[citation][nom]mayankleoboy1[/nom]Nice review!1. Are there plans to release any K CPU's without the HD4000? will they OC higher?2. Any chance of intel releasing a second stepping of K-series IB chips?[/citation]
That would not affect how well they overclock and Intel releases K edition CPUs with the top IGP. The only difference between an 5 with HD 2500, an i5 with HD 4000, and an i7 with HD 4000 is binning. That hardware for HD 4000 and HTT is all there, just disabled because either Intel wanted to sell the chip as a cut-down version to address an under-supply of i5s, or because the chip had a damaged part involved with the IGP that forced Intel to activate it only as HD 2500 instead of 4000. That means that the chip is an inferior chip that would not be sold as a top end overclocking chip, so there is no chance of Intel selling K edition CPUs with HD 2500.
[citation][nom]vilenjan[/nom]Good old Intel. No competition and look what happens, the new generation is barely an upgrade over the previous. Anyone remember Intel PII 450s and the PIII 450s?[/citation]
Everyone knew from the start that Ivy Bridge was not a CPU performance leap over Sandy Bridge. If not for Intel using the crap paste instead of solder or at least better paste, Ivy Bridge would have beaten Sandy Bridge at least in overclocking performance in addition to it's efficiency win. Haswell is the performance leap over Sandy Bridge. Intel has been using the tick-tock strategy for years now and we've all known that Ivy wasn't a big leap, so you're not only a troll, but a fail troll at that.
[citation][nom]ringsTrue[/nom]Without the baseline clock for clock comparison (4.5 sandy vs 4.5 ivy), i'm afraid these results are pretty much useless. It's like doing an uncontrolled experiment and passing it off as real science.[/citation]
That was already done in a previous article and we all know that Ivy has more performance per Hz.
[citation][nom]digiex[/nom]as the manufacturing process gets smaller = Smaller die size, supposed to be cooler temperature,but, with small die size = small area for heat dissipation,...an irony that needs to be solved.[/citation]
Not using such crap paste would have solved that. Leaving it with the solder, Ivy Bridge would have trumped Sandy Bridge greatly in overclocking and even with the higher quality paste that the Japanese site used makes it better than Sandy Bridge.
[citation][nom]A Bad Day[/nom]One possible counter to the rapid thermal ramp up is to remove the die cover and get rid of the thermal paste bottleneck.The hard part is getting the heatsink to fit, and not crush the die chip into silicon sand...[/citation]
If you're willing to cut off the IHS, then you're willing to put in higher quality paste like the Japanese site did which was shown to improve thermals dramatically.
[citation][nom]spookyman[/nom]So I guess Haswell will just 2-5% faster then Ivy Bridge.Sounds like an improvement to me.[/citation]
Haswell will be more like 20% to 35% faster than Ivy Bridge. You trolls just don't learn, do you?
[citation][nom]BVKnight[/nom]Why is the TH test throttling your chip at high 90s-100 C if the TJMax for the Ivy Bridge chips is 105? I thought that the chip wouldn't throttle itself until it hit that limit.[/citation]
They probably didn't use a physical temperature gauge, just a software one. Those aren't as accurate, but using a real gauge can damage the processor (look at Hard OCP's CPU cooler reviews to see how).
http://www.hardocp.com/article/2011/11/17/thermalright_true_spirit_cpu_air_cooler_review/
[citation][nom]jaquith[/nom]Very nice Article! IF it's entirely true of 20% cooling efficiency loss seems like Intel made a very poor decision on shifting to Paste from Fluxless Solder. I'd need to see that Japanese test substantiated and repeated by a few others before converting myth to fact.I always get a kick out of efficiency and consumer computing power. In my case, at $0.12/kWh more cores win and shorter processing time is a no brainer - compare an hourly wage of $30~$50/hr+. The efficiency is dramatically more important in mobile computing. Any of the OC's listed are going to throwing a ton of heat off, and unless you have a very cool room with AC and you can tolerate the subsequent noise -- you need to fully evaluate high OCs. Even with the OCs listed I don't recommend e.g. an H100 unless you're deaf, and instead with those listed OCs I'd look into a better cooling solution i.e. EK, Koolance, etc with a ~1000W radiator with decent fans.[/citation]
20% gain from Intel's paste to higher quality paste. Solder would be even better.
Facebook
Twitter
Instagram