Intel Sandy Bridge 2600K and 2500K Processors @ [H]

I took a quick look at the datasheet but couldn't find what intel is calling max vcore and abs max vcore.

Anyone know these numbers?
 
ziddey said:
I took a quick look at the datasheet but couldn't find what intel is calling max vcore and abs max vcore.

Anyone know these numbers?
Click to expand...

It's been established that the max Vcore is 1.52v which is surprisingly the highest VID they produce.

There has been one person on OCN that reported seeing some degradation at 5GHz+ with ~1.51v. So I would keep it below 1.45v for now.
 
SolidBladez said:
It's been established that the max Vcore is 1.52v which is surprisingly the highest VID they produce.

There has been one person on OCN that reported seeing some degradation at 5GHz+ with ~1.51v. So I would keep it below 1.45v for now.
Click to expand...

What about the stock Vcore?
I didn't see what it was in the article

1.25v like previous I7's?

or 1.2v

Just wondering.
 
so far for me I am at 1.25 volts with 4.4GHz stable. Will try for 4.5GHz tomorrow.
 
Not sure if this has been mentioned yet or not (didn't read through all 15 pages of this thread) but the i5 655k is mislabeled as an i7 655k in at least some of the graphs.
 
I'm whelmed by these new mid range chips. But it gives me good hope for the high end 2011 chips coming later.

Until then I see no reason to upgrade from a x58 board. Most 67 boards lack greater than 2 way SLI/Crossfire and lower memory capacity. Both which wuld be a PITA coming from x58, for no real performance gain.

One question however, How has the memory controller improved? Does filling all the channels of the sandy bridges controller still slightly limit overclock like with other intel chips? Or does it not really effect it?
 
Nothing new here except if you notice the test set-up they only used this ram... 4GB G.Skill Perfect Storm 2,200MHz CL8 ... and came to the conclusions without even testing others. Now why would you call this a good review?
 
Who cares about the model of the ram used, what counts is it's speed and timings.
They were obviously able to run those sticks at plenty of different settings, covering most of whats available on the market. What more do you want ?
 
Infrabasse said:
Who cares about the model of the ram used, what counts is it's speed and timings.
They were obviously able to run those sticks at plenty of different settings, covering most of whats available on the market. What more do you want ?
Click to expand...

So...you test a Corvette at different speeds and assume a Chevette will suffice? :rolleyes: If you are going to do a review then review all the ram....assumptions are a dime a dozen.
 
That metaphore, it just doesn't work.
RAM is pretty much defined by its frequency and latencies. This isn't a RAM review, it's a memory controller benchmark.
 
Infrabasse said:
That metaphore, it just doesn't work.
RAM is pretty much defined by its frequency and latencies. This isn't a RAM review, it's a memory controller benchmark.
Click to expand...

Yep... it doesn't matter other than frequency & latency. Not sure why there's confusion as to this.
 
rickss69 said:
So...you test a Corvette at different speeds and assume a Chevette will suffice? :rolleyes: If you are going to do a review then review all the ram....assumptions are a dime a dozen.
Click to expand...

No, I'll review a corvette from Jim's dealership and assume the one at Bob's dealership will preform the same. You clearly don't understand how ram works if you think there's a significant difference between brands running the same speeds.
 
vengence said:
No, I'll review a corvette from Jim's dealership and assume the one at Bob's dealership will preform the same. You clearly don't understand how ram works if you think there's a significant difference between brands running the same speeds.
Click to expand...

You clearly did not read the review. They tested this ram "4GB G.Skill Perfect Storm 2,200MHz CL8" and would have you believe that any ram running 1333/1600 would suffice or be optimal for this set-up...seemingly pulling it out of their ass so to speak.
 
I really don't see what the issue is. They took fast RAM, and ran it at a variety of speeds and timings. Are you saying that running 2133 RAM at 1600 is somehow different than running 1600 RAM at 1600? 1600 CL9 is 1600 CL9 no matter what the RAM is actually rated for. Why bother switching in 10 different physical sets of RAM when you can use one and change the settings. No different than running a fast processor at a lower speed to simulate a lower-speed processor.
 
The issue is someone making broad claims with no concrete data to back it up. By your estimation if I downclocked my Gulftown to a lower frequency, then any other cpu regardless of type would perform as well...homey don't think so.
We all know most any ram will work but to claim an optimal set-up without thorough testing is less than impressive. If one really believes there is no difference between brands of memory they are in for a rude awakening.
 
Last edited:
Forceman said:
I really don't see what the issue is. They took fast RAM, and ran it at a variety of speeds and timings. Are you saying that running 2133 RAM at 1600 is somehow different than running 1600 RAM at 1600? 1600 CL9 is 1600 CL9 no matter what the RAM is actually rated for. Why bother switching in 10 different physical sets of RAM when you can use one and change the settings. No different than running a fast processor at a lower speed to simulate a lower-speed processor.
Click to expand...

To some degree this is correct. With regard to CPUs that only works if they are adjustable via multiplier and assuming that L2/L3 cache are the same as are any other features such as instruction set support, manufacturing process and of course, the processor core count.

rickss69 said:
The issue is someone making broad claims with no concrete data to back it up. By your estimation if I downclocked my Gulftown to a lower frequency, then any other cpu regardless of type would perform as well...homey don't think so.
We all know most any ram will work but to claim an optimal set-up without thorough testing is less than impressive. If one really believes there is no difference between brands of memory they are in for a rude awakening.
Click to expand...

Wow, just wow.

Yes, if you down clock RAM and adjust the timings it will perform like any other RAM at the same settings. It's not the same as down clocking a six core processor with more cores to the same frequency as a quad core and comparing them. It's not in the same ball park and it's not even the same sport.

Just how is RAM "different" between brands when you set latency values manually? If I take Corsair RAM and G.Skill RAM, run them at the same frequencies and timings, I'll get the same exact performance results. Now that doesn't address board compatibility issues, voltages or automatic settings via SPD or XMP values but from a performance stand point they'll be the same. I'm not sure why you believe this to be incorrect but your wrong.
 
Ok - So everyone here just throws a dart when they order memory? Not a thought about brand or any other consideration? Perhaps the very cheapest ram you can purchase? Please...

Tell you what...I'll take my GTX Dominator's and you take some generic set of your choosing and I will prove my point with any benchmark you choose. All ram are not created equal...there is a difference between running memory and using memory.
 
Last edited:
rickss69 said:
The issue is someone making broad claims with no concrete data to back it up. By your estimation if I downclocked my Gulftown to a lower frequency, then any other cpu regardless of type would perform as well...homey don't think so.
We all know most any ram will work but to claim an optimal set-up without thorough testing is less than impressive. If one really believes there is no difference between brands of memory they are in for a rude awakening.
Click to expand...

The difference in RAM is the speed and timing - DDR3-1600 at CL9 from Corsair is not going to different in any significant way from DD3-1600 at CL9 from G.Skill. Half the time they all use the same RAM chips anyway, the only difference is the PCB they put them on (or the heatspreader they attach). The testing they did was to find the difference between different RAM speeds and timings, not to try to determine what the "optimal" memory was for SB. Did we even read the same article?

The Gulftown is a bad example because it is 6 core, but if I downclock an i7-950 to i7-920 speeds, it is going to be indistinguishable from an actual i7-920.

Edit: Dan_D beat me to it. And yes, I'd say most people will buy whatever name brand memory is cheapest for the speed and timings they want.
 
Forceman said:
The difference in RAM is the speed and timing - DDR3-1600 at CL9 from Corsair is not going to different in any significant way from DD3-1600 at CL9 from G.Skill. Half the time they all use the same RAM chips anyway, the only difference is the PCB they put them on (or the heatspreader they attach). The testing they did was to find the difference between different RAM speeds and timings, not to try to determine what the "optimal" memory was for SB. Did we even read the same article?

The Gulftown is a bad example because it is 6 core, but if I downclock an i7-950 to i7-920 speeds, it is going to be indistinguishable from an actual i7-920.

Edit: Dan_D beat me to it. And yes, I'd say most people will buy whatever name brand memory is cheapest for the speed and timings they want.
Click to expand...

From the review...


Conclusion of what to Buy
If you're the type of person that runs dozens of applications all at once, then a higher memory frequency does help, particularly when you're running demanding software. However, our testing shows that memory rated at over 1,866MHz doesn't give much extra performance. Worse still, in some applications only 1,333MHz memory gives a performance penalty, meaning that 1,600MHz memory is fine.

If you're doing anything other than heavy multi-tasking - this goes for gamers in particular - then a 1,600MHz or 1,866MHz kit is plenty. You could opt for CL8, as we saw some advantage in the video encoding test, but we wouldn't obsess over this factor, especially if a CL9 kit is much cheaper.

Always remember to buy from a reputable manufacturer, though, as cheap and unbranded memory tends to cause more trouble than the saving is worth. We'd also strongly advise you to opt for at least a 4GB, dual-channel kit. We expect 8GB kits to be common this time around, and these are worth considering if you run a few resource-heavy applications concurrently.


More interesting is to read above this portion.
 
While its nice to see the new midrange chips smoke everything out currently I see no reason to upgrade my setup.
 
rickss69 said:
From the review...


Conclusion of what to Buy
If you're the type of person that runs dozens of applications all at once, then a higher memory frequency does help, particularly when you're running demanding software. However, our testing shows that memory rated at over 1,866MHz doesn't give much extra performance. Worse still, in some applications only 1,333MHz memory gives a performance penalty, meaning that 1,600MHz memory is fine.

If you're doing anything other than heavy multi-tasking - this goes for gamers in particular - then a 1,600MHz or 1,866MHz kit is plenty. You could opt for CL8, as we saw some advantage in the video encoding test, but we wouldn't obsess over this factor, especially if a CL9 kit is much cheaper.

Always remember to buy from a reputable manufacturer, though, as cheap and unbranded memory tends to cause more trouble than the saving is worth. We'd also strongly advise you to opt for at least a 4GB, dual-channel kit. We expect 8GB kits to be common this time around, and these are worth considering if you run a few resource-heavy applications concurrently.


More interesting is to read above this portion.
Click to expand...

No one is saying that all brands of memory or even specific model numbers of modules are created equal. What we are saying is that running a particular set RAM modules at different speeds and timings produces the same performance metrics as other brands of RAM at those speeds and timings would. We are saying this because it's true. A more detailed example of this: Corsair Dominator GT 2GB, DDR3 2000MHz modules at DDR3 1600MHz with timings of 8,8,8,21,1T will produce identical performance results as G.Skill DDR3 1600MHz modules running at 8,8,8,21,1T timings. Functionally they are identical when speed and timings are set the same.
 
Dan_D said:
No one is saying that all brands of memory or even specific model numbers of modules are created equal. What we are saying is that running a particular set RAM modules at different speeds and timings produces the same performance metrics as other brands of RAM at those speeds and timings would. We are saying this because it's true. A more detailed example of this: Corsair Dominator GT 2GB, DDR3 2000MHz modules at DDR3 1600MHz with timings of 8,8,8,21,1T will produce identical performance results as G.Skill DDR3 1600MHz modules running at 8,8,8,21,1T timings. Functionally they are identical when speed and timings are set the same.
Click to expand...

Some people believe that their 400hp Camaro is faster than your 400hp Camaro because theirs is red and yours is silver. smh

can't please everyone all the time
 
rickss69 said:
Ok - So everyone here just throws a dart when they order memory? Not a thought about brand or any other consideration? Perhaps the very cheapest ram you can purchase? Please...

Tell you what...I'll take my GTX Dominator's and you take some generic set of your choosing and I will prove my point with any benchmark you choose. All ram are not created equal...there is a difference between running memory and using memory.
Click to expand...

You don't get it.

It's not that people buy dominators because it's faster than some other brand at 2000Mhz, it's that if you buy el cheapo 1333mhz memory, you'd be lucky to clock it to 1600mhz and even if you did, you're going to have to loosen the timings to shit compared to the dominator. If you managed to get the el cheap running at 2000mhz and the same timings as a set of dominators then they'll preform exactly the same.
 
HydroBudz said:
Some people believe that their 400hp Camaro is faster than your 400hp Camaro because theirs is red and yours is silver. smh

can't please everyone all the time
Click to expand...

It is faster though, while each engine is rated at 400hp, the red paint adds 100 hp for a combined 500hp. What a noob.
 
its a sure sign the thread is dead when the it deteriorates into a car debate.
 
Well, Cars and racing are something that I know alot about, but this is not the board for that.
I have been just pulling my hair out trying to get the hardware together to order my new rig. I will be going with the i7-2600K and Asus Deluxe mobo. The RAM is a sticking point as well as the rest of the peripherals. I am getting there and I will post the setup on another post later for comment. I will be looking to see some critical comments on it.

note- There is a case when my 400hp Camaro is faster than your 400 hp Camaro...And that can be determined by year, drivetrain, etc. . ie. 1967 vs 2011. I'll let you all decide the victor. I know the answer...LOL


Thanks [H]
 
My 1999 Z-28 went 11.66 in the quarter at Rockingham and Red River Raceway on nitrous...12.78 on motor alone at Moroso. I had the answer lol.
 
rickss69 said:
My 1999 Z-28 went 11.66 in the quarter at Rockingham and Red River Raceway on nitrous...12.78 on motor alone at Moroso. I had the answer lol.
Click to expand...

Nice, but now you are in a different class than the 400hp being described here.
My 1975 Corvette can run a 10.44 QM. @ 138mph without nitrous...LOL

My new Sandy Bridge i7-2600K will run without nitrous as well, hopefully near 5.0 ghz... LOL (easier to build a car)
 
Not so different class...mine was licensed for the street and I did so on occassion. (NMCA) ;)
 
Well, my Camaro had a 17 second quarter-mile, so hah!

'Course, it was an '85 2.8 V6... But it WAS a Camaro!

And on that note... I see mid-range processors have gotten significantly faster than my Q6600 (even OC'd), but where are the mid-range GPUs to blow the doors off my GTX260 Black Edition? It's almost seeming like I should just try to find another GTX260 and go SLI for bang-to-buck...

Or just upgrade to SB first and let my GPU be the bottleneck.
 
Sorry to get OFF-Topic of the Cars.....

Just stopped down at Microcenter and they have restocked the 2600K......

Axe
 
I wish the review would have talked about why Intel went back to dual channel memory from the earlier triple channel.
 
Mike89 said:
I wish the review would have talked about why Intel went back to dual channel memory from the earlier triple channel.
Click to expand...

Perhaps because triple channel is more costly? Remember this is a mid-range platform, according to Intel's product positioning anyway. The high-end LGA2011 X68 platform will have quad channel, I believe.
 
Mike89 said:
I wish the review would have talked about why Intel went back to dual channel memory from the earlier triple channel.
Click to expand...

They didn't ?
Socket 1156 was dual channel, socket 2011 which will replace 1366 will be quad channel
 
Goodbye brave q6600 @ 3.2. You have served me honorably. I will ensure you get a good home to retire in.

/Salutes


(time to try out an SSD in the new rig also:D)
 
You must log in or register to reply here.
Back
Top