's that apps haven't caught up yet, Sandy bridge isn't exciting for many people because most of us aren't doing anything thats really taxing our i5's/i7's, hell with a decent overclock even a venerable Q6600 breezes through everything. It's hard to get excited over something that you won't even notice most of the time.
 
paying for the ability to overclock goes against OC principles. Intel just attempting to cash in on enthusiasts and their willingness to part with their hard earnd cash.

I hope BD can bring something new to the table. It's been too long since Intel had some serious competition and when theres competition the main people that win are us the consumers
 
pinkpotato said:
Definately not worth upgrading any x58 system, it is not aimed to beat it, it doesnt even support tripple channel memory for a start...:rolleyes:
Click to expand...

Your right it's not worth upgrading from an x58 setup or even from something like a Core i5 750.

Tripple channel memory though is no big loss it doesn't really make much difference to performance.
 
The way i understand Sandy Bridge is it will turbo upto whatever limit (turbo ment to be fair chunk higher than before) then within its thermal capacity it can go beyond that briefly yeilding much higher turbo state as when required until it reaches it TDP.

So within reason with good cooling it can effectively overclock itself by a healthy margin.

The unlocked versions i read will overclock just as high as locked versions but for longer with maybe higher TDP via multiplier making manual overclocking possible upto similar levels the locked chip can briefly reach.

Therefore only differences i see with locked vs unlocked chip is power consumption and that people moaning about restricted overclocking a non-issue.
 
Well it all depends on usage - when running lightly threaded applications it should be able to turbo boost itself up to its limit without hitting the TDP limit and throttling back. The maximum turbo of the i7 2600(K) is 3.8GHz (compared to 3.4GHz stock) and for the i5 2500(K) it is 3.7GHz (compared to 3.3GHz stock).

However, when running heavily threaded applications (like video encoding or rendering) all cores will be running at full whack and the thermal limit will be reached at stock speeds - so there will be no turbo boosting at all. Hence, turbo boost is not really overclocking and why overclockers usually turn it off.

Looking at this the unlocked i7 2600K can reach 5GHz using air cooling, so i'm certain a lot of people will pay for the K versions, they don't cost much more than the standard version. It certainly is a bit of price gouging by Intel - but then again these CPUs will not be extortionately priced and will offer some of the best performance on the market.
 
Impressive that it can reach 5Ghz on air cooling. If lga1155 cpu can reach such speed, what do reckon the lga2011 high end cpu will be capable of ?
 
WingZero30 said:
Impressive that it can reach 5Ghz on air cooling. If lga1155 cpu can reach such speed, what do reckon the lga2011 high end cpu will be capable of ?
Click to expand...

To be honest, based on the performance of s1156 vs s1366, I reckon the Sandy Bridge (45nm) s2011 CPUs will overclock about as far the s1155 chips. However, s2011 will throw hex and octo cores into the mix - so for multi-threaded applications they should be MUCH faster (as well as much more expensive, all-in).
 
cmndr_andi said:
To be honest, based on the performance of s1156 vs s1366, I reckon the Sandy Bridge (45nm) s2011 CPUs will overclock about as far the s1155 chips. However, s2011 will throw hex and octo cores into the mix - so for multi-threaded applications they should be MUCH faster (as well as much more expensive, all-in).
Click to expand...

How many cores are going to be be kept on adding to the cpu before the limit is reached? Will it be defined by the die process? As Rockwell is going to be on 16nm die, will it be capable of introducing 10/12 cores:eek:? Does picometre come after nanometre? Seems to be that technology is reaching atomic size lol:eek:.
 
The limit is fast approaching because atoms are only ~0.1 nm in diameter. Beyond 11 nm, the number of atoms per transistor would probably drop below 1000 plus you get quantum effects interfering. I don't think there's any plans for current CMOS technology to reach below 11 nm and a completely new approach would probably be needed (nanoelectronics).
 
Last edited:
cmndr_andi said:
However, when running heavily threaded applications (like video encoding or rendering) all cores will be running at full whack and the thermal limit will be reached at stock speeds - so there will be no turbo boosting at all. Hence, turbo boost is not really overclocking and why overclockers usually turn it off.
Click to expand...

With Sandy Bridge this is where the difference lies, SB can turbo all cores not just 1 or 2 single threaded application regardless of temperature.

The other difference being TDP figures no longer relate to its maximum safe TDP amount as they currently do, which is the reason for brief moments the turbo can boost even higher until it reaches the cpu true thermal limit at say 125w or something when cpu is rated at 75w for example.
 
You must log in or register to reply here.
Back
Top Bottom