Techgage logo

Intel’s 32nm Clarkdale – Nehalem for Everyone

Date: January 3, 2010
Author(s): Rob Williams

To help kick 2010 off right, Intel has filled out the rest of its current-gen processor line-up with the help of Westmere. We’re taking a look at the desktop variant here, which brings a lot to the table compared to the previous generation. For those who’ve been holding out for that next affordable PC upgrade, the wait has been worth it.



Introduction

This past September, Intel released a follow-up to its successful Nehalem architecture, called Lynnfield. Throughout its launch campaign, Intel touted Lynnfield as being “Nehalem for the mainstream”, hence the title of our launch article. That title was appropriate, because Nehalem, for the first year of its life, was the only choice for those who wanted in on the company’s latest architecture, but thanks to the $300+ price tag, many held off and waited for more budget-conscious models to come along.

Lynnfield improved the situation, because its lowest-priced model, the Core i5-750, came in at around $200. Aside from its lack of HyperThreading, that particular model had everything that made Nehalem so great. In fact, it even improved on certain aspects, such as with its more effective Turbo feature. In the end, the entire Lynnfield line-up proved to be packed full of winners, as we discovered in our launch article.

However, even at $200 for the lowest-priced chip, there are many who are missing out on what Nehalem and its sub-microarchitectures bring to the table, and that’s where Westmere comes into play, with its dual-core models set to sell for between $100 – $284. At the low-end, the prices finally allow virtually anyone to take advantage of Intel’s latest architecture, and with a few extra benefits tossed in.

At its heart, Westmere is based on Lynnfield, which was derived from Nehalem. That means that these Westmere chips share the equally-impressive Turbo multipliers of Lynnfield, along with the removal of the Northbridge. Compared to Core 2 processors, Westmere has other benefits as well, such as HyperThreading, the AES-NI instruction set, and of course, improved power efficiency. Oh, and not to mention integrated graphics for its dual-core models.

Introducing Intel’s Westmere Microarchitecture

Before we dive deep into Westmere, let’s take a moment to clear up all of the codenames that are floating around. Westmere, first and foremost, is the latest microarchitecture out of Intel, and is not the name of the processors which use it. Rather, those names are Clarkdale (Desktop) and Arrandale (Mobile). The former isn’t to be confused with Clarksfield, which is Intel’s mobile Nehalem (45nm). On the chipset side, all LGA1156 models thus far (P55, H55, et cetera) are classified as Ibex Peak. Future six-core Westmere’s (which will not include an IGP) use the Gulftown codename. Finally, Westmere’s IGP is codenamed Ironlake.

To help confuse things further, Westmere is ultimately based on Lynnfield, which was derived from Nehalem. That means that the same Nehalem features that were shaved off for Lynnfield are also gone from Westmere. This is putting it all simply, however, as Westmere features a few things Lynnfield processors do not, such as the AES-NI instruction set. We’ll tackle this and more throughout the article.

One of the biggest and most notable features of Westmere is that it’s built on Intel’s 32nm process, which makes it the first of its kind in a desktop processor. Intel was also first with 45nm, which we saw in the form of Penryn just a little over two years ago (oh, how the time flies). If Intel successfully sticks to its tick/tock cadence, we should be seeing a 22nm launch at some point during 2011.

Thanks in part to this die shrink, Intel was able to do some interesting things, including adding in new features, and also adding a secondary chip onto the same processor, which in this case is an IGP (integrated graphics processor). Because both the CPU and GPU are separate, this isn’t a hybrid design, and given that the GPU is built on a 45nm process, it simply couldn’t be. But, what makes this design all the more interesting is that we’re one step closer to that hybrid design, which will undoubtedly occur in the future.

Because the IGP is located on the processor, and not a Northbridge or Southbridge, the chipset should run much cooler. At the same time, we’re able to cool both the CPU and GPU (the single chip) with a single cooler, rather than have one for the CPU, and a smaller, and whinier one, for the Northbridge/Southbridge. That in itself is a huge advantage to the new chip design, and should make HTPC building all the more easier.

The stock image above, showing off Clarkdale’s exploded view, makes it easy to understand how the two processors are laid onto this single chip, with the larger of the two being the GPU. With the help of the labeled die shots below, we’re able to see each chip in greater detail:

This to-scale image details a lot, and it’s easy to understand just how much larger the GPU is than the CPU. There’s a good reason for that, and it’s not 100% tied to the fact that the GPU is built on a larger process. Rather, you’ll notice that the CPU, unlike Lynnfield and Nehalem, does not have an integrated memory controller. Instead, that’s built into the GPU. With this design, it means that the GPU is going to be inherently larger to begin with, and the CPU smaller than it would have had been with the IMC.

There are a couple of reasons Intel might have taken this route, but the most likely one would be for the sake of ease. Since the GPU here isn’t hugely different from the one used in the G45 chipset, Intel decided to carry its design over to Clarkdale. Since previous IGP designs had the IMC built into the GPU, it seems Intel decided that sticking to that design would be easiest, and that’s the reason Westmere doesn’t have an IMC built into the CPU. Future Westmere chips, like Gulftown, will not take this same route, as it won’t have an IGP.

Should this be a concern? It depends, but for the most part, no. Because of this design, the memory latency is higher than it is on other current-gen processors, by around 25 – 30%, according to our Sandra tests. The reason that this is of little concern to most people is that applications typically aren’t going to respond much faster to tighter latencies (a fact I’ve exhaustively explored in the past). I’m sure Intel took its time in deciding on this design, and in the end, it found that the higher latencies simply weren’t an issue. For enthusiasts, it could be, but for all intents and purposes, this is a mainstream processor, not an enthusiast one.

We’ll talk a bit more about the Westmere architecture and its integrated graphics chip more on the following pages, but before we do that, let’s review Intel’s current processor line-up:

Intel CPU
Cores
Threads
Clock
Turbo
Cache
GPU
TDP
1Ku Price
Core i7-975 (1)
4
8
3.33GHz
3.60GHz
8MB
N/A
130W
$999
Core i7-960 (1)
4
8
3.20GHz
3.46GHz
8MB
N/A
130W
$562
Core i7-950 (1)
4
8
3.06GHz
3.33GHz
8MB
N/A
130W
$562
Core i7-870 (2)
4
8
2.93GHz
3.60GHz
8MB
N/A
95W
$562
Core i7-860 (2)
4
8
2.80GHz
3.46GHz
8MB
N/A
95W
$284
Core i7-920 (1)
4
8
2.66GHz
2.93GHz
8MB
N/A
130W
$284
Core i5-670 (3)
2
4
3.46GHz
3.73GHz
4MB
733MHz
73W
$284
Core i5-750 (2)
4
4
2.66GHz
3.20GHz
8MB
N/A
95W
$196
Core i5-661 (3)
2
4
3.33GHz
3.60GHz
4MB
900MHz
87W
$196
Core i5-660 (3)
2
4
3.33GHz
3.60GHz
4MB
733MHz
73W
$196
Core i5-650 (3)
2
4
3.20GHz
3.46GHz
4MB
733MHz
73W
$176
Core i3-540 (3)
2
4
3.06GHz
N/A
4MB
733MHz
73W
$133
Core i3-530 (3)
2
4
2.93GHz
N/A
4MB
733MHz
73W
$113
Pentium G6950 (3)
2
2
2.80GHz
N/A
3MB
533MHz
73W
~$100
Microarchitecture: (1) Nehalem, (2) Lynnfield, (3) Clarkdale

With the introduction of Westmere, I think it’s safe to say that we’ll be seeing the Core 2 line-up phased out very shortly, and with these new models and their pricing, there’s little reason to run out and purchase a Core 2 in lieu of one, even if you don’t need the integrated graphics. For this launch, Intel will be releasing a total of four Core i5’s, and two Core i3’s, along with their respective chipsets: H55 (mainstream), H57 (upper mainstream) and Q57 (business).

Of all the processors, which are spread across three microarchitectures, six models belong to the Core i7 family, five to Core i5 and two to Core i3. There’s also the lone Pentium G6950, which will surely be followed-up with other models in the year to come. Intel hasn’t disclosed pricing on this chip, so I put an estimate on it at around ~$100. Like Core i3, Pentium models will not feature a Turbo multiplier, and like the Core i5-750, it also lacks HyperThreading.

Westmere’s launch is one worth celebrating, because finally, everyone will have access to Intel’s robust Nehalem-based architecture. We’re not into the sub-$100 realm yet, and won’t be for some time, but with prices ranging from $100 – $999, there’s certainly a processor model for most anyone looking to build a new PC, regardless of what your needs are.

Westmere’s Goals & Industry Trends

Westmere’s launch might seem tame overall, but in reality, we’re being brought into a rather significant shift of focus in the computer industry. For as long as there will be PCs, there will be enthusiast PCs and parts, but with Westmere, and particularly Clarkdale and Arrandale, we’re seeing an increase of integration, and also a focus on smaller PCs.

At Techgage, we try our best to cater our content to all segments, although I admit that we’ve not touched on the integrated, or low-end, segments much at all. With Clarkdale, we were forced into it, as the processor has an integrated GPU. But, this is for a reason, because it’s the beginning of things to come. Although the GPU is simply placed next to the CPU at the moment, we can expect a total fusion design in the future, from both AMD and Intel. Integration is king, and there’s no sign of that changing.

In a press deck Intel developed for its Clarkdale press kits, there are a couple slides worth looking at, because they showcase the important trends to take note of in the industry. As you can see in the foil below, both the minitower (mid-tower) and tower (full-tower) numbers are gradually sliding, whereas all-in-one PCs, and SFF/USFF are growing quite significantly.

As a PC enthusiast, this can be considered a disheartening trend, but I’m certain that there will always be a market for enthusiast components and PCs. The real reason we’re seeing such an increase in small PCs and decline in larger PCs is part in thanks to the fact that more people than ever are computing, and likewise, more people than ever are building secondary PCs, many of which are small in form-factor, and designed for home server or HTPC use.

We’re not covering Westmere’s mobile part, Arrandale, today, but I still wanted to point our trends seen in the notebook scene, because they’re equally as interesting. In the foil below, there are a couple interesting happenings. It’s no surprise that the sub-12″ market has seen stark growth, and will continue to see that going forward, but it is somewhat surprising to see the 15″ notebooks decreasing in popularity, while 16″+ notebooks are growing, albeit at a much slower rate than the others.

It can be assumed that in the 16″+ space, consumers are beginning to prefer notebooks to desktops, and as a result, they’re using these notebooks as DTR’s (desktop replacements). It’s also clear that 15″ is a dying form-factor, as 14″ is creeping up to almost match its sales in 2013.

I won’t include slides for everything Intel points out, but I’d like to touch on a couple of trends that I found interesting, although I admit not many have much to do with Clarkdale in particular. From an Internet users perspective, at the end of September, there were 1.73 billion users online. Compared to the 361M at the end of 2000, that’s some stark growth… 380% in a mere nine years. Adoption doesn’t seem to be slowing down, either, which is important, since the Internet is increasingly becoming much more than just a luxury.

In a similar vein, Intel’s PDF also points out the insane growth in online video, and I’m sure it will come as a surprise to no one that this is one area that’s simply exploded. According to Pew Internet & American Life Project, a staggering ~90% of people online between the ages of 18-29 are using the net to watch video. That number drops a bit to ~70% for 30-49, and continues to decline at 50+.

As a result of this increased video viewing online, the number of people creating their own content to publish online has also increased. In 2008, that number hovered around 80 million for the US, while by 2013, it’s estimated that it will be closer to 115 million. Seems like a small boost, but realize that this is US-specific, and that 35 million boost is just over 10% of the population.

What does Westmere have to do with any of this? Faster media performance, of course. Because online video and creation is more popular than ever, Intel is pushing these numbers to show just how beneficial its new and improved processors are. In the company’s internal surveys, it found that 27% of computer users edit photos a couple of times a week, while 39% play casual games. Westmere, and more specifically Clarkdale and Arrandale, enable consumers to do all of this, faster, and easier.

More than ever, we’re seeing a push from Intel to let people know that upgrading that old PC might be a good idea. According to the PC Gaming Alliance, 250 million people worldwide play casual games on a regular basis, and with Westmere’s IGP, consumers will be able to load up and play those same games without a discrete GPU. See? The goals of Westmere, and its targets, are becoming increasingly clear.

In another interesting foil, preliminary forecasts show that between 2010 – 2011, about 475 million older PCs will be “retired” in order to make room for faster machines. By retired, it means that they will not be the main PC, not necessarily that they will be thrown into the trash. As more and more people begin using the PC for more than just e-mail, the demand for faster performance is higher than ever.

To give a real-world example of performance increases, Intel notes that while on a 2006 notebook, using the Core Duo T2250, creating an HD vacation video took about 3 hours and 2 minutes. On today’s Core i5-430M processor, the same process takes just 49 minutes. These are the numbers that people want to see, because even if you’re not a PC enthusiast, shaving up to 4x the time off of a project is going to be enticing.

Mentioned above was the fact that Westmere and its IGP were designed to cater to casual gamers, so it’s appropriate that we tackle that, and more, on the next page.

Clarkdale’s Integrated GPU

For Westmere’s launch, Intel is putting emphasis on the fact that the processor is equipped with an integrated GPU, and probably for good reason. The company is the first to do this in a typical desktop processor, and although we’re not being given a fusion design, what we have here is still very noteworthy. But ultimately, what benefit does this design offer the consumer? There’s one big one.

With Lynnfield’s launch a couple of months ago, we saw the first consumer motherboards that lacked a Northbridge. This design was able to happen because the integrated memory controller (IMC) was moved onto the CPU. With a shifting of other components, the Northbridge was obliterated, and we were left with the CPU and Southbridge. In previous architectures that included graphics, the IGP+IMC was its own chip, the Northbridge. With that moved straight onto the processor, we’re also able to rid the typical Northbridge for Westmere, like we saw happen with P55.

The benefit is that instead of having two chipsets on a motherboard along with the CPU, we now have a single chipset and the CPU. This allows for greater flexibility from motherboard vendors, since they won’t have to make room for a Northbridge and all of its traces. And because the GPU is much closer to the CPU, the memory bandwidth and latency may improve also. I’d be willing to bet that the differences would be minimal, however, and not noticeable in the real-world. If the CPU and GPU were fused together, that would change.

In past generations, Intel applied simple names to its GPUs, and with Westmere, that doesn’t change. Throughout all of Intel’s marketing materials, I’ve actually had a had time figuring out what the IGP was called since it’s been called something different all over. The term I’ve seen most often is “Intel HD Graphics”, so I assume that’s the final and most appropriate name. Although the GPU will be clocked different on some CPUs, the name of the GPU itself doesn’t change. All of the IGPs in Clarkdale / Arrandale have equal feature-sets.

High-Definition / Blu-ray Support

As I touched on elsewhere in this article up to this point, Intel is pushing its IGP in Westmere as being best-suited for the casual consumer. That means that all of the graphical features of Windows 7 will work just fine, and likewise, the same can be said for casual games. But, there’s another important feature that can’t be ignored, given it’s growing popularity: HD Playback. As you’d expect, Intel’s newest IGP touts full Blu-ray support, and even includes dual decode, for those movies that include a Picture-in-Picture option.

When G45 launched last summer, it wasn’t that well-received, due to it’s Blu-ray playback issues. It took a couple of months to see those ironed out, but they were, and things have improved ever since. The IGP found in Westmere isn’t a simple copy/paste of G45, but rather includes some upgrades worth noting. To help me explain those upgraded or new features, we can refer to another slide from Intel:

From an acceleration standpoint, the new feature here is dual decode, which means that Blu-ray movies (or other HD content) that features Picture-in-Picture content (such as commentary) will playback just fine. For post-processing, we’re given sharpness acceleration, and also the ability to take advantage of the xvYCC extended color space. Display support has also improved, with the addition of dual HDMI, and also increased color depth, at 12-bit, for both HDMI and DisplayPort. Finally, audio is also improved, with the addition of audio through DisplayPort and also native support for Dolby True HD and DTS HD.

For those looking to build either a regular or home theater PC with Westmere, there’s really no important features missing, and unless your a devout home theater enthusiast, chances are that all of the features here will suit you just fine. It should be noted that Intel works closely with developers of Blu-ray playback software, which include Cyberlink, Corel and Arcsoft. All three of these companies develop software that can take good advantage of Westmere’s IGP.

Currently, none of these offer acceleration for DVD upscaling with the help of the IGP, but the feature will work just fine regardless. In the case of Westmere, the CPU will simply be relied on much more than the GPU, as you’d expect. If there’s one fact to take away from all this information so far, it’s that Westmere, in the form of Clarkdale and Arrandale, can handle Blu-ray and other HD playback just fine, while supporting leading AV technologies many enthusiasts want to see.

Tech Specs & Gaming

Once again, I’ll use one of Intel’s slides in order to help me explain all of what’s new with Intel’s latest IGP from a technical standpoint. It should be stressed that this IGP is designed for casual gaming, not enthusiast gaming. Don’t expect to pick one up to play games at higher than 1280×1024, because that won’t be the case anytime soon. In all of the exhaustive surveys done, though, it’s clear that the market-leading genre for games is casual, and as a result, Intel is stressing that its latest IGP handles the vast majority of those games just fine.

By “casual”, I mean games like Sims 3, Bejeweled, Nancy Drew, anything from companies like PopCap and so forth. The GPU doesn’t stop there, because it can also handle games like World of Warcraft, but that’s about as far as you’ll go. I’ll touch a lot more on the gaming capabilities of this upgraded IGP later in the article, but for now, let’s take a brief look at why this IGP is better than last-gen:

At the forefront, the latest IGP uses Intel’s 3rd generation Unified Shader Architecture (I wonder why they don’t abbreviate this??) and also bumps up its execution units from 10 to 12. Its graphical capabilities have also been increased with improved vertex processing, hierarchical Z and fast Z clear rendering and of course, support for Windows 7. The core clock has also been boosted up to 900MHz, and coupled with the bump of execution units, it’s safe to say that this IGP is faster than what we saw with G45.

One thing worth pointing out is that while Intel boasts 2560×1600 resolution support, that’s only if you’re using a DisplayPort connection. For DVI (due to its single-link connection), you’ll be limited to 2048×1536, while for HDMI, the limit is 1920×1200. Intel fully believes that DisplayPort is the future, and given its growing popularity in consumer monitors, that seems to be true, so no argument there.

The lacking resolution for DVI and HDMI might be a problem for some, but I think it’s safe to say that the vast majority will not be running anything higher than 1920×1200, so from that standpoint, Westmere’s IGP will suit most people just fine. If you do need 2560×1600, then chances are you have a newer display with DisplayPort built-in. Either way, the IGP here will handle at least 1920×1200 on all connections it offers, and of course, 1080p for all your media needs.

Revamped Graphics Control Panel

With Westmere’s IGP being a big focus for this launch, Intel wanted to help ring in the celebration right with a completely revamped control panel, which looks much, much better than what we’re used to seeing. I first got a glimpse at this back at IDF during a meeting, and I have to say, a UI really doesn’t matter at the end of the day as long as it works, but this one looks very nice. It might help that I have an affinity towards the color blue, though…

Intel's Graphics and Media Control Panel

The control panel doesn’t only look better, but it offers much better overall tweaking ability as well, from being able to set your custom resolutions to having more robust graphics options, with the ability to tweak texture detail, AF and vertex processing. Another feature I like is the ability to quickly and efficiently change the rotation of your display. With a simple three-key combo, you can turn your screen to portrait mode, and even upside down. The best thing about is that it does its action fast… much faster than I’d expect.

Overall, the revamped control panel packs in more features, and looks better while doing so. For the consumer software side, this is one of the best releases from Intel, even if it’s only based on the looks alone!

Intel’s H55, H57 & Q57 Chipsets

With all we’ve discussed so far, it would make little sense to skip over mention of the chipsets themselves, so let’s take a moment here to look at them a bit deeper. As of today, Intel will be launching three new chipset models: H55, H57 and Q57. It’s a little important to note the differences between these, because it may affect your decision-making process when looking for a board.

The trio of chipsets here can be best compared to P55, as there’s no Northbridge, and everything is divided up between either the processor or the Southbridge. Since the graphics are moved to the CPU, the Southbridge’s job is to handle the PCI-E lanes, the various display support, S-ATA and eSATA, USB, LAN and also Intel’s Management Engine, which handles all of the on-chip components in the OS.

One of the more interesting things about the H55 chipset, is that up to now, the majority of motherboards I’ve seen to use it have been mATX, so we’re truly entering a time where small PCs are pushed more than ever. I might be an enthusiast at heart, but it’s easy to understand this direction. Most people don’t need an ATX motherboard, or three PCI-E slots, or 10 S-ATA. Rather, they need a single PCI-E for graphics, and another slot for some other peripheral, whether it be WiFi or audio.

Intel's Q57 Chipset

So with that all in mind, the most popular chipset of these three will surely be the H55, as it’s slightly less expensive ($40 vs. $43), and lacks nothing truly important. But… there is one notable difference between the H55 and H57 chipsets that may sway your decision… AHCI. H55 doesn’t support it, but H57 does. Why Intel took this route, I have no idea, but it’s a bit inconvenient if you ask me.

The situation around this right now is a little murky, and in talking to board vendors, I’m only left even more confused. One company in particular told me that they were looking into supporting AHCI on its H55 boards, but didn’t know if it was going to be possible. The person I talked to led me to believe that it could be possible to write a specific driver to enable the support, or to just go with another chipset. But at that point, it would make much more sense to just opt for H57, given the minor price difference.

AHCI isn’t the only other benefit H57 brings to the table. It also increases the number of available PCI-E lanes and number of USB ports (12 to 14). I sure most would agree that neither of these two increases will affect much people, so if the lack of AHCI doesn’t bother you, then the H55 chipset is going to suit you just fine.

The third chipset is Q57, which is essentially an H57 with Intel’s Active Management Technology added in. This feature allows IT environments extra features for efficient handling of a large number of computers, by being able to diagnose a machine remotely, isolate it from all other computers (except the one you are accessing it with) and much more. This is an IT-type feature, and one that would never be used in the home (unless you own a serious mansion, maybe), so for the end-consumer, the choice would between H55 and H57.

Intel’s DH55TC Motherboard

When Intel launches a processor built on a new architecture that requires an also-new socket, it will send along its mainstream board in order for us to get our testing done. In some ways, this could be considered a “reference” board, but unlike reference cards in the graphics world, Intel actually sells all of the boards it creates, so there’s no reference, or prototypes here.

Though Intel will have more than one H55 board at launch, the one sent to us is the ~$100 DH55TC, an mATX offering with a focus on media. It offers all we’d expect a Clarkdale board to have, and lacks little that might be of some use. It’s interesting to note that this board does not offer an IDE or floppy connector, which is something I’m sure will bother few.

Like most other Intel boards, this one is designed for a mainstream audience, and not for overclocking. You can tell this by the lack of power phases, and also thanks to the fact that it only uses a 4-pin motherboard connector. The DH55TC is instead designed for those who want a non-complicated board that offers all of the features they need. In that regard, this board delivers, and at a reasonable price-point.

If you’ve seen Intel boards before, the DH55TC will come as no surprise to you. It’s mATX, it features many of the same colors as other Intel boards, and it’s pretty simple in design, with no emphasis on bling (this isn’t entirely a bad thing).

Since Clarkdale, like most other current CPU architectures out there, utilizes a dual-channel memory controller, we’re given a 4x DIMM configuration, with support for up to 16GB of DDR3-1333.

For storage connectivity, Intel provides six S-ATA ports for use with hard drives or optical storage. Note that the board does not include an IDE or floppy connector.

For discrete graphics, the board includes a single PCI-E 16x slot, along with two PCI-E 1x and a single legacy PCI slot. To the right of these slots is the Southbridge, in all its passive glory.

Around the CPU socket, we can see there’s plenty of room for even the beefiest of coolers. We can also see some old-school voltage regulators, which aren’t all too common on mainstream boards today. As mentioned before, this board requires a 4-pin motherboard connector, not an 8-pin. If you have a PSU with an 8-pin cable that cannot be split into two parts, you can still properly plug in the appropriate four.

Looking at the back I/O port, we see six USB 2.0, a LAN, PS/2 Keyboard/Mouse and audio connectors (this is the first time in a while I’ve seen three connectors on any board, in lieu of the usual six). Aside from all this, for display purposes there’s HDMI, VGA and DVI. It’s interesting that despite DisplayPort being required for any resolution higher than 2048×1536, there’s no such connector here.

Intel’s boards are generally always simple compared to the competition, and to many, that’s part of the appeal. Intel focuses highly on stability, not overclocking, so when you pick up one of its boards, you know you’re going to be getting a reliable product. As long as you don’t try to overclock it, of course. But, that’s going to be hard to do anyway, given some of the limitations in the BIOS, as we’ll see next.

Intel’s DH55TC – BIOS

I have mentioned many times in the past, most recently in our DP55WG review, that I’m not entirely a fan of Intel’s BIOSes. All of the issues I have with them have to do with clunkiness, which rears its ugly head in various ways. For one, switching from one top menu to another actually redraws the screen. It’s not a quick transition like on most other boards. Second, the overclocking options are totally lacking, and the DH55TC takes that to a new level.

I won’t explain everything about the BIOS here, because the pictures do a good job of making less work for me in that regard. All of the importnat options are here, but interestingly enough, the overclocking options are highly limited. You can adjust the BCLK and CPU voltage, but not much else. Want to tighten your memory timings? You can’t. No joke. With my DDR3-1333 kit, I was forced to use 9-9-9 timings, rather than the 7-7-7 timings the kit is capable of.

It’s for this reason that I opted to perform all testing not on this board, but rather ASUS’ P7H55D-M EVO, which I take a look at on the following page. That board is much more open when it comes to flexibility, so it made more sense to use it for testing. Chances are also good that it will be a more popular choice than the DH55TC, for reasons I discussed here.

ASUS’ P7H55D-M EVO

When I received a package from ASUS and discovered the P7H55D-M EVO inside, I was stoked, and couldn’t wait to get it installed. I have little against Intel’s board for simple use, but I knew ASUS’ offering would blow the doors open for greater customization, features and of course, overclocking. While Intel’s board offers very little in terms of overclocking, the P7H55D-M EVO is all about it.

At the time of writing, I am unsure of the retail price of this board, but I wouldn’t be surprised to see it at around $130, or even a bit higher, given its feature-set. It’s a robust board in all regards, from its connectivity options, to its layout and feature-set. It aims to be the ultimate board for either an HTPC user, gaming enthusiast or simply a regular desktop user who wants the most from their PC.

By looking that the board, you can see that ASUS has loftier goals than Intel. ASUS supplies a an 8+3 power phase solution, with the 3 being dedicated to the GPU/IMC. Unlike Intel’s board, ASUS’ requires an 8-pin power connector, proving that the board is built for overclocking, and as we’ll see later, it does a great job in the regard.

If there’s just one thing I notice right off about this board that I’d change, it’s that it only has three on-board fan connectors. Given this is an mATX board, that’s not a huge problem given that it’s destined to be used in a smaller chassis, but with one for a memory cooler, and another for the CPU, that only leaves one for the chassis. Fortunately, many chassis fans today include molex power connectors that plug into the PSU cables, so this still shouldn’t be much of a problem for many people.

Like Intel’s and all other Westmere-based boards to be released, the P7H55D-M EVO includes 4x DIMM slots with support for up to 16GB of RAM at DDR3-1333 speeds. In the picture below, you can see two of the phases that are dedicated to the memory. Hidden right above the 24-pin motherboard connector is the MemOK! button, which when pushed, allows the PC to boot up with memory that may be riddled with bad SPD information, or is designed to run at much faster speeds than the CPU or motherboard could handle at a given voltage.

ASUS includes a total of six S-ATA connectors on this board, all of which are 3Gbit/s, not 6Gbit/s. ASUS will offer other H55 boards that offer S-ATA 3.0 (6Gbit/s) support, but this one doesn’t have it. Instead, this board includes just USB 3.0 support, which we’ll see shortly. Unlike Intel’s DH55TC, this board does include an IDE port, for those who still wish to hold onto their older drives.

Once again, our PCI configuration is almost identical to Intel’s, except ASUS has opted to put the PCI-E graphics slot in the middle of the two PCI-E 1x. This was likely done so as to not cover the BIOS battery, which is highly appreciated as that’s an issue commonly overlooked by many motherboard vendors. Along the bottom, you’ll find the FireWire and USB 2.0 internal headers. The USB headers would allow for 6 extra ports in total.

Most of the power phases available on this board are located right around the CPU socket, with futuristic-looking heatsinks behind them. As we’d expect, there’s an ample amount of room around the CPU socket for most any cooler on the market – especially those that will typically be used for an mATX motherboard.

Because the H55 chipset limits the overall USB port count to 12, six are dedicated as internal headers, while the other six are found at the back I/O section. Along with those, ASUS includes a FireWire, eSATA, LAN, full audio ports (including S/PDIF), PS/2 keyboard and of course, the three same display connectors we saw on Intel’s board. Though Intel is pushing DisplayPort, we’re likely only to see that connector available on higher-end H55/H57 motherboards.

The P7H55D-M EVO is the board I used throughout the majority of my testing, and from a design to features standpoint, I couldn’t have been much happier. The layout is great, installation was made easy, and I ran into no single hitch during my use. Fortunately, the BIOS stacks up to the rest of the board, so let’s take a look at that next.

ASUS’ P7H55D-M EVO – BIOS

I might not much care for Intel’s BIOSes, as I mentioned, but between all others out there, my favorite over time has been the ones ASUS has put out. Similar to Intel’s own BIOSes, ASUS uses a top-menu for the sorting of things, which I’ve come to like. It’s easier to go from left to right than it is up to down, across a two-sided menu. Plus, as soon as you go to a top-menu, it’s automatically opened, so there’s no need to hit Enter.

Like its desktop boards, the BIOS here is very robust, and as mentioned before, it focuses quite a bit on overclocking. The box boasts DDR3-2133 support, and with the 8+3 phase solution, along with the robust BIOS, high overclocks can, and will be reached. And quite easily, as we’ll see later in the article.

As I did with the Intel BIOS page, I’ll let the screenshots here speak for themselves. As ASUS’ board is the first non-Intel we received, it will be the first H55 board we officially review. If all goes well, I hope to get that up later this week.

Test System & Methodology

At Techgage, we strive to make sure our results are as accurate as possible. Our testing is rigorous and time-consuming, but we feel the effort is worth it. In an attempt to leave no question unanswered, this page contains not only our testbed specifications, but also a fully-detailed look at how we conduct our testing.

If there is a bit of information that we’ve omitted, or you wish to offer thoughts or suggest changes, please feel free to shoot us an e-mail or post in our forums.

Test System

The table below lists the hardware for our two current machines, which remains unchanged throughout all testing, with the exception of the processor. Each CPU used for the sake of comparison is also listed here, along with the BIOS version of the motherboard used. In addition, each one of the URLs in this table can be clicked to view the respective review of that product, or if a review doesn’t exist, you will be led to the product on the manufacturer’s website.

Component
AMD AM2+/AM3 Test System
Processors

AMD Phenom II X4 965 Black Edition – Quad-Core, 3.40GHz, 1.325v
AMD Phenom II X4 955 Black Edition – Quad-Core, 3.20GHz, 1.325v
AMD Phenom II X3 720 Black Edition – Tri-Core, 2.80GHz, 1.325v
Motherboard
Gigabyte MA790GP-DS4H – 790GX-based, F3 BIOS (01/13/09)
Memory

Corsair XMS3 DHX 2x2GB – DDR2-1066 5-5-5-15-2T, 2.10v
Graphics
Audio
On-Board Audio
Storage

Intel X-25M 80GB SSD
Seagate Barracuda 500GB 7200.11
Power Supply
Corsair HX1000W
Chassis
SilverStone TJ10 Full-Tower
Display
Gateway XHD3000 30″
Cooling
Thermaltake V1
Et cetera
Windows Vista Ultimate 64-bit

Component
Intel LGA1156 Test System
ProcessorsIntel Core i7-870 – Quad-Core, 2.93GHz, ~1.25v
Intel Core i5-750 – Quad-Core, 2.66GHz, ~1.25v
Intel Core i5-661 – Dual-Core, 3.33GHz, ~1.10v
Motherboard
Lynnfield: Gigabyte P55-UD5 – P55-based, F3 BIOS (08/01/09)
Westmere: ASUS P7H55D-M EVO – H55-based, 0503 BIOS (12/02/09)
Memory

Corsair XMS3 DHX 2x2GB – DDR3-1333 7-7-7-20-2T, 1.65v
Graphics
ATI Radeon HD 4870 512MB (Catalyst 8.11)
Audio
On-Board Audio
Storage

Intel X-25M 80GB SSD
Seagate Barracuda 500GB 7200.11
Power Supply
Corsair HX1000W
Chassis
SilverStone TJ10 Full-Tower
Display
Gateway XHD3000 30″
Cooling
Thermalright MUX-120
Et cetera
Windows Vista Ultimate 64-bit

Component
Intel LGA1366 Test System
Processors
Intel Core i7-975 Extreme EditionQuad-Core, 3.33GHz, 1.30v
Intel Core i7-920 Quad-Core, 2.66GHz, 1.30v
Motherboard
ASUS Rampage II Extreme – X58-based, 0705 BIOS (11/21/08)
Memory

OCZ Gold 3x2GB – DDR3-1066 7-7-7-20-1T, 1.56v (920/940)
OCZ Gold 3x2GB – DDR3-1600 7-7-7-20-1T, 1.56v (965)
Graphics
Audio
On-Board Audio
Storage
Intel X-25M 80GB SSD

Seagate Barracuda 500GB 7200.11
Power Supply
SilverStone DA1200
Chassis
SilverStone TJ10 Full-Tower
Display
Gateway XHD3000 30″
Cooling
Thermalright TRUE Black 120
Et cetera
Windows Vista Ultimate 64-bit

Component
Intel LGA775
Processors

Intel Core 2 Quad Q9650 – Quad-Core, 3.00GHz, 1.30v (Sim)
Intel Core 2 Quad Q9550 – Quad-Core, 2.83GHz, 1.30v (Sim)
Intel Core 2 Quad Q9400 – Quad-Core, 2.66GHz, 1.30v
Intel Core 2 Quad Q8200 – Quad-Core, 2.33GHz, 1.30v
Intel Core 2 Duo E8600 – Dual-Core, 3.33GHz, 1.30v
Intel Core 2 Duo E8500 – Dual-Core, 3.16GHz, 1.30v (Sim)
Intel Core 2 Duo E8400 – Dual-Core, 3.00GHz, 1.30v
Intel Pentium Dual-Core E5200 – Dual-Core 2.50GHz, 1.30v
Motherboard
ASUS Rampage Extreme – X48-based, 0501 BIOS (08/28/08)
Memory

Corsair XMS3 DHX 2x2GB – DDR3-1333 7-7-7-15-1T, 1.91v (1333FSB)
Corsair XMS3 DHX 2x2GB – DDR3-1066 6-6-6-15-1T, 1.91v (1066FSB)
Corsair XMS3 DHX 2x2GB – DDR3-800 6-6-6-15-1T, 1.91v (800FSB)

Graphics
Audio
On-Board Audio
Storage
Intel X-25M 80GB SSD

Seagate Barracuda 500GB 7200.11
Power Supply
Corsair HX1000W
Chassis
SilverStone TJ10 Full-Tower
Display
Gateway XHD3000 30″
Cooling
Thermalright TRUE Black 120
Et cetera
Windows Vista Ultimate 64-bit

(Sim) represents models that were tested using a faster, but under-clocked processor. For example, for the Q9550, we used the QX9770, since the specs are identical all-around, except for the clock speeds. Those were adjusted appropriately, effectively giving us a Q9550 to test with.

When preparing our testbeds for any type of performance testing, we follow these guidelines:

To aide with the goal of keeping accurate and repeatable results, we alter certain services in Windows Vista from starting up at boot. This is due to the fact that these services have the tendency to start up in the background without notice, potentially causing slightly inaccurate results. Disabling “Windows Search” turns off the OS’ indexing which can at times utilize the hard drive and memory more than we’d like.

Application Benchmarks

To help test out the real performance benefits of a given processor, we run a large collection of both real-world and synthetic benchmarks, including 3ds Max, Adobe Lightroom, TMPGEnc Xpress, Sandra 2009 and many more.

Our ultimate goal is always to find out which processor excels in a given scenario and why. Running all of the applications in our carefully-chosen suite can help better give us answers to those questions. Aside from application data, we also run two common games to see how performance scales there, including Call of Duty 4 and Half-Life 2: Episode Two.

Game Benchmarks

In an attempt to offer “real-world” results, we do not utilize timedemos in any of our reviews. Each game in our test suite is benchmarked manually, with the minimum and average frames-per-second (FPS) captured with the help of FRAPS 2.9.5.

To deliver the best overall results, each title we use is exhaustively explored in order to find the best possible level in terms of intensiveness and replayability. Once a level is chosen, we play through repeatedly to find the best possible route and then in our official benchmarking, we stick to that route as close as possible. Since we are not robots and the game can throw in minor twists with each run, no run can be identical to the pixel.

Each game and setting combination is tested twice, and if there is a discrepancy between the initial results, the testing is repeated until we see results we are confident with.

The two games we currently use for our motherboard reviews are listed below, with direct screenshots of the game’s setting screens and explanations of why we chose what we did.

Call of Duty: World at War

1680×1050
2560×1600




Crysis Warhead

1680×1050
2560×1600




Half-Life 2: Episode Two

1680×1050
2560×1600

Workstation: Autodesk 3ds Max, Cinebench, POV-Ray

Autodesk’s 3ds Max is without question an industry standard when it comes to 3D modeling and animation, with DreamWorks, BioWare and Blizzard Entertainment being a few of its notable users. It’s a multi-threaded application that’s designed to be right at home on multi-core and multi-processor workstations or render farms, so it easily tasks even the biggest system we can currently throw at it.

For our testing, we use two project files that are designed to last long enough to find any weakness in our setup and also allows us to find a result that’s easily comparable between both motherboards and processors. The first project is a dog model included on recent 3ds Max DVD’s, which we infused with some Techgage flavor.

Our second project is a Bathroom scene that makes heavy use of ray tracing. Like the dog model, this one is also included with the application’s sample files DVD. The dog is rendered at an 1100×825 resolution, while the Bathroom is rendered as 1080p (1920×1080).

Given their frequencies, the Core i5-661 is most comparable to the Core 2 Duo E8600, also at 3.33GHz. In that match-up, Intel’s Clarkdale comes out far ahead, rendering our bathroom scene about 41% faster, and our dog model about 10% faster. As we first discovered last fall with Nehalem, Intel’s latest architectures excel greatly where ray tracing is concerned, so the stark gains in our bathroom scene (which heavily uses ray tracing) is unsurprising.

Cinebench R10

Like 3DS Max, Cinema 4D is another popular cross-platform 3D graphics application that’s used by new users and experts alike. Its creators, Maxon, are well aware that their users are interested in huge computers to speed up rendering times, which is one reason why they released Cinebench to the public.

Cinebench R10 is based on the Cinema 4D engine and the test consists of rendering a high-resolution model of a motorcycle and gives a score at the end. Like most other 3D applications on the market, Cinebench will take advantage of as many cores as you can throw at it.

Cinebench doesn’t utilize ray tracing techniques, but our results here are still quite interesting. On a single-thread basis, the Core i5-661 scored 4668, or about 13% higher than the E8600. But thanks to HyperThreading, we saw a 35% gain in the multi-threading test. Any CPU built on Nehalem proves to be good for ray tracing, but the same can be said for multi-threading. HyperThreading makes such a large difference that it’s almost a non-option to go against using it.

POV-Ray 3.7

Similar to Cinebench, the “Persistence of Vision Ray Tracer” is as you’d expect, a ray tracing application that also happens to be cross-platform. It allows you to take your environment and models and apply a ray tracing algorithm, based on a script you either write yourself or borrow from others. It’s a free application and has become a standard in the ray tracing community and some of the results that can be seen are completely mind-blowing.

The official version of POV-Ray is 3.6, but the 3.7 beta unlocks the ability to take full advantage of a multi-core processor, which is why we use it in our testing. Applying ray tracing algorithms can be extremely system intensive, so this is one area where multi-core processors will be of true benefit.

For our test, we run the built-in benchmark, which delivers a simple score (Pixels-Per-Second) the the end. The higher, the better. If one score is twice another, it does literally mean it rendered twice as fast.

Continuing its ray tracing excellence, the Core i5-661 once again soars past the older, but equally-clocked, Core 2 E8600. Note that part of the reason for this isn’t only due to architecture enhancements, but also the introduction of Turbo, which auto-overclocks the CPU to a small degree during any sort of load.

Multi-Media: Adobe Lightroom, TMPGEnc Xpress

Photo manipulation benchmarks are more relevant than ever, given the proliferation of high-end digital photography hardware. For this benchmark, we test the system’s handling of RAW photo data using Adobe Lightroom, an excellent RAW photo editor and organizer that’s easy to use and looks fantastic.

For our testing, we take 100 RAW files (in Nikon’s .NEF file format) which have a 10-megapixel resolution, and export them as JPEG files in 1000×669 resolution, similar to most of the photos we use here on the website. Such a result could also be easily distributed online or saved as a low-resolution backup. This test involves not only scaling of the image itself, but encoding in a different image format. The test is timed indirectly using a stopwatch, and times are accurate to within +/- 0.25 seconds.

In almost all scenarios where the CPU is stressed, any Nehalem-derived processor is going to deliver faster performance than the previous generation, thanks to its Turbo, but with Lightroom, we still see a near 30% boost, an increase that Turbo itself has little hand in (Turbo in itself would make about a 10% performance boost in a totally multi-threaded application). So, the architecture enhancements shine through in tests like these.

TMPGEnc Xpress 4.5

When it comes to video transcoding, one of the best offerings on the market is TMPGEnc Xpress. Although a bit pricey, the software offers an incredible amount of flexibility and customization, not to mention superb format support. From the get go, you can output to DivX, DVD, Video-CD, Super Video-CD, HDV, QuickTime, MPEG, and more. It even goes as far as to include support for Blu-ray video!

There are a few reasons why we choose to use TMPGEnc for our tests. The first relates to the reasons laid out above. The sheer ease of use and flexibility is appreciated. Beyond that, the application does us a huge favor by tracking the encoding time, so that we can actually look away while an encode is taking place and not be afraid that we’ll miss the final encoding time. Believe it or not, not all transcoding applications work like this.

For our test, we take a 0.99GB high-quality DivX H.264 AVI video of Half-Life 2: Episode Two gameplay with stereo audio and transcode it to the same resolution of 720p (1280×720), but lower the bit rate in order to attain a modest file size. This test also utilizes the SSE instruction sets, either SSE2 or SSE4, depending on what the chip supports.

Not one to break tradition, the Core i5-661 continues to dominate, compared to its older dual-core siblings, with performance gains of between 15 – 30%, depending on codec or video workload.

Multi-Media: ProShow Gold, Sandra Multi-Media

While TMPGEnc XPress’ purpose is to convert video formats, ProShow from Photodex helps turn your collection of photos into a fantastic-looking slide show. I can’t call myself a slide show buff, but this tool is unquestionably definitive. It offers many editing abilities and the ability to export in a variety of formats, including a standard video file, DVD video and even HD video.

Like TMPGEnc and many other video encoders, ProShow can take full advantage of a multi-core processor. It doesn’t support SSE4 however, but hopefully will in the future as it would improve encoding times considerably. Still, when a slide show application handles a multi-core processor effectively, it has to make you wonder why there is such a delay in seeing a wider-range of such applications on the marketplace.

In many of our tests up to this point, where we tested with two different scenarios, we have seen one scenario experience a larger boost than the other. Here, that’s not the case, as both our HD and DVD encodes saw close to a 34% boost in performance.

Sandra 2009 Multi-Media

This test here stresses the CPU’s ability to handle multi-media instructions and data, using both MMX and SSE2/3/4 as the instruction sets of choice. The results are divided by integer, floating point and double precision, three specific numbering formats used commonly in multi-media work.

Real-world tests are the absolute best way to gauge a processor’s worth, but synthetics play an important role as well, as we can easily see where a product excels, or lacks. This test is one good example of that, because we can see just how much the Core i5-661 excels for float computation compared to the others. It proves to be the first CPU in our entire line-up that sees float performance faster than int and double. Overall, the performance is quite good all-around, with stark increases between the CPU on the table and the older Core 2 E8600.

Mathematics: Sandra Arithmetic, Crypto, Microsoft Excel

With each new processor launch, one thing that’s bound to prove faster are mathematical equations, which when all said and done, plays a massive role in a lot of our computing today. The faster an equation can be completed, the faster a math-heavy process can finish.

Sandra includes applications designed to specifically test the mathematical performance of processors, with the main one being the arithmetic test.

Similar to what we just saw with Sandra’s multimedia test, we can see here that once again, the Core i5-661 is the oddball of the bunch, with faster MFLOPS performance than its MIPS performance. Could this have something to do with the addition of the AES-NI instruction set? If I were a betting man, I think my odds would be good if I settled on that theory.

Sandra 2009 Cryptography

Crypto is a major part of computing, whether you know it or not, and certain processes can prove slower than others, depending on their algorithms. User passwords on your home PC are encrypted, as are user passwords on web servers (like in our forums). Past that, crypto is used in other areas as well, such as with creating of unbreakable locks on files or assigning a hash to a particular file (like MD5).

In Sandra’s Cryptography test, the results are outputted as MB/s, higher being better. Although this is somewhat of an odd metric to go by, generally speaking, the higher the number, the faster the CPU tears through the respective algorithm, which comes down to how fast a password is either encrypted, decrypted, signed, et cetera.

Go ahead and laugh… it’s fine. From both an AES and SHA perspective, the Core i5-661 is beyond excellent. Compared to the E8600, the i5-661 performs close to 35% faster. Where AES is concerned, there’s no competition… not even Intel’s highest-end chips. Compared to the E8600 once again, the i5-661 proved a staggering 12.39x faster for AES computation, and 5.6x faster than Intel’s highest-end processor, the Core i7-975 Extreme Edition.

What do these kinds of gains mean for you? Depending on your computing lifestyle, it could range from not much, to a whole lot. If you take advantage of encryption using AES, and the program you are using supports AES-NI, you’ll see incredible performance gains.

Like most new instruction sets, it’s going to take a little while before we see wide support for AES-NI, but given the performance gains, you can be sure that developers won’t be waiting around to support it, because it could mean having a much more successful product than the competition. Clock for clock, we saw a 12x performance boost… that’s incredibly significant. I can’t wait to see this properly supported in some tool, whether it be hard drive encryption, archiving or what have you.

Microsoft Excel 2007

Most, if not all, businesses in existence have to crack open a spreadsheet at some point. Though simple in concept, spreadsheets are an ideal way to either track information or compute large calculations all in real-time. This is important when you run a business that deals with a large amount of expenses.

Although the importance of how fast a calculation takes in an Excel file is, we include results here since they heavily test the mathematical capabilities of each processor. Because Excel 2007 is completely multi-threaded (it can even take advantage of an 8-Core Skulltrail), it makes for a great benchmark to show the scaling between all of our CPUs.

I’ll let Intel explain the two files we use:

Monte CarloThis workload calculates the European Put and Call option valuation for Black-Scholes option pricing using Monte Carlo simulation. It simulates the calculations performed when a spreadsheet with input parameters is updated and must recalculate the option valuation. In this scenario we execute approximately 300,000 iterations of Monte Carlo simulation. In addition, the workload uses Excel lookup functions to compare the put price from the model with the historical market price for 50,000 rows to understand the convergence. The input file is a 70.1 MB spreadsheet.

CalculationsThis workload executes approximately 28,000 sets of calculations using the most common calculations and functions found in Excel*. These include common arithmetic operations like addition, subtraction, division, rounding and square root. It also includes common statistical analysis functions such as Max, Min, Median and Average. The calculations are performed after a spreadsheet with a large dataset is updated with new values and must re-calculate many data points. The input file is a 6.2 MB spreadsheet.

It’s impossible to follow-up to the incredible AES results we saw in the previous test, but here, we can again see the Core i5-661 perform a fair bit faster than the E8600. The differences here are a bit smaller than what we saw in our other tests, but when the entire test takes under 30s, that’s no real surprise.

System: Sandra Memory, Multi-Core Efficiency

Generally speaking, the faster the processor, the higher the system-wide bandwidth and the lower the latency. As is always the case, faster is better when it comes to processors, as we’ll see below. But with Core i7, the game changes up a bit.

Whereas previous memory controllers utilized a dual-channel operation, Intel threw that out the window to introduce triple-channel, which we talked a lot about at August’s IDF. Further, since Intel integrates the IMC onto the die of the new CPUs, benefits are going to be seen all-around.

Before jumping into the results, we already had an idea of what to expect, and just as we did, the results seen are nothing short of staggering.

As I mentioned on the front page of this article, the integrated memory controller on Clarkdale is not on the CPU, but rather on the GPU. That means that whenever the CPU needs to access memory, it needs to go to another chip to do it. Simple thinking would tell you that this method shouldn’t prove much slower than the design of previous generations, where the IMC was on the chipset, but that’s not the case.

For whatever reason, with the IMC on the GPU, we see our latencies greatly increased – to the point where the brand-new Clarkdale has the same latency as the low-end Pentium E5200. From the cache performance standpoint, though, nothing at all is lacking, and that feature alone is more important than memory latency.

I queried Intel to see if it considered the higher latency to be a problem, and the response was:

The memory latency on Clarkdale will be higher than what was seen on Lynnfield due to the memory controller being on the GMCH (Ironlake) and not directly on the CPU. However, we don’t see this memory latency as a significant impact on most client applications + the memory BW that you see vs. a Pentium E5200 (Wolfdale).

In the past, I’ve done a lot of personal testing with both high bandwidth and low latency modules, and from what I saw with that, I’d have to side with Intel’s thinking. I don’t believe that higher memory latencies aren’t going to effect certain workloads, because they certainly will, but for the majority of people, no real difference will ever be seen. In an article I published last fall, which took a look at memory performance on Core i7, only a single application of all we tested saw a difference with tighter timings (Adobe Lightroom). Clarkdale = Mainstream, so these higher latencies are likely to be of very low concern for most people, and for good reason.

That’s not to say that I wouldn’t like to see even tighter numbers, though, because I certainly would. If the IMC was built into the CPU, we wouldn’t see this issue. Essentially, Clarkdale’s implementation is similar to previous integrated designs, where the GPU is part of the Northbridge. There, the IMC is also built into the GPU, because for a platform like this, high performance just isn’t needed. When the time comes where we see the GPU and CPU fused together (whenever that happens), we’ll undoubtedly see the latencies decrease. But for now, I can’t view this as a real problem, given that we’ve yet to see throttled performance in our real-world tests due to it.

Sandra 2009 Multi-Core Efficiency

How fast can one core swap data with another? It might not seem that important, but it definitely is if you are dealing with a true multi-threaded application. The faster data can be swapped around, the faster it’s going to be finished, so overall, inter-core speeds are important in every regard.

Even without looking at the data, we know that Core i7 is going to excel here, for a few different reasons. The main is the fact that this is Intel’s first native Quad-Core. Rather than have two Dual-Core dies placed beside each other, i7 was built to place four cores together, so that in itself improves things. Past that, the ultra-fast QPI bus likely also has something to do with speed increases.

Continuing the domination trend, the Core i5-661 proves superior compared to its lower competition where both cache bandwidth and latencies are concerned. For Nehalem-based processors, there’s no competition in this regard.

Gaming: Call of Duty: World at War, Half-Life 2: Episode Two

Note: Our gaming tests on this and the next page were completed with an ATI Radeon HD 4870 installed, not the integrated graphics processor. Our goals here is to see how Westmere stacks up compared to all of the other CPUs we’ve tested in recent memory. Many people who purchase Westmere will purchase a discrete GPU, so this information is important to note. We have specific IGP benchmarks and performance three pages over.

While some popular game franchises are struggling to keep themselves healthy, Call of Duty doesn’t have much to worry about. This is Treyarch’s third go at a game in the series, and a first for one that’s featured on the PC. All worries leading up to this title were all for naught, though, as Treyarch delivered on all promises.

To help keep things fresh, CoD: World at War focuses on battles not exhaustively explored in previous WWII-inspired games. These include battles which take place in the Pacific region, Russia and Berlin, and variety is definitely something this game pulls off well, so it’s unlikely you’ll be off your toes until the end of the game.

For our testing, we use a level called “Relentless”, as it’s easily one of the most intensive levels in the game. It features tanks, a large forest environment and even a few explosions. This level depicts the Battle of Peleliu, where American soldiers advance to capture an airstrip from the Japanese. It’s a level that’s both exciting to play and one that can bring even high-end systems to their knees.

Luckily for hardcore CoD players, the game’s performance doesn’t change with a faster CPU, which is rather impressive. Here, the game ran just as well on our lowly E5200 as it did on our i7-975.

Half-Life 2: Episode Two

The original Half-Life 2 might have first seen the light of day close to four years ago, but it’s still arguably one of the greatest-looking games ever seen on the PC. Follow-up versions, including Episode One and Episode Two, do well to put the Source Engine upgrades to full use. While playing, it’s hard to believe that the game is based on a four+ year old engine, but it still looks great and runs well on almost any GPU purchased over the past few years.

Like Call of Duty 4, Half-Life 2: Episode Two runs well on modest hardware, but a recent mid-range graphics card is recommended if you wish to play at higher than 1680×1050 or would like to top out the available options, including anti-aliasing and very high texture settings.

This game benefits from both the CPU and GPU, and the skies the limit. In order to fully top out the available settings and run the highest resolution possible, you need a very fast GPU or GPUs along with a fast processor. Though the in-game options go much higher, we run our tests with 4xAA and 8xAF to allow the game to remain playable on the smaller mid-range cards.

Unlike CoD, HL2: Episode Two does love extra CPU power, and that’s evidenced above, but only at the highly-sporadic 1680×1050 resolution. That resolution has proven to be a chore, because the average FPS can fluctuate a great deal. What’s important to note here is that at our top setting of 2560×1600, the differences are almost zero.

Gaming: Crysis Warhead, 3DMark Vantage

As PC enthusiasts, we tend to be drawn to games that offer spectacular graphics… titles that help reaffirm your belief that shelling out lots of cash for that high-end monitor and PC was well worth it. But it’s rare when a game comes along that is so visually-demanding, it’s unable to run fully maxed out on even the highest-end systems on the market. In the case of the original Crysis, it’s easy to see that’s what Crytek was going for.

Funny enough, even though Crysis was released close to a year ago, the game today still has difficulty running at 2560×1600 with full detail settings – and that’s even with overlooking the use of anti-aliasing! Luckily, Warhead is better optimized and will run smoother on almost any GPU, despite looking just as gorgeous as its predecessor, as you can see in the screenshot below.

The game includes four basic profiles to help you adjust the settings based on how good your system is. These include Entry, Mainstream, Gamer and Enthusiast – the latter of which is for the biggest of systems out there, unless you have a sweet graphics card and are only running 1680×1050. We run our tests at the Gamer setting as it’s very demanding on any current GPU and is a proper baseline of the level of detail that hardcore gamers would demand from the game.

Our previous games didn’t show real differences between CPUs, and Crysis Warhead is no different. You can be rest-assured that no matter your PC, this game is going to run like molasses!

Futuremark 3DMark Vantage

Although we generally shun automated gaming benchmarks, we do like to run at least one to see how our GPUs scale when used in a ‘timedemo’-type scenario. Futuremark’s 3DMark Vantage is without question the best such test on the market, and it’s a joy to use, and watch. The folks at Futuremark are experts in what they do, and they really know how to push that hardware of yours to its limit.

The company first started out as MadOnion and released a GPU-benchmarking tool called XLR8R, which was soon replaced with 3DMark 99. Since that time, we’ve seen seven different versions of the software, including two major updates (3DMark 99 Max, 3DMark 2001 SE). With each new release, the graphics get better, the capabilities get better and the sudden hit of ambition to get down and dirty with overclocking comes at you fast.

Similar to a real game, 3DMark Vantage offers many configuration options, although many (including us) prefer to stick to the profiles which include Performance, High and Extreme. Depending on which one you choose, the graphic options are tweaked accordingly, as well as the resolution. As you’d expect, the better the profile, the more intensive the test.

Performance is the stock mode that most use when benchmarking, but it only uses a resolution of 1280×1024, which isn’t representative of today’s gamers. Extreme is more appropriate, as it runs at 1920×1200 and does well to push any single or multi-GPU configuration currently on the market – and will do so for some time to come.

The results here are just as we expected. Generally, the better the CPU, the higher the score. The overall 3DMark Score doesn’t vary much, however, as the benchmark doesn’t weigh the CPU score that heavily, which after taking a look at our three games tested here, is a good thing.

Power Consumption

It goes without saying that power efficiency is at the forefront of many consumers’ minds today, and for good reason. Whether you are trying to save money or the environment – or both – it’s good to know just how much effort certain vendors are putting into their products to help them excel in this area. Both AMD and Intel have worked hard to develop efficient chips, and that’s evident with each new launch. The CPUs are getting faster, and use less power, and hopefully things will stay that way.

To help see what kind of wattage a given processor draws on average, we use a Kill-A-Watt that’s plugged into a power bar that’s in turn plugged into one of the wall sockets, with the test system plugged directly into that. The monitor and other components are plugged into the other socket and is not connected to the Kill-A-Watt. For our system specifications, please refer to our methodology page.

To test, the computer is first boot up and left to sit at idle for five minutes, at which point the current wattage is recorded if stable. To test for full CPU load, LinX is run with 2560MB memory usage for a total of five minutes. During that run, the highest point the wattage reaches on the meter is captured and becomes our “Max Load”.

Throughout all of our tests so far, Clarkdale has proven to be in a unique class. Compared to all other dual-core CPUs, it’s extremely fast, and showcases gains of between 10 – 35% on average. Of course, being built on a 32nm process, along with its architecture upgrades, we could assume that the Core i5-661 will also excel where power consumption is concerned, and we’re spot on with that… just look at the results.

Because all of our other configurations were tested with a discrete graphics card, I included two sets of results here. One has the GPU installed (Radeon HD 4870), and the other without, so the IGP steps up to the plate. With or without the discrete card, the i5-661 turns out to be the most power efficient of the entire bunch. It’s idle power draw doesn’t change much from its closest competitors, but the only CPU to come close to its low load is AMD’s budget 240e dual-core.

Clarkdale’s GPU: Gaming

Throughout the article thus far, we’ve made it clear that the graphics processor in Clarkdale isn’t designed for the enthusiast, and I’m sure that comes as a surprise to no one. Integrated graphics in general are nowhere near as advanced as discrete cards where gaming performance is concerned, which is one of the reasons Intel touts its IGP as being perfectly suited for the casual gamers, who play games like Sims 3, and nothing more advanced than World of Warcraft.

During my time with Clarkdale, I spent a fair amount of time on the gaming aspect, and tested out a variety of casual games that Intel mentioned. Sure enough, the performance was just fine, and I ran into no show-stopping issues. For casual games, I tested out a couple of PopCap titles, along with Sims 3, Nancy Drew: Ransom of the Seven Ships, Crazy Machines 2, and had no problem with any.

The rule of thumb is, if it’s a casual game with relatively simple graphics, then it should run fine on Clarkdale. Sometimes, though, even casual games have a hard time in running, so it really depends on just how much acceleration the game wants to use. One example is Gumboy: Crazy Adventures. This 2D platform game makes heavier use of accelerated graphics than most casual games, so I had a difficult time getting it to run smoothly. Oddly, the game ran at about 30FPS, but every couple of seconds, the game would stick for about 3 – 4 seconds, then unstick and run fine again.

I couldn’t test out every casual game out there, but just be aware that if the one you’re after makes interesting use of 3D, it may not run perfectly. This seems to vary from game to game, as I played games much more advanced than Gumboy just fine, so it’s a little hard to pinpoint what will work fine and what won’t. Still, we’ve established that for casual gamers, Clarkdale is fully suitable, but what about the more advanced games, such as FPS and racing titles?

To help give Clarkdale an appropriate test in this regard, I loaded up a variety of current games to see what it would take to make them run, and I admit, I had a fair bit of luck with most titles. Below, you can see a variety of screenshots from the four main games I tested with, and though the graphics aren’t top-rate, even at the given resolution, it’s certainly nice to be able to get any gaming like this on an IGP.

In Call of Duty: Modern Warfare 2, Clarkdale was able to handle a 1280×1024 resolution, with the lowest detail settings possible. During my testing, I ran into a fair amount of issue with this particular title, but it’s frustrating, because I feel like I shouldn’t have. For the most part, the game ran just fine, and was smooth enough for me to complete multiple online matches, but for whatever reason, about 50% of the time, the game would crash during a round, locking up the entire computer.

I talked to Intel about this, and it was a little confused as to the problem. One member of their benchmarking team has been running the game on an Arrandale notebook for the last little while, and while he did note that this issue occurred once, he also mentioned that he completed half the single-player mode without issue.

During all the crashes I experienced, there was no telling the reason, but if I had to guess, I’d blame the limited 128MB GPU memory. We’re using a low resolution here, but games use more memory than ever, so being able to dedicate more than 128MB would be nice. The only reason I assume this could be the issue is that one crash happened at the same time as an explosion occurred in front of me.

Call of Duty: Modern Warfare 2

“Can it run Crysis?” Yes, yes it can. Albeit at 1024×768 resolution and the lowest possible detail settings. Despite all that, the game actually didn’t look too bad, and it ran quite well, with no random sticking or lag. I played through most of the first level in the game to come to this conclusion, and also a level later in the game which takes place in a snowy world. Overall, the performance here was rather impressive. 1280×1024 was simply impossible, though, for probably obvious reasons. Crysis is one heck of a glutton when it comes to PC games.

Crysis Warhead

Similar to Modern Warfare 2, F.E.A.R. 2 ran at 1280×1024 just fine, with the lowest detail settings. There were no real anomolies to mention here, except that by looking at the screenshots now, the game appeared a bit dark. I don’t believe that to be the fault of the GPU at any rate, but rather that I simply didn’t adjust the in-game brightness at all. Because my monitor tends to automatically brighten up dark scenes, I didn’t notice it at the time.

F.E.A.R. 2: Project Origin

Left 4 Dead 2 was a bit trickier to get working well, but in the end, 1280×1024 was again possible on the lowest detail settings. The performance wasn’t perfect, but it was certainly playable, even in the heavier action scenes. But, given that performance isn’t superb, I’m not sure I’d recommend this for serious online gamers, as it’s harder to be a precision gamer with lower resolutions and weaker performance.

Left 4 Dead 2

While all of the above games worked relatively painlessly on Clarkdale, there were a few other titles I tried out that didn’t work out well at all. One example is seen below, Call of Juarez: Bound in Blood. As you can see, the game is textureless, and though it’s not easy to tell by the picture, the models themselves were stick figures… unable to move their arms, legs and so forth. Rather, they simply slid along the landscape, which was pretty funny to see.

Call of Juarez: Bound in Blood

Another game I tested, Race Driver: GRID, failed to load at all. The screen would turn black, but at that point, nothing would ever happen, so I was forced to Alt-Tab out in order to kill the process. Overall, I had
a couple of games with bizarre behavior, so it’s worth noting. But again, Intel is stressing this as a chip to handle casual and mainstream gaming – game titles that make up near 95% of the market’s sales. It does that, so Intel’s mark has been hit.

You can see that for the most part, the performance isn’t extreme, but it’s definitely manageable. On lower resolutions, you can seem to get by a bit easier on lower FPS, so even with ~25FPS sounding awful, all of these games were surprisingly playable, even Crysis, with its minimum FPS of 12.

I really wish I knew what was causing my Modern Warfare 2 issues, though, because most of the time, the game ran very well. It was just those random crashes that got in the way. Intel notes that it’s probably a software issue, not an issue with its hardware or driver, and whether or not that’s the case or not, I’d hope to see improvements made in the future.

Clarkdale’s GPU: Blu-ray Playback

As we touched on the GPU earlier in the article, it became clear that one of the biggest features of Clarkdale’s IGP is full HD support, which currently tends to always point to Blu-ray. In addition to full AVC/VC-1/MPEG2 acceleration, dual-decode has been added, for those flicks that offer a picture-in-picture feature, usually for commentary. That feature alone proves that Clarkdale is more than capable of delivering a compelling Blu-ray experience, given it can essentially handle two playbacks at the same time.

For a more thorough look at what else Clarkdale’s IGP brings to the HD arena, I recommend looking back through page three of this article, as it was all tackled there. With Blu-ray such a focus of Clarkdale, I couldn’t have found a much better excuse to take time away from work to kick back and watch a couple of movies. So that’s just what I did, and all of my findings are below.

As mentioned above, Clarkdale’s IGP supports acceleration for the three most common Blu-ray encoding formats, AVC (MPEG 4), VC-1 and also MPEG 2, which is typically used for alternate features, not the main film itself. Either way, if you have a Blu-ray movie, you can be sure it will benefit from the acceleration offered here, which at the end of the day, means lower CPU usage and smooth-as-a-baby’s-butt playback.

Throughout all of my testing with Clarkdale, I tested out a wide variety of Blu-ray’s from my collection, to see if any stuttering or any other anomaly occurred. I’m pleased to say that I couldn’t find a single issue, and my experience overall was excellent. I watched through both 2 Fast 2 Furious (AVC) and Fast & Furious (VC-1) in their entirety, using CyberLink’s PowerDVD 9 Ultra, and there were no hiccups whatsoever to speak of. The movies loaded fast, paused and restored fast, and had no graphical issues whatsoever.

Below are direct screengrabs from a couple of Blu-ray’s I tested rather thoroughly, and as you can see, all look excellent, with rich color, sharpness and clarity.

Fast & Furious
Fight Club
Iron Man
No Country for Old Men

To see how Clarkdale fared performance-wise with its Blu-ray playback, I ran two movies, with different encodings, through three of the most popular DVD/Blu-ray playback suites on the market, Corel’s WinDVD, CyberLink’s PowerDVD and also ArcSoft’s TotalMedia Theatre. To test, I played the same 30 minute portion of each movie with each player, and used Everest Ultimate Edition 5 in order to track CPU usage. The results below are the average and maximum CPU usage percentage over that half-hour span.

It’s clear that Blu-ray playback is hardly an issue for Clarkdale, with an average of 10 – 14% of the entire CPU being used. Even the maximum figure we saw, 23% with Fast & Furious using PowerDVD, is modest… that’s a mere single thread, out of the four that the CPU offers. It’s easy to understand why dual-decode is possible, with so much horsepower left over. Overall, Clarkdale can handle Blu-ray without an issue, it goes without saying.

Note that CPU usages may vary when you’re using additional high-end features, and perhaps even display connectors, but from what we saw here (running through HDMI), it would take a lot to make a dent in the CPU usage, regardless of your setup.

Final Thoughts

Clarkdale might be a mainstream processor, but it’s easily one of the most interesting we’ve ever taken a look at here at Techgage. Likewise, putting together this launch article required a lot more work than what’s typical, because we had so many new factors to take a look at, such as with the integrated graphics chip. All in all, though, benchmarking and playing around with Clarkdale was a lot of fun, and for many different reasons.

At the forefront, the biggest feature of Clarkdale is the fact that it’s built on a 32nm process. Intel is officially the first out the door with such a processor, and as we saw throughout our power consumption tests, there are some benefits to be reaped from that, as both the idle and load power draw decreased a fair bit, even with a discrete GPU.

Clarkdale is our first real foray into dealing with IGP, so unfortunately we don’t have comparative results to G45 or other IGP solutions. However, what I found extremely interesting is that at absolute full load, our machine didn’t go an inch above 124W, and it idled at 76W. During a regular workload, these results are going to be even lower. Compared to a regular gaming PC, these results are almost 1/4 of typical power draw, so that’s exciting to see.

From a gaming perspective, the Clarkdale IGP delivers where it needs to. It handles most casual games just fine, and even some more advanced titles, such as F.E.A.R. 2: Project Origin and Left 4 Dead 2. The IGP is by no way a replacement for a discrete GPU, because even a $50 – $80 graphics card would improve performance at least 5x, but it’s certainly better than having no gaming potential at all.

What I am most impressed with is the IGP’s Blu-ray performance. Intel’s G45 was riddled with issues from the get-go, but after those were fixed, we saw the company go on a road to recovery, and with its Westmere launch, it looks as though we’ll all see a smooth launch. I spent quite a bit of time with Blu-ray testing, and didn’t once see any issue, whether it be stuttering, slow playback or anything of the like.

Up-Close Photo of a Westmere Wafer

When Intel launched its Nehalem processors last fall, I think few people were left unimpressed. Of all the features I dug about the architecture, it was probably HyperThreading that I came to like so much, because it made a real difference in many different scenarios – especially 3D rendering. The new microarchitecture also made huge strides with ray tracing performance, as is easily seen in our 3ds Max and POV-Ray tests. Then there’s Turbo… which is essentially free performance for those who don’t want to overclock.

With Lynnfield, certain Nehalem features were chopped off, such as a triple-channel memory controller, but overall, nothing was dropped that would affect most people. If there was one thing, it would be the lack of HyperThreading on the Core i5-750. What I find interesting about Westmere, at least up to this point, is that all of the models (aside from the Pentium) support HyperThreading. That’s a bit bizarre, if you ask me, because out of the entire Core i line-up, the Core i5-750 is the only model currently lacking the feature.

I supposed this will begin to make a little more sense if Intel continues to release follow-up models that also axe the feature, but it’s a bit strange to see all of these sub-$200 Westmere chips support it, and not the Core i5-750, which comes in at $195. But while that’s a potential downside for the i5-750, it’s a huge benefit to all the Westmere chips, because as I mentioned before, HyperThreading is a very good thing, and in applications that can take advantage of it, huge benefits are seen.

From the perspective that Westmere is essentially a dual-core Lynnfield scaled down to 32nm with AES-NI and an IGP tacked on, there’s absolutely nothing to dislike about the chip. Last fall, I was wowed by Nehalem thanks to its incredible performance, and with Lynnfield, my love for the architecture only increased. With Westmere, we finally see Intel’s latest and greatest microarchitecture available to the masses. That’s fantastic.

If there’s one aspect of Westmere, or in particular, Clarkdale, that really sticks with me, it’s not the product itself, but the direction computing is going. It’s no surprise to anyone that computers are getting smaller and smaller, and integration is more hyped than ever, but with Clarkdale’s launch, we’re really starting to see an increased focus from vendors for small PCs.

For Clarkdale, the most popular chipset choice will be H55, and from what I’ve seen up to this point, the majority of boards equipped with it will be of the mATX form-factor, not ATX. Rather, vendors will focus on H57 for the latter form-factor. In talking to various board vendors, the general consensus seems to be that we are indeed heading quickly into the small PC realm, and Clarkdale is the first real push, thanks in part to its H55 counterpart.

There’s been a lot of discussion lately on our forums regarding this very subject, and it’s becoming increasingly clear by the day that for most people, mATX boards suffice. I can’t possibly disagree with that, because for most people, all they need is a single GPU, and another slot for some other peripheral. It’s not as though mATX boards are lacking on storage or peripheral expansion, so the smaller form-factor makes a ton of sense.

It can be argued that previous generations had similar launches, where mATX boards were the emphasis. That’s true, because with G45 as an example, all of the boards to use it were mATX. But the difference here is that the Clarkdale line-up is Intel’s latest mainstream chip, and it’s likely to be paired with an mATX board. When Core 2 Duo’s launched, we still saw ATX boards as the focus, not mATX. Either way, the shift to smaller computing is definitely the focus by many vendors today.

With that, it makes sense to call Clarkdale both a great desktop chip, and also one that’s perfectly suited for HTPC use. Regardless of how you use it, Clarkdale is a very, very fast dual-core, and it makes many strides in both performance, efficiency and other features when compared to the Core 2 line-up. As much as Core 2 will always have a place in my heart, I’m stoked that Nehalem is finally available to everyone.

Discuss this article in our forums!

Have a comment you wish to make on this article? Recommendations? Criticism? Feel free to head over to our related thread and put your words to our virtual paper! There is no requirement to register in order to respond to these threads, but it sure doesn’t hurt!

Copyright © 2005-2017 Techgage Networks Inc. - All Rights Reserved.