Intel i7-4930K & i7-4820K Ivy Bridge-E Review

[SKYMTL]

720P Gaming: Super Street Fighter IV / Torchlight

720P Gaming Benchmarks (pg.2)

Processors play a huge roll in gameplay performance since they have to process copious amounts of information for the GPU. In the following tests, we use a simple 720P resolution and the lowest possible detail settings in an effort to remove the graphics processor from the equation and place additional pressure on the CPU. 720P was used since it is a resolution that is extensively used by gamers sporting lower end HDTVs and it doesn’t put as much stress upon the GPU as 1080P.

For every one of the following titles, a simple 1 minute gameplay walkthrough was used and the average frames per second was logged via FRAPS.

 

[SKYMTL]

1080P Gaming: Deus Ex: HR / Dirt 3 / Skyrim

1080P Gaming Benchmarks

While lower resolution gaming highlights processor bottlenecks, most people use slightly higher resolution monitors and want to play with increased detail settings. In these situations, the CPU tends to take a back seat to the graphics processor but even at 1080P (ie: 1920x1080) a slower CPU can still have a drastic impact upon in-game performance. In order to illustrate this, we have carried over the games from our previous tests, pumped detail levels to their max and used the increasingly popular 1080P resolution standard.

 

[SKYMTL]

1080P Gaming: Super Street Fighter IV / Torchlight

1080P Gaming Benchmarks (pg.2)

While lower resolution gaming highlights processor bottlenecks, most people use slightly higher resolution monitors and want to play with increased detail settings. In these situations, the CPU tends to take a back seat to the graphics processor but even at 1080P (ie: 1920x1080) a slower CPU can still have a drastic impact upon in-game performance. In order to illustrate this, we have carried over the games from our previous tests, pumped detail levels to their max and used the increasingly popular 1080P resolution standard.

 

[SKYMTL]

1080P Gaming; Current Games

1080P Gaming; Current Games

While our standard in-game testing consists of slightly older but still popular titles, we decided to throw in a number of additional, newer games. Each of these puts a significant amount of load on the CPU but many are also multi-core optimized.

All of the standard testing guidelines apply here as well though there is one major difference: instead of using FRAPS, we are utilizing the FCAT tool. This captures frames in real time before analyzing them, allowing for a much more accurate depiction of what will eventually be displayed onscreen. Once again, a GTX 670 has been used.

 

[SKYMTL]

1080P Gaming; Current Games (pg.2)

1080P Gaming; Current Games (pg.2)

While our standard in-game testing consists of slightly older but still popular titles, we decided to throw in a number of additional, newer games. Each of these puts a significant amount of load on the CPU but many are also multi-core optimized.

All of the standard testing guidelines apply here as well though there is one major difference: instead of using FRAPS, we are utilizing the FCAT tool. This captures frames in real time before analyzing them, allowing for a much more accurate depiction of what will eventually be displayed onscreen. Once again, a GTX 670 has been used.

 

[SKYMTL]

System Power Consumption / Temperature Testing

System Power Consumption

Our power consumption numbers are broken down into two categories: one which simply stresses all of the CPU cores with WPrime and another which puts a high amount of load on both the CPU cores and the IGP. The latter will only be included if a given processor includes a dedicated internal graphics sub-processor.

For the CPU power consumption test, we use the standard testing system (with an NVIDIA GTX 670 installed) and wait until the system and discrete GPU are at idle speeds in order to log the idle power consumption. After this, WPrime 1024M is looped for 15 minutes while the power consumption is logged with a calibrated power meter to determine the peak watts.

Please note that after extensive testing, we have found that simply plugging in a power meter to a wall outlet or UPS will NOT give you accurate power consumption numbers due to slight changes in the input voltage. Thus we use a Tripp-Lite 1800W line conditioner between the 120V outlet and the power meter.

The power consumption of both the 4930K and 4820K met with expectations for processors using Intel’s 22nm technology. They are significantly more efficient than their predecessors while also requiring less power than the 4960X.

Temperature Testing

In order to test temperatures, we set all processors on a loop of Prime95’s Blend Test for 25 minutes and logged the peak temperature core over that period of time using Intel’s XTU 4.2 utility. Ambient temperatures were kept at a constant 24°C. For cooling, a Noctua NH-U14S with two 140mm fans was used.

Naturally, the lower power needs tend to translate directly into reduced heat production. The i7-4820K in particular has plenty of room for overclocking from a thermal perspective, though temperatures on any of these processors tend to shoot up dramatically when additional voltage is applied.

 

[SKYMTL]

Overclocking Results

Overclocking Results

Overclocking Intel’s latest processors hasn’t been an easy task from our end. The Haswell i7-4770K hit a wall at 4.5GHz while even the 4960X –which is supposedly tailor-made for overclockers- edged it out at a mere 4.55GHz. Neither of those represents a particularly good result considering the 4.8GHz to 5GHz speeds we’d seen on previous architectures. Simply put, 22nm processors don’t seem to have all that much headroom….or do they?

As we’ve seen with Ivy Bridge-E, temperatures are kept to a minimum through the use of a revised IHS design which effectively disperses heat over a large surface area. This means the “hot spot” nature of Intel’s 3D transistor design doesn’t become a significant thermal roadblock like it did in mainstream Ivy Bridge and Haswell chips. On paper this should also mean the i7-4930K’s reduction in cache size should give it a bit more overclocking headroom. Consequently, the lower number of cores and a further reduced cache footprint on an i7-4820K should significantly reduce the amount of heat being produced and potentially lead to higher frequencies.

In theory, all of the items we mentioned above lead to better overclocks but they really don’t mean all that much in real life. Due to chip to chip differences, every processor will overclock to a different level. In the case of Intel’s 4960X, according to motherboard vendors, only one in ten samples will hit a frequency above 4.7GHz so anything above that should be considered a spectacular overclock. Among Haswell processors, one in fifteen i7-4770K's were able to hit 4.75GHz. On the other hand, preliminary information seems to suggest those numbers are improved this time around, with the 4930K and 4820K being able to go beyond 4.7GHz 20% and 25% of the time respectively.

With this in mind we set out to find the limits of our chips using both air and water cooling and set 1.35V as our absolute voltage ceiling with 24/7 stability being a prerequisite for a successful overclock. Believe it or not though, our Noctua NH-U14S and Corsair H100i returned identical clock speed results, proving that temperature wasn’t a limiting factor with either processor. Just bear in mind that our results are based on single-source samples and don’t necessarily represent what you will achieve.

​

Starting off with the i7-4820K, we were reasonably surprised and pleased to hit a stable frequency of 4.84GHz on all four cores / eight threads. That represents a great improvement over the 4960X’s rather anemic showing and outpaces our lackluster i7-4770K sample by a good amount. Unfortunately, after hitting this mark, we couldn’t go one iota further while still maintaining stability on air or water cooling.

​

The 4930K provided decent results here as well, though it couldn’t hit the same levels as its little brother. It ended up topping out at 4.65GHz but anything higher and our system simply gave up, refusing to POST until the CMOS was cleared.

All in all, we were satisfied with the overclocking results posted by these two processors. Neither was able to hit the 5GHz mark but they outpaced the i7-4960K and the i7-4820K may have just made a convincing argument in its favor over an i7-4770K.

 

[SKYMTL]

Conclusion

Conclusion

Since the initial reviews were published, Intel’s Ivy Bridge-E has received its fair share of negative press. To many the price was simply too high and the platform lacked convincing features that would differentiate it from what Intel is offering with Haswell. Many of those conclusions were based on reviews of the 4960X and we’ll be the first to say that the i7-4820K and i7-4930K have softened our opinion somewhat.

That 4960X may be targeted at a very specific market which few can afford to buy into but the i7-4930K makes it look particularly overpriced. Intel’s $550, 12-core processor offers literally 96% of its sibling’s performance at a fraction of the cost and can overclock to potentially higher levels. If you are looking to throw a ton of processing power at a multi-threaded application, this is an excellent companion, though still quite expensive.

The i7-4820K provides an interesting counterpoint to the two costly 12-core models and it is actually the CPU we would recommend for someone upgrading from a four core, eight thread i7-900 series processor. It provides excellent out of box performance, particularly in gaming where its high clock speeds allow for framerates on par with the 4930K. It does however struggle to differentiate itself from the i7-3820 and fails to compete on a level footing with the i7-4770K in many instances.

Ever since Ivy Bridge-E’s day one reviews, everyone has been talking about the i7-4820K versus i7-4770K battle. Which one should someone buy? That really depends on how you approach the perspective of value in higher end CPUs. On one hand the 4820K can (at least in our tests) theoretically overclock to a higher level, costs $10 less, is granted huge memory bandwidth from its quad channel layout and uses a platform that includes two full speed x16 PCI-E 3.0 slots. However, the i7-4770K has its own set of strengths like lower power consumption, IPC enhancements that provide a significantly better out-of-box experience and motherboards that feature native support for the latest connectivity technologies rather than using a mish-mash of third party controllers.

With Ivy Bridge-E, it feels like Intel has very much adhered to their old hope that multi-thread processing would become a reality, thereby overcoming the need to boost clock speeds. Unfortunately, that way of thinking is a thing of the past. The multicore revolution didn’t happen nor will it happen anytime soon. Rather, in many respects, highly threaded applications just moved to more capable GPUs, essentially leaving Intel’s Extreme Editions looking flat footed and outclassed by Haswell CPUs that cost less and use a more capable platform. From DirectCompute to OpenCL to CUDA to QuickSync, leveraging a GPU’s resources is simply more efficient than throwing information to and from the x86 cores. Intel has realized this which is why they are pouring resources into their integrated graphics unit. There will always be a market for CPUs with the capability to process 12 threads or more but gamers and other enthusiasts would be much better served with slightly less cores alongside higher clock speeds.

This situation leaves us with a simple question: are the i7-4930K and i7-4820K really worth buying right now? We’d like to think so. The i7-4930K is an excellent all-round processor that won’t cause buyer’s remorse like the 4960X likely will. Plus, there’s always going to be something intrinsically compelling about having one of the best processors money can buy without spending a thousand bucks.

The i7-4820K on the other hand is the star of Intel’s Ivy Bridge lineup for budget-focused users. Most examples should be able to overclock extremely well and it can easily keep up with the i7-4930K in games. So, while it may be tied at the hip to a slightly inferior platform from a connectivity standpoint and looses to an i7-4770K in a bone stock comparison, we’d recommend the i7-4820K for gamers who want to build a new system around a solid, adaptable and overclockable core. The only problem here is making the jump to this relatively inexpensive IVB-E chip means participating in a game of Russian roulette: if you don’t get a decent overclocker, the i7-4770K will ultimately be a much better value. The question is whether or not you're willing to bet $310 on a that outcome.