Fastest Single Core Performance?

[Shaggy]

At work I'm doing a lot of engineering simulations. The problem is these programs only utilize a single thread. Taking that into account would an upgrade to an i5 really be the best option?

I'm looking for the fastest single core performance. Given these are work PC's OC'n really isn't an option.

[Lu(ky]

Sandy 2500k on air should reach 4.5GHz no problems on water maybe 5GHz depends on CPU.. Even though this is a QUAD CORE 4 cores it is really really fast not to mention very easy to overclock as well. Also cores can be turned off to only using 1,2,3 or all 4 etc.. Go to use a i3-530 as well sells for 100 bucks OC ok as well.

[Shaggy]

Like I said, I'm not overclocking a work computer.

[zalbard]

At work I'm doing a lot of engineering simulations. The problem is these programs only utilize a single thread. Taking that into account would an upgrade to an i5 really be the best option?
Yep, it would be.

[RacingTurtle]

Anything new will be much faster than older generations even in single core performance.
i7, i5, Sandy Bridge etc.

[[XC] Synthetickiller]

Since you're not OCing and apparently not using more than 4 cores (right, not running multiple simulations?), i5 2500 (non-k) is the answer.

[Shaggy]

Each sim uses only 1 thread. Each sim also is backed by a $100,000 license. We have 3 licenses.

So we load up the que with a ton of sims to be performed and it knocks them out 3 at a time using a total of three threads.

Two computers are running Wolfdales while a third is running a Deneb.

[lkiller123]

Like I said, I'm not overclocking a work computer.

Even if you won't overclock it, the stock Turbo Boost can reach up to 3.7GHz on the 2500.

[Johnny87au]

Seeming as Sandybridge are getting recalled due to a design flaw i'd grab a i5 or i7.. Anywhere from a 870 to a 930 would do you quite well with HT!

[Filter]

wow seriously 100k per thread.

[endlesszeal]

Synthetickiller;4727942']Since you're not OCing and apparently not using more than 4 cores (right, not running multiple simulations?), i5 2500 (non-k) is the answer.

+1. i5-2500 because 2600 has HT which youre not using, $100 more than 2500 and has only a 100mhz advantage.

In these synthetic single-threaded bench marks you can see how much faster the 2500 is over the i7980x (the second fastest single-core processor).

20% faster on Cinebench 10 Single Thread:
http://www.anandtech.com/show/4083/the-sandy-bridge-review-intel-core-i7-2600k-i5-2500k-core-i3-2100-tested/17

13% faster on SuperPI 32M:
http://www.hardwarecanucks.com/forum/hardware-canucks-reviews/39555-intel-sandy-bridge-core-i5-2500k-core-i7-2600k-processors-review-9.html

And i disagree JohnnyAU, the defect only affects the motherboard's SATA 3.0 ports, not the cpu itself. And many companies are giving replacements, so it shouldnt deter you from buying sandy bridge. The performance advantage (over 20% depending on program) outweighs the con.

[TJ TRICHEESE]

No it affects sata 2 ports and not sata 3 ports.

[OldChap]

Will the terms of your software license allow you to run more than one instance on a single machine .on a VM?

[Johnny87au]

Will the terms of your software license allow you to run more than one instance on a single machine .on a VM?

He could always re image and do it that way without losing software licensiing..

[[XC] Synthetickiller]

Depending on the set up, I still say go i5 2500. Wait for new boards to ship.

The jump in single thread performance is worth it. IF you do plan on trying to run 6 at a time, the 2600 is an option. If time matters the most, just build a second or 3rd i5 since real cores are much faster than HT.

My i5 760 @ 4ghz still gets killed by sandybridge.

[Johnny87au]

Synthetickiller;4732254']Depending on the set up, I still say go i5 2500. Wait for new boards to ship.

The jump in single thread performance is worth it. IF you do plan on trying to run 6 at a time, the 2600 is an option. If time matters the most, just build a second or 3rd i5 since real cores are much faster than HT.

My i5 760 @ 4ghz still gets killed by sandybridge.

I wouldnt say killed dude, in benchmarks YES by alot, In gaming hardly a difference imo..

[445533]

I wouldnt say killed dude, in benchmarks YES by alot, In gaming hardly a difference imo..

Clearly the OP is not using it for gaming so who gives a :banana::banana::banana::banana:?

[Johnny87au]

Clearly the OP is not using it for gaming so who gives a :banana::banana::banana::banana:?

And you are? purpose of your post?

[BrightCandle]

Whether a Sim will see an advantage its hard to say. Do you know what is limiting the performance of the Simulations today? Is it floating point, memory bandwidth or integer math limited? How big in total memory set is a simulation and does it benefit greatly from cache performance?

[Thrmionicshaman]

Whether a Sim will see an advantage its hard to say. Do you know what is limiting the performance of the Simulations today? Is it floating point, memory bandwidth or integer math limited? How big in total memory set is a simulation and does it benefit greatly from cache performance?

+1 sir I concur:yepp: