Maximum PC phenom II RSS Feed

Intel's Core i5 Analyzed and Tested (150+ Benchmarks)

Gordon Mah Ung — Tue, 08 Sep 2009 10:00:00 -0500

Intel’s latest troika of new CPUs brings Nehalem goodness to the masses

Nehalem for everyone! That simple sentence best explains Intel’s brand-new series of CPUs, which is sure to please budget users everywhere while confounding power users.

Why would a new CPU that gives you the best bang for the buck in town be greeted nervously? Because Intel’s new CPU brings with it a new socket as well as a new infrastructure. This new infrastructure is essentially a fork in the road that forces users to make a difficult choice: Save money today but get locked out of the high-end, or splurge today knowing that the budget CPU is damn near as good as the top-end part.

For the details on Intel’s new budget monster, savor our full report, consume the specs, and then digest the benchmarks to see just which path your next PC should take.

Meet Lynnfield

We normally don’t use a CPU’s code-name once its real name is known, but to help keep your head from popping off over Intel’s confusing naming scheme we’re going to rely on some code-names here for clarity.

Intel’s newest CPU family is code-named Lynnfield. The lineup includes the 2.93GHz Core i7-870, the 2.83GHz Core i7-860, and the 2.66GHz Core i5-750. Lynnfield chips use essentially the same microarchitecture as Intel’s original Core i7 CPUs, which were code-named Bloomfield, but the new CPUs are incompatible with existing Core i7 motherboards. That’s right, you could walk into a store and buy a Core i7 CPU that will not work with the Core i7 motherboard you just bought. Likewise, the Core i7 heatsink cooler you bought may not work with a new Core i7, either.

Despite its smaller size, Intel’s new Lynnfield Core i7/Core i5 CPU (right) actually features an increased transistor count of 774 million and a larger die size of 296 square millimeters, compared with the 731-million-tranny Bloomfield Core i7 (left) and its 263 square-millimeter die.

Socket Switcheroo

The most notable difference in this new crop of Core i7s is the socket. For Lynnfield, Intel is introducing the LGA1156 socket. This socket is, as stated, incompatible with the current LGA1366 motherboards and CPUs. To irk you even more, even the heatsink cooler mounting holes are incompatible, so you probably couldn’t use an LGA1366 cooler, even if you happened to have one. And even more annoying to enthusiasts, LGA775 coolers are also incompatible. Earlier this year, Intel execs told Maximum PC that LGA775 cooler compatibility was being considered for the new chip, but obviously the company has since ruled that out. To break it down: LGA1366 uses a 12cm gap, LGA1156 uses an 11cm gap, and LGA775 uses a 10cm gap. Yes, one centimeter difference and you have to dustbin your pricey high-rise cooler even though it’s capable of handling the thermals of the new chip.

We asked Intel if it was doing this just to piss people off and the company said no, it did it for legitimate engineering reasons. Intel actually lowered the height of the new direct socket load mechanism that clamps the CPU in place, which required moving the mounting holes out. Existing heatsinks capable of the thermal load should work, Intel said, so long as consumers obtain updated mounting brackets from the cooler maker. We have to also note that very new high-end coolers are coming with mounts for LGA1156 too. Still, make sure that if the box says Core i7, support for LGA1156 is included.

Although Intel wouldn’t confirm this, we’ve been told by high-end system builders that certain LGA1366 motherboards and coolers would flex enough to create a gap between cooler and CPU. The new design presumably fixes that problem.

Sockets Compared

Lynnfield Socket

Bloomfield Socket

The new budget Core i5/Core i7 CPUs use a new LGA1156 socket design (top image), which is incompatible with existing LGA1366 Core i7 motherboards and coolers (bottom image). The design eliminates traces for the third memory controller, but builds in wires that will accommodate forthcoming CPUs with integrated graphics cores.

Memory Loss

Enthusiasts will also question the move from tri-channel DDR3 to dual-channel DDR3—why go back if wider is better? Intel’s decision is based on pragmatism and cost. The tri-channel circuitry in the CPU doesn’t add much cost to the processor, but it’s not cheap to implement when building a motherboard. Those added traces from the socket to the RAM slots mean more layers and pricier boards. That’s one of the contributing factors to excessively priced X58 boards this past year.

Should you be concerned about shifting down to dual-channel? Generally, no. For the most part, only the most memory-bandwidth-intensive apps will actually see any performance hit. The fantastic latency and overall bandwidth capacity of the Nehalem design is more than adequate for today’s applications. This doesn’t mean it’s a non-issue. The vast majority of Lynnfield motherboards we’ve seen are opting for four-DIMM slots. That means a maximum of 8GB using affordable 2GB DIMMs (4GB DIMMs are currently cost-prohibitive). The only board we’ve seen with more DIMM slots is Gigabyte’s GA-P55-UD6, which features six, for a total of 12GB, using 2GB DIMMs. The memory controller in Lynnfield tops out at 16GB, while Bloomfield maxes out at 24GB. Realistically, 24GB of RAM is way overkill for 99 percent of us. Our experience has shown us that most apps do not consume that much RAM; 4GB to 6GB is the sweet spot today.

PCI-E at the Core

Another significant change for Lynnfield is that PCI-E comes directly off the CPU core. In X58/Bloomfield, X48/Core 2 Quad, and 790FX/Phenom II, PCI Express is external to the CPU, residing in the north-bridge chipset. Data is pumped out to the chipset where it must then be directed to the PCI-E slots that connect to the GPUs. As you can imagine, the extra hop creates a latency issue—that’s eliminated with Lynnfield. Lower latency is better, right? Yes and no. While latency is better on Lynnfield, the CPUs feature but a single x16 PCI-E 2.0 lane in the chip. That’s fine for a single GPU, but multi-GPU configurations will have to split the bandwidth. That means an SLI or CrossFire X rig will run both cards in x8 mode. Remember, however, that we’re talking x8 at PCI-E 2.0 speeds, which is 8GB/s for each card. From what we’ve seen and from what vendors have told us, only at the highest resolutions with antialiasing do you see any impact.

But, you say, what about tri-SLI? That, too, can be theoretically supported. Even though Lynn-field supports but a single x16 PCI-E 2.0 lane, additional PCI-E lanes are added through the P55 chipset—at the cost of latency. Previous designs that had PCI-E plumbed from both the north bridge and south bridge were dinged for doing just this. Some vendors are working around the lack of bandwidth by adding Nvidia’s nForce 200 chip to their boards. The nForce 200 doesn’t magically add bandwidth but it does manage the available bandwidth across multiple slots. Thus, a board vendor could add three or even four x16 PCI-E slots and have the nForce 200 chip manage the load for multi-GPU configurations. This would eliminate the need to have the GPUs feed off of the higher-latency connection in the chipset.

CrossFire X support is a given in the vast majority of P55 motherboards. And SLI will be as well for the board vendors who pay Nvidia to “certify” that their boards are SLI-ready. Fortunately, the big names are already onboard with that, including Asus, Gigabyte, MSI, EVGA, and even Intel.

If you’re wondering why Intel doesn’t just fix the lack of bandwidth by adding additional PCI-E lanes in the CPU, consider this: Intel would have to add additional traces from the CPU and the socket, and even Intel is loathe to introduce yet another new socket standard so soon. The company also doesn’t think it’s worth it as anyone who truly wants full dual-x16 GPU support should really be buying X58: Remember, folks, Lynnfield and P55 are for the “mainstream.”

The Lynnfield LGA1156 again rejiggers the design of the modern CPU. With the Athlon 64 and the Core i7, the memory controller was moved from the north bridge directly into the CPU core. With Lynnfield, the PCI-Express ports are now handled directly by the CPU as well. Furthermore, relatively low-speed data from the hard drives and USB ports flows though a single DMI link at 2GB/s with Lynnfield.

P55: Downsizing the Chipset

In the old days, new core-logic chipsets were almost as exciting as a new CPU. Not so today. With X58 and Bloomfield, the chipset got fired from its job of managing the memory controller. With Lynnfield, it even loses its responsibilities for managing PCI-E, as both features are now integral to the CPU itself. Today, the chipset is nothing more than a glorified south bridge, managing USB, SATA, PCI, PS/2, and other decidedly unsexy items. The P55 might have been sexier if it had USB 3.0 or SATA 6 functionality, but that won’t appear until next year. And even if it had those features, it’s pretty clear where the P55 stands: Its desk has been moved to the basement and its red Swingline stapler has been confiscated.

Tweakability

There was some initial confusion when Bloomfield was introduced. Early reports had it that the memory multipliers were unlocked on the high-end Extreme versions of the CPU, but locked on the lower-end versions. It turned out that was only the case for samples first sent to the press and system vendors. Retail versions of the lower-end Bloomfields were unlocked as well, making it possible for users to set the RAM at speeds higher than the rated DDR3/1066. This time around, there’s no such confusion. All three new Lynnfields feature unlocked memory multipliers and the chips are actually officially rated for operation at DDR3/1333, with higher speeds obtainable through “overclocking.” Of course, the chips are upwardly clock-locked, so you can’t simply set your $200 Core i5 to run at 4GHz by changing the multiplier.

Though open about memory locks, Intel has been cagier about Turbo mode. With Bloomfield, Intel never revealed to consumers the top clock speeds the CPUs could hit under Turbo mode, but with Lynnfield, it’s now publicizing the top speed that a single core can hit. The Core i7-870, for example, can top out at 3.6GHz by overclocking individual cores based on the thermals and power consumption of the chip. That’s actually far greater than the Bloomfield modes can top out at on default. The take-away is that, with the months it has had to tinker with Lynnfield, Intel has improved Turbo mode (now officially Turbo Boost). However, the feature is locked. Only on the Extreme parts will users be able to tinker with Turbo beyond the default caps.

The Big Decision

All this techno-speak is meaningless if the processor doesn’t perform as expected. We don’t want to give away everything here but let’s just say we’re not disappointed. Lynnfield is everything you’d expect of a new iteration of Nehalem and Intel’s now-very-mature 45m process. As such, overclocking, is also fruitful. By simply boosting the base clock of the cheapie $200 Core i5-750, we were able to take it from 2.66GHz to a very stable 3.5GHz without any additional voltage on our very first attempt using a Gigabyte GA-P55-UD6 board.

System builders have been equally impressed with these budget parts and have achieved overclocking results every bit as good as, if not better than, the most expensive Core i7-975 Extreme Edition parts.

That brings us to the main question: Why even build a Bloomfield LGA1366 at this point if building a Lynnfield system will save you at least $100 on the board and a little on the RAM, and even more money if you opt for the ultra-budget Core i5-750?

Here’s that fork in the road: Lynnfield is cheaper and gets you 90 percent the performance of a Bloomfield system, but early next year Intel will introduce a CPU code-named Gulftown, aka Core i9. Core i9 adds two more physical cores to the CPU and will likely be the first consumer hexacore CPU. With Hyper-Threading, that’s 12 threads available to the OS and enough to make the most jaded enthusiast perk up. Core i9, however, will only be available on the Bloomfield/LGA1366 platform. If you were to build a Lynnfield LGA1156 box there’d be no six-core for you! At least, not at this point. Intel said it has no plans for an LGA1156 hexacore. You see the dilemma. Save money now and build a really kick-ass LGA1156 or spend the extra $200 to build an LGA1366 that has an easy upgrade path to six cores with Hyper-Threading. It’s not an easy choice to make under normal circumstances, but in this economic climate, it’s even harder—that $200 goes a long way toward a better GPU, better PSU, more RAM, or a bigger hard drive. The choice, however, is up to you.

Next, the CPU buyers guide and benchmarks!

CPU Buyers Guide

What’s the best budget chip available today for those interested in getting good performance on the cheap? We’ll walk you through the top five chips and tell you which one to buy.

CPU Specs

CPU	Core i7-920	Core i7-870	Core i7-860	Core i5-750	Core 2 Quad Q9550/Q9550s	Phenom II X4 965 BE
Socket	LGA1366	LGA1156	LGA1156	LGA1156	LGA775	AM3
Price (Volume)	$284	$562	$284	$196	$266 / $320	$245
Price (Street)	$280	N/A	N/A	N/A	$220 / $350	$245
TDP	130Watts	95Watts	95Watts	95Watts	95Watts / 65Watts	140Watts
Codename	Bloomfield	Lynnfield	Lynnfield	Lynnfield	Yorkfield	Deneb
QPI/HT	4.8GT/s	4.8GT/s	4.8GT/s	4.8GT/s	N/A	4GHz
Core Clock	2.66GHz	2.93GHz	2.8GHz	2.66GHz	2.83GHz	3.4GHz
Turbo Boost (Max 1 Core)	2.93GHz	3.6GHz	3.46GHz	3.2GHz	N/A	N/A
HyperThreading?	Yes	Yes	Yes	No	No	N/A
Cores/Threads	4/8	4/8	4/8	4/4	4/4	4/4
L1 Cache	256KB	256KB	256KB	256KB	256KB	512KB
L2 Cache	1MB	1MB	1MB	1MB	12MB	2MB
L3 Cache	8MB	8MB	8MB	8MB	N/A	6MB
Die Size (mm^2)	263	296	296	296	214	258
Transistor Count (million)	731	774	774	771	820	758
Process (nm)	45	45	45	45	45	45

AMD 3.4GHz Phenom II X4 965 BE

It’s bad news for AMD’s recently released quad-core. Its best and brightest could never compete with the Core i7-920 in performance, but it certainly outgunned it in price. Now with Intel’s LGA1156 Lynnfields here, it can’t even compete on price. As of this writing, the Phenom II X4 965 Black Edition had a list price of $245. The Core i5-750 has a list price of $200 and it mercilessly punts the Phenom II X4 965 BE up and down the field. Only in the mostly-single threaded Photoshop CS3 and POV Ray 3.7 did the Phenom’s 700MHz advantage put it over the top. Against the Hyper-Threaded Core i7’s though, the virtual cores vaulted the Nehalems over the Phenom II X4 965 everywhere. The good news for the Phenom II X4 965 is that it isn’t dead ass last. That falls to its old nemesis, the 2.83GHz Core 2 Quad Q9550. There, the Phenom II X4 965 BE serves space-cold revenge to the only chip here not to sport an on-die memory controller. Yeah, so you Intel fan boys say so what? Core 2 is an end of life CPU big whoop, but you gotta take your victories as you can get them.

The real bad news for AMD is that it’s not expected to get its next-generation Bulldozer core out now until 2011 so it can build it on a 32nm process. With Core i5 pushing the $200 range and Core i3 around the corner and expected to push into the ultra-budget range, there’s going to be very little maneuvering room for AMD’s rather dated design.

Still, there is place for the Phenom II X4 965 BE: existing upgrades. The processor will drop into a large assortment of existing AM2+ boards (you should check your mobo maker’s web site first for support) and even though it’s slower than a Core i5/i7, it’s a hell of a lot cheaper and easier to remove your current Athlon 64 X2 part and drop in the Phenom II X4 965 BE. Once you do that, you can thumb your nose at the Intel boys who have to replace their LGA775 boards and likely do an OS reinstall to upgrade to Core i5 and Core i7. Heck, there’s even rumors of a six-core AM3 part which would likely drop into modern AM2+ board so there’s some solace for AMD fans even if the performance isn’t there.

Intel 2.83GHz Core 2 Quad Q9550

We’ve said for months that Core 2 was a lame duck and it gets even lamer with the introduction of the LGA1156 parts. The Core 2 can’t compete with its Nehalem brethren on any front. Even the lowly – and cheaper Core i5-750 – gives the Core 2 Quad Q9550 such a bad beat down, that AMD would likely feel bad for it. And as we said, even the Phenom II X4 965 BE mostly smokes the Core 2 Quad. Sure, the Core 2 Quad surprisingly outperforms the Phenom II X4 965 BE in a couple of places, but it’s still the loser. We could have reached for the highest bin Core 2 Quad, the 3GHz Q9650, but at $319 on the street and list, it makes no sense.

Hell, even at the street price of $220, it’s hard to justify the Core 2 Quad Q9550 over a new Core i5 rig. Even AMD’s Phenom II has a better roadmap as the company will support it through at least through 2010 with newer, faster CPUs. The same can’t be said of Intel which is unlikely to introduce faster Core 2 parts. There is still a place for Core 2 though: If your box is rolling a dual-core and the LGA775 board supports 45nm quads, it would be dumb not to get one more upgrade out of it. Outside of that, it’s clear Core 2’s glory days are long, long gone.

Intel 2.66GHz Core i5-750

Of the troika of new chips, Intel’s Core i5-750 is likely to be the big hit. With wholesale prices of $200, the average joe or jane can now build a Nehalem-based system for almost as much as a Core 2 or Phenom II box. The key price differentiator is the motherboards. Board’s we’ve seen will likely top out about $225 but many will dip into the $150 range. That makes it significantly different from when Intel’s Core i7-920 launched. Sure it was only $280 (and still a year later) but the boards for the processor all pushed the $300 mark. Even today, it’s hard to get an LGA1366 board for $250.

In performance, the lower clocked Core i5 schools Core 2 Quad Q9550 as well as the Phenom II X4 BE. The Phenom II X4’s massive clock difference actually gets close to the Core i5 in a few benchmarks and surpasses it in two but the cheap Lynnfield is clearly the winner. Against the Core i7, as expected, the Core i5 loses but for many, giving up 10 percent to 25 percent in performance is worth it to save, oh, $300 over the price of the top-end Core i7-870. The performance delta between the i5 and i7 is the greatest in multi-threaded apps. In gaming and apps not optimized for quad cores, they’re a lot closer.

In the final analysis, the Core i5 falls right where Intel’s bewildering branding scheme puts it: faster than Core 2 and Phenom II but slower than i7 and really damned cheap to boot too. Buy this chip is the vast majority of your applications are not optimized for quad-core or more. If, however, your work counts on an increased amount of cores, go the extra mileage for a Core i7 with Hyper-Threading.

Intel 2.8GHz Core i7-860

If you use the Goldilocks formula, this new quad-core, Hyper-Threaded CPU is “just right.” It’s just the right blend of performance to price. At $266 for a 2.83GHz, its most natural enemy is the Core i7-920 in LGA1366. The 920’s primary advantage is increased bandwidth thanks to its tri-channel DDR3 support. But since the vast majority of apps really don’t need that much bandwidth, the 860 is generally faster by five percent thanks to its 5 percent clock advantage. That’s pretty much what we saw in our tests too – about five percent difference in the vast majority of the tests in favor of the 2.8GHz 860. And of course, the 860 also gets a boost from its increased Turbo Boost clocks. A stock 920 will only Turbo as high as 2.93GHz under some loads. The 860 will max out at 3.46GHz under some loads. Why is the Turbo higher on the 860? The 920 was a first generation Nehalem processor and Intel played it very conservatively with the Turbo. With the 860, the company feels more comfortable pushing it higher.

So what should you buy? Both procs are priced the same at $266, but each has its strengths. The 920 has true dual x16 PCI-E 2.0 support thanks to the X58 chipset and discrete PCI-E as well as tri-channel DDR3. Frankly, we think both of those features are pretty minor reasons to choose 920. The main reason to pick a 920 is the upgrade path. In a few months, Intel will introduce a 32nm-based six-coire processor codenamed Gulftown. With Hyper-Threading, you get 12 threads as well as any goodness Intel can squeeze out of the new 32nm process (both the 860 and 920 are 45nm parts). A rig built on 860 will also not see Gulftown and probably will not see a six-core part for quite some time if ever. Again, Intel intends for LGA1156 to be for the mainstream and the 99 percent of mainstream users don’t need 12-threads for their apps nor will they pay out $1,000 for a Gulftown CPU.

That takes us back to the Core i7-860 part. If you save possibly $150 to $200 by building an 860 rig, that can be put toward a faster GPU, a bigger hard driver or a larger monitor. In this economic climate, that’s pretty appealing and really the pick of the litter if you ask us.

Intel 2.93GHz Core i7-870

In this comparison of five CPUs, the 870 was head and shoulders the fastest CPU. Generally, we saw a 5 percent advantage in the 870 – just what you would expect from a 5 percent clock advantage. We did see some interesting results though. The multi-threaded Cinebench 10 actually saw the 870 with a 13 percent higher score. World in Conflict also saw the 870 run away with a 24 percent higher frame rate than the 860. So here’s the real question: Would you pay nearly 100 percent more for that? For most folks, probably not. At $555 for the Core i7-870 and $266 for the Core i7-860, it just makes a hell of a lot sense to give up some of that performance to put towards something else in your machine. And if you really are into being on the ragged edge of performance, it makes a lot more sense to jump in with both feet for an LGA1366 platform and the promise of a six-core upgrade in a few month. So we’re a little ambivalent about the Core i7-870. But make no bones about it, the Core i7-870 is the king of the CPUs that battled today, it’s just not really a “budget” CPU is it?

Intel 2.66GHz Core i7-920

If you haven’t figured it out by now, Intel’s 920 is the chip all three LGA1156 Lynnfield’s were gunning for. A huge hit among the performance budget shoppers, it overclocked well and got you into the Core i7 club without paying through the nose.

With its three new siblings here, the 920 is mostly pushed to the back in performance and pricing competitiveness. The other chips run at higher clocks, automatically Turbo Boost to higher levels and can save you from $150 to $200 in associated building costs. So why bother with the 920? There is still some magic to this old favorite. In this roundup, the 920 was generally in third or second place in the vast majority of the tests. And in some benches, the 920 interestingly bubbled up to the top. Even though its superior tri-channel doesn’t always pay off, it apparently does in Photoshop CS3 where the 920 took top honors and we suspect its superior bandwidth helps keep it in hunt elsewhere too. We had heard early rumors that Intel would axe the 920 once the LGA1156 procs were out but the company has since told us it had no plans to discontinue 920 right now. That’s a good decision too. While LGA1136 is definitely an exclusive club, there’s no reason for Intel to make it even harder to get in. If your only choice to getting an LGA1366 platform was to shell out $500 for a Core i7-950 or $1,000 for a Core i7-975 Extreme Edition, very few people would choose that route.

Of course, the biggest reason to buy 920 is for the path to the Core i9 “Gulftown” CPU that Intel will introduce in a few months. Ideally, you could build a 920 box and use it for a year or so until the Gulftown derivatives drop down enough to become “budget” chips themselves.

The Benchmarks

To run our test, we tried to balance each particular CPU’s platform as closely as possible but given the differences in each chip’s requirements that’s almost impossible. We used the same make and model Western Digital Raptor 150GB drive for all four builds, the same 64-bit Windows Vista Home Premium SP2 build, the same make/model/clocked EVGA GeForce GTX280 and the same drivers. Windows 7 was available at the time of our reviews but we felt it would be best to use Vista as drivers were readily available for it.

The only real differences, of course, were the motherboards and RAM configurations. Although the Core i7-920 does unofficially support higher clocked RAM, we stuck with the official DDR3/1066, but 6GB of it. For the dual-channel Phenom II X4 and Core i5/i7, we used 4GB of DDR3/1333. For the dual-channel Core 2 Quad, we used the far more common DDR2/1066 as the vast majority of Core 2 Quad folks are running DDR2. For motherboards, we used three Gigabyte boards: the new GA-P55-UD6 for the LGA1156 procs, a GA-EX58-UDR3 for the LGA1136, and a GA-MA790GPT-UD3H for the Phenom II. For the Core 2, we relied on an Asus Maximus II Formula.

Benchmark Test Setup

CPU	Core i7-920	Core i7-870	Core i7-860	Core i5-750	Core 2 Quad Q9550/Q9550s	Phenom II X4 965 BE
RAM Mode	Triple	Dual	Dual	Dual	Dual	Dual Unganged
RAM Size	6GB	4GB	4GB	4GB	4GB	4GB
RAM Speed	DDR3/1066	DDR3/1333	DDR3/1333	DDR3/1333	DDR2/1066	DDR3/1333
RAM Latency	7-7-7-20-1T	9-9-9-24-1T	9-9-9-24-1T	9-9-9-24-1T	5-5-5-18-2T	9-9-9-25-1T
Hard Drive	WD 150 Raptor	WD 150 Raptor	WD 150 Raptor	WD 150 Raptor	WD 150 Raptor	WD 150 Raptor
OS	Vista HP 64-bit SP2	Vista HP 64-bit SP2	Vista HP 64-bit SP2	Vista HP 64-bit SP2	Vista HP 64-bit SP2	Vista HP 64-bit SP2
GPU	Geforce GTX 295	Geforce GTX 295	Geforce GTX 295	Geforce GTX 295	Geforce GTX 295	Geforce GTX 295
GPU Clocks 3D	670/1458 /1215	670/1458 /1215	670/1458 /1215	670/1458 /1215	670/1458 /1215	670/1458 /1215
Motherboard	Gigabyte GA-MA790 FXT-UD5P	Gigabyte GA-P55 -UD3	Gigabyte GA-P55 -UD3	Gigabyte GA-P55 -UD3	Asus Maximum II Formula	Gigabyte GA-EX58 -UD3R

BENCHMARKS

CPU	Core i7-920	Core i7-870	Core i7-860	Core i5-750	Core 2 Quad Q9550/Q9550s	Phenom II X4 965 BE
MainConcept Reference 1.0	1235	1115	1170	1337	1644	1388
MainConcept Refernece 1.0 Pro AVC	696	635	664	769	986	840
Premiere Pro CS3	671	610	630	620	741	733
Cinebench 10 64-bit	16140	18275	16085	14442	12280	14083
Handbrake iPod Classic	994	939	993	1198	1366	1220
PCMark Vantage 64-bit Overall	6929	7536	7299	7208	6241	6824
POV Ray 3.7	3470	3888	3702	2773	2669	3045
Photoshop CS3	116	119	126	128	132	123
ProShow Producer	636	640	617	700	862	911
Everest 5.0 RAM Read (MB/s)	14387	12997	13641	12867	7807	8154
Everest 5.0 RAM Write (MB/s)	11639	10811	10992	9881	7085	6794
Everest 5.0 RAM Copy (MB/s)	15790	15414	15393	14684	7455	10246
Everest 5.0 RAM Latency (ns)	61	53	52	31	64	54.3
Sandra RAM Bandwidth (GB/s)	22.4	16.7	17.2	16.8	7.2	12.7
Fritz Chess Benchmark	21.4	23.94	22.38	17.38	16.97	17.04
3DMark Vantage Overall	15008	15002	14985	14947	14681	14544
3DMark Vantage GPU	12306	12231	12247	12249	12013	11978
3DMark Vantage CPU	44002	46815	45525	44066	40644	40679
Valve Particle Test	143	159	151	124	99	95
Valve Map Compilation	146	128	133	121	129	125
Crysis	146	156	150	147	119	104
Resident Evil 5 Fixed DX9	114	115	118	109	85	89
Resident Evil 5 Fixed DX10	117.7	106.5	119.7	109.4	83.8	89.2
Resident Evil 5 Variable DX9	145.8	155.9	147.2	155.4	133.7	140.3
Resident Evil 5 Variable DX10	150.4	155	157.8	160	133.9	140.2
World in Conflict	221	282	227	266	159	160
WinRar 3.20	581	588	594	706	868	805

Best scores in bold.

Next, our detailed benchmark analysis.

Benchmark Analysis

Sisoft Sandra RAM

(higher is better)

Sisoftware’s Sandra RAM benchmark lets the Core i7-920 strut its tri-channel DDR3 stuff. Keep in mind, we tested the Core i7-920 at its officially rated speed of DDR3/1066 vs. DDR3/1333 for the five other parts here. That extra bandwidth of the 920, in fact, may help it keep pace with the newer processors despite a general clock deficit. Still, as we’ve said, the vast majority of folks just don’t need that bandwidth.

WinRar 3.20

(lower is better)

Now that we’ve said don’t pay attention to bandwidth, here’s a test that does actually show how much it can help. We used WinRar 3.20 to compress a folder of RAW files shot with a Canon EOS 5D. The files, pretty much, are near uncompressible so we thought it would be a good way to stretch the processors. The Core i7-920 just edges out the Core i7-870 despite a 266MHz gap between the two chips.

World in Conflict

(higher is better)

We ran World In Conflict at low resolutions to take the GPU out of the equation for performance. Some have questioned this but it tells you what you would get if you had the fastest GPU in ze vorld. The results were a bit odd but we saws the Core i7-870 up front and then the budget Core i5-750 next in line. Huh? Why would the results be so whacky? We have two theories: Turbo Boost can play havoc with unpredictability. Afterall, it automatically overclocks individual cores for a given thermal load and power consumption load so it’s possible the Turbo Boost for the 750 just kicked up into high gear in the most single-threaded game. Our other theory is that the oddly super low latency that the 750 exhibitied may have contributed to its scores. The other conclusion you can draw from this is that the Phenom II and Core 2 Quad should have just stayed off the field.

Crysis

(higher is better)

We used the same low quality and low resolution mode for our Crysis runs to show you just what kind of frame rates you’ll get in four years when the newest GPUs finally make Crysis its bitch. If you happened to still be running the same rig in 2013, here’s how your CPU would shape up. Unlike some of the multi-threaded benchmarks, we actually saw very predictable results with the most expensive 870 on top, the budget builder’s favorite, the 860, in second, and the 750 and 920 pretty even.

Valve Particle Test

(higher is better)

Valve’s Particle Test is a quad-core optimized test that measures CPU performance. Here, our results were predictable with the Lynnfields seemingly scaling with clock speeds. The surprise again is how well the 920 does which can only be attributed to its third-channel of DDR3.

Proshow Producer

(lower is better)

Our Proshow Producer benchmark hammers all cores available and again, there’s that pesky Turbo Boost messing with us. The Core i7-870 should have been the fastest, but the Core i7-860 just edges past all contenders. If you’ve wondered if Hyper-Threading pays off? Yes. Notice the roughly 10 percent gain the 2.66GHz 920 has against the 2.66GHz 750. That’s generally what we’ve seen from Hyper-Threading in other benchmarks. One other conclusion: Core 2 Quad and Phenom II X4, why are you here again?

AMD Releases New Flagship Processor: Phenom II X4 965 Black Edition

Paul Lilly — Thu, 13 Aug 2009 08:30:00 -0500

AMD today adds to its Phenom processor line with a new flagship part, the Phenom II X4 965 Black Edition. The new chip updates the company's Dragon platform, which combines a Phenom II CPU with an ATI Radeon 4870/4890 graphics card and 790GX-based motherboard.

We're told the silicon in the X4 965 BE is unchanged from the X4 955 BE, so you're essentially looking at a clockspeed bump with a slightly higher TDP. Specifically, the 45nm chip comes clocked at 3.4GHz and contains 6MB of L3 cache, and 8MB of total cache (2MB total L2 per processor). And because it's a 'Black Edition' part, the new CPU is unlocked.

AMD also tells us that its own internal testing has shown the X4 965 BE to be a better overclocker than the previous 955, which isn't always the case when releasing a faster-clocked processor built on the same architecture. We currently have one of these chips in our Lab, so look for our own performance impression in the very near future.

Best of all for the AMD faithful, AMD has set an MSRP of $245 for the 965, the same official MSRP the 955 previously.

Image Credit: AMD

Socket AM3 Arrives -- AMD Releases Five New CPUs that Support DDR3

Gordon Mah Ung — Mon, 09 Feb 2009 12:00:00 -0600

Who says AMD moves too slowly? Just a month after releasing its well regarded Phenom II mid-range CPUs, the company is back with no fewer than five new P-II chips and its new AM3 socket that support DDR3. We give you the skinny on AMD’s latest quad and tri-cores and help you sort through AMD’s bewildering array of CPU choices.

War. What is it good for? Absolutely nothing. Well, except when it’s a CPU war. In that case, it’s good for consumers. Really good for us. With the unveiling of five new AMD’s latest Phenom II CPUs supporting DDR3, it’s pretty clear that the CPU war that started with the unveiling of the Phenom II in January is escalating.

AMD’s new lineup includes the 2.6GHz Phenom II X4 for $175, the 2.8GHz Phenom II X3 720 Black Edition at $145, and the 2.6GHz Phenom II X3 710 for $125. AMD’s two other new chips: the 2.6GHz Phenom II X4 910 and the 2.5GHz Phenom II X4 805. The 910 and 805 are OEM only CPUs and pricing was not released but you can expect that gray-markets will carry them and that the prices will follow the numbers. The 805, for example, should be slightly cheaper than the $175 810 and the 910 should be cheaper than the $195 Phenom II X4 920.

Lost in the numbers? So where we. AMD’s lineup is so bewildering to us today that we had build a spread sheet just to sort it out!

There are five prominent things to note with the new CPUs: They all support AM3, the HyperTransport speeds are higher, the L3 cache size is different, the tri-cores are back and the thermals are lower. We’ll address these in order.

AM3 socket

AM3 is AMD’s new socket standard that is built to support DDR3. The good news is that AM3 CPUs feature both DDR2 and DDR3 controllers. This means you can install an AM3 CPU in an AM2+ (and even some AM2) motherboards. You cannot, however, install an AM2+ CPU in an AM3 board. To prevent damage, the AM3 sockets have two fewer pins so you can’t even physically insert an AM2+ CPU in the socket. All Phenom II CPUs except for the two original launch CPUs should be AM3-based.

Can you spot the two fewer pins on the AM3 CPU on the right vs. the AM2+ CPU on the left?

Why didn’t AMD make the two original chips (the Phenom II X4 940 and 920) AM3 too? The company said it wanted to get them out as soon as possible and ditching AM3 support a cut quite a bit of engineering time off. We can understand that but it’s only been a month since the 940 and 920 were unveiled so couldn’t it have waited just a little longer so as not to confuse the hell out f people and piss off 940 and 920 owners? Apparently not. In fact, we’ve been told by AMD officials that the 940 and 920 actually had the AM3 controllers in them but they not turned on. If we ever get the time, we’ll have to snip the pins off a Socket AM2+ 920 and see what happens when it’s inserted into an AM3 motherboard.

HyperTransport Speeds

The newest Phenom II all sport 4GHz HyperTransport speeds. The original Phenom II X4 940 and 920 only ran at 3.6GHz HyperTransport speeds. Why? Again, AMD said it need to cut a few corners to get the 940 and 920 out as soon as possible and limiting the speeds to 3.6GHz help it do that. The company notes that it’s not like the CPUs are saturating the bandwidth anyway so it should have no real impact on performance. We, frankly, haven’t noticed that much of a difference either and we have to note that only the original Phenom X4 9950 and Phenom X4 8850 sported the 4GHz HyperTransport speeds. All others were lower.

If you’re really interested in how AMD’s Phenom’s look from an HT speed context here’s our chart sorted by HyperTransport speeds:

Cache

AMD is now also resorting to smaller cache versions to differentiate its models. This is an old technique long used by Intel and AMD and helps maximize the yields. If a CPU has a bit of defective cache, Intel or AMD turn off that portion and sell it as a lower model which is why these new chips often have the same die size as the larger L2/L3 chips. As the process matures and the yields get good enough that all of the cache is good, the companies have been known to actually produce smaller cache versions to even further maximize the yield. Generally, the largest cache models cost the most. For AMD’s Phenom II lineup, a 9XX denotes the larger 6MB L3, while an 8XX denotes 4MB of L3 cache.

Again, here’s the view of AMD’s Phenom CPUs from an L3 cache perspective:

Tri-cores

Like the cache, AMD is also maximizing its yields by taking quad-core procs that have one bad core and selling them as tri-cores. Initially, enthusiasts scoffed at the idea of a quad-core minus one, but they’ve gradually been accepted. AMD has also had some success by putting the tri-cores against Intel’s dual cores. For the most part, three execution cores will indeed give you better performance in multi-threaded applications and multi-tasking than dual-cores. Here’s a break down of the X3’s by clock speed.

TDP

The final difference with these new chips is the thermal or TDP ratings. The original Phenom II X4 940 and 920 both had enthusiast-class TDP ratings of 125 watts. As more mainstream parts, all five new CPUs run are rated to disperse about 95 watts of heat under full load. It appears that AMD is now pushing 125 watts as its maximum TDP for desktop parts. Only one chip, the original Phenom X4 9950, hits 140 watts. All others are 125 watts or lower. Again, here’s a view from the TDP perspective.

We can’t touch on this CPU launch without tackling the big question: DDR3.

And you can’t tackle DDR3 without openly wondering why the hell is AMD so slow in adopting new memory standards? We find this to be especially ironic because it was AMD that made DDR what it is today.

Years ago, Intel decided that a fast, serialized, memory standard would benefit PCs. The solution to this was Rambus’ Direct RDRAM memory. Intel bet it all on RDRAM and restricted the Pentium 4’s chipsets to Direct RDRAM only. The problem is the RAM maker’s didn’t want RDRAM due to the licensing fees that they would have to pay Rambus. But what could they do if Intel made the CPUs and chipsets for it?

AMD saw it’s opening and led memory makers in a mutiny against Intel by supporting DDR with Athlon. The mutiny was successful, Direct RDRAM was tossed under the bus and Intel embraced DDR with a bear hug. Although we now believe the industry and the media (including Maximum PC) made a mistake by not moving to Direct RDRAM, or at least, something similar to it, DDR was the standard.

So how the hell did AMD turn from the darling of the memory industry into a perceived drag ass? DDR2 was adopted by Intel two years before AMD introduced AM2. And Intel’s DDR3 chipset has been around since late 2007.

We’ve long had a pet theory that moving the memory controller into the CPU has taken some flexibility out of memory choices. The original Athlon 64 was hard wired to only run DDR. Likewise, AM2 Athlon 64s and Phenoms could only run DDR2. AMD’s solution to supporting both DDR2 and DDR3 is to build a memory controller that supports both types of RAM into the AM3 procs.

AMD’s official explanation for its seemingly slow memory updates is that it only adopts new memory standards when its cost effective and when people actually want it. Hence, even with AM3, the company is still pooh poohing DDR3. AMD says it still believes the vast majority would rather have the cost savings of DDR2 over DDR3.

DDR2’s cost performance over DDR3 isn’t what it once was though. A year ago, 2GB of DDR3 would fetch several hundred dollars. Today, you can buy 4GB of Crucial DDR3/1066 for $79. Change your selection to 4GB of Crucial DDR2/800 and it would cost you $38. So DDR2 does cost 100 percent more, but we’re talking 80 bucks here folks for 4GB of RAM.

Finally there’s price.

AMD’s pricing on its CPUs has been a great deal for consumers. It is a bit confusing though. At first glance, you’d think it was 22 CPUs spread out in a price band from $225 to $101. But two of those chips are OEM only. Another four are aimed at servers and media center PCs and another four are for business desktops.

Performance

How we tested

We used an MSI DKA790GX board outfitted with 4GB of Patriot DDR2/1066, a PC Power and Cooling 1200 Watt PSU, a GeForce 8800GTX and 150GB WD Raptor drive to test the three AMD procs. As a comparison, we used a 2.83GHz Core 2 Quad Q9550 in a Gigabyte GA-X48-GQ6 with 4GB of DDR3/1333 with a WD Raptor 150 and GeForce 8800GTX. For the Core i7-920, we used an Intel DX58SO with 3GB of DDR3/1066, GeForce 8800GTX and WD Raptor 150 drive. All configurations used Windows Vista Home Premium in 64-bit flavor.

The results were mostly what we expected. Performance against its sibling, the 3GHz Phenom II X4 940 was what you would think a CPU with 400MHz fewer clock cycles would score. We did see some unexpected results though. The AM3 Phenom II X4 810 part slightly outscored the Phenom II X4 940 in several of the memory benchmarks. We didn’t expect this given that we were running it in the same board with the same RAM and with same RAM speeds and timing set. This is either a hiccup in our test or an errant setting from our earlier test of the 940. We unfortunately didn’t have time to go back and rerun our tests with the 940. It is also possible that AMD has taken the extra few months it had with the newer AM3 parts to tweak the memory controller.

Between the Intel and AMD chips there was no comparison but that’s no surprise. AMD doesn’t expect the 810 to take on the 2.83GHz Core 2 Quad Q9550. Instead, AMD believes the CPU is better matched against the budget 2.33GHz Core 2 Quad Q8200 part. The $163 Q8200 has 2MB less L2 cache than the Q9550 and runs about 500MHz slower. We didn’t have Intel’s ultra budget part handy to test but subtract 500MHz from the Q9550’s scores and take away a small bit for the cache and both parts are likely competitive with each other.

The $284 Core i7-920, of course, is the fastest of the bunch but it’s also more expensive to buy and build a machine around.
Realistically, this comes down to Phenom II vs. Core 2. There, it’s a competitive crowd as Intel has as many or more CPUs than AMD does. There is only one advantage an AMD builder would have over an Intel machine: future upgrades.

Intel is pretty set to push the superior performing Core i7 as the platform of the future and is unlikely to spend the money and engineering to say, qualify a 3.5GHz CPU for the Core 2 platform. AMD, on the other hand, is committed to AM3 for now. That means it’s possible we’ll see a 3.4GHz or 3.6GHz Phenom II down the road. And even if that chip comes out in AM3 the backwards compatibility with AM2+ means those people won’t get left behind either.

While Core 2 Quad certainly has some legs left in it, those with an eye towards future upgrades should look to Core i7 if they want performance. And if they just want a good performing budget chip, AMD’s Phenom II is actually looking like a more stable platform over Core 2 right now. We’re not at all saying that Core 2 is dead, especially since in many ways, it still far outperforms all of AMD’s CPUs, but in six or nine months, Core 2 will feel stale as only Core i7 and Phenom II will get performance upgrades.

3DMark Drama: Intel Employee Claims AMD Never Held The Top Score

Justin Kerr — Sun, 01 Feb 2009 19:31:04 -0600

It’s no secret that we here at Maximum PC are fans of Intel’s new Core i7. In fact, Intel has held a place of distinction in our best of the best round up pretty consistently now ever since Athelon’s day came and went several years ago. Despite this fact, we are pretty fickle with our affections, and are all secretly still rooting for the underdog. We are also the first to admit that we are glad AMD is still around to keep Intel on its toes. Though both Phenom & Phenom II failed to set the world on fire, we were all pretty impressed when we discovered how much overclocking headroom we received as a result of the die shrink. We were even more excited when we saw the videos of AMD pushing the new CPU past 6.5Ghz, setting a new record in terms of clock speed.

Intel however, never wanting to concede its speed crown, was quick to go on the attack. In an email exchange with TGDaily, an Intel employee pointed out that the AMD 3DMark score of 45,474 submitted on January 12^th 2009 was actually 1,170 points lower than a Core i7 score turned in by Intel just 8 days earlier. He also stated that the AMD results were achieved with unapproved drivers, and curiously were only run when the clock speed was at 4.481 Ghz. So as for who holds the 3DMark speed crown, I guess it all depends on who you ask.

It’s good to know that even if Phenom II didn’t quite bring them up to where they need to be, at least they have Intel taking notice of them again. And I for one can’t wait until I see the portable liquid helium cooling system that lets me duplicate these AMD scores at home! They are working on that right?

Overclocked AMD Phenom II Hits 6.5GHz with Liquid Nitrogen/Helium Cooling

Andy Salisbury — Mon, 26 Jan 2009 16:01:30 -0600

In the past, the AMD Phenom II has been overclocked to an extremely impressive 5 GHz. And while this was extremely impressive, it would seem that AMD wouldn’t want to be outdone.

AMD’s own Pete Hardman and Sami Makinen were able to overclock an AMD Phenom II to a blazing 6.5GHz, at an operating temperature of –230 degrees Celsius using liquid nitrogen and liquid helium as their cooling agents.

Should you be interested in seeing the whole process play out, be sure and check out the video here (and, as is usual with videos of this nature, prepare your ears for some awful trance music).

Image Credit: AMD

AMD Already Dropping Prices of Recently Released Phenom II Processors

Paul Lilly — Wed, 21 Jan 2009 08:38:51 -0600

Well that was fast. Just two weeks after launch, AMD has significantly cut pricing on its Phenom II X4 processors. And the chips weren't that expensive to begin with. The Phenom II X4 920 and 940 debuted at $235 and $275 (per 1,000-unit trays) respectively, meaning you could pick up the company's flagship CPU for under $300 when normally a top of the line processor commands a grand. Now the 920 and 940 are being sold on Newegg for $195 and $235 respectively.

While it might seem AMD is being a bit hasty in slashing prices, you can chalk it up to competitive pressure from Intel, who could care less about AMD's financial situation. Not only is Core i7 in no danger of conceding its performance crown, but two days ago Intel announced price cuts on its Core 2 Quad Q9650, Q9550, and Q8200 processors, which now sell on Newegg for $334, $283, and $170 respectively.

No BS Podcast #92: The Long Walk Home Edition

The Maximum PC Staff — Fri, 16 Jan 2009 14:47:01 -0600

Is there anybody out there? We're back for our first podcast of 2009 with a lively discussion of CES, Windows 7, Phenom II, the skiing conditions in Utah, and a whole lot more. What exactly is a whole lot more? Well, pretty much it's reader questions and Gordon's rant. (The rant is powered by three weeks of bottled-up rage though.

Do you have a tech question? A comment? A tale of technological triumph? Just need to get something off your chest? A secret to share? Email us at maximumpcpodcast@gmail.com or call our 24-hour No BS Podcast hotline at 877.404.1337 x1337--operators are standing by.

Subscribe: http://feeds.feedburner.com/maximumpc/1337

AMD Strikes Back with Phenom II -- Full Analysis and First Benchmarks!

Gordon Mah Ung — Thu, 08 Jan 2009 11:00:00 -0600

The production of a sequel typically implies that the original creation is worth revisiting. However, considering that the original Phenom was the hardware version of Ishtar, many enthusiasts didn’t think Phenom deserved to be revisited.

AMD certainly thinks it does—and it hopes Phenom II is Star Trek II: The Wrath of Kahn to Phenom’s Star Trek: The Motion Picture. And why shouldn’t AMD be able to pull off a reversal of fortune? Phenom II isn’t just Phenom joined by a Roman numeral—it’s a die shrink with a boatload of additional cache and an improved core. In short, AMD hopes to erase memories of the original Phenom and put smiles on the faces of disappointed overclockers with its reimagined Phenom II chip.

Come with us as we review, critique, and dissect Phenom II and find out how it stacks up against a stack of Intel CPUs.

Phenom Reimagined

AMD’s trip back to the drawing board

The Phenom launch certainly didn’t go as AMD had planned. Rather than christening a new line that would change the company’s fortunes, AMD CEO Hector Ruiz broke a bottle of champagne over the bow of a ship that promptly sank under the waves—but only after smashing into a nearby pier with a bait shop and a busload of tourists on it: Phenom was a year late and had a performance-crippling TLB bug, yield issues, and a performance gap with Intel’s older generation of CPUs.

Fast-forward a year and the picture looks far different for the underdog chipmaker. Phenom II is actually ahead of schedule. And doubts about overclocking were quashed months ago when the company invited elite overclockers to its headquarters to get medieval on the new chip with liquid nitrogen and other exotic toys. The result? Overclocking feats beyond 5GHz.

Not to belabor the sequel talk, but it’s clear that AMD doesn’t intend for its pair of new Phenom II chips to be cheesy follow-up. These CPUs are intended to erase all doubts that the original chip created and help quell uneasiness about the company’s ability to make good parts.

The Dynamic Duo

The Phenom II family consists of two CPUs: the 2.8GHz Phenom II X4 920 and the 3GHz Phenom X4 940 Black Edition. Both use the company’s new 45nm process and can be paired with the majority of Socket AM2+ boards (and even some AM2 boards.) Both CPUs are native quad-core designs with all four execution cores residing on a monolithic die. AMD will continue its practice of repackaging defective quad-core dies as tri-cores (denoted with X3 rather than X4).

New under the Hood

For the most part, Phenom II isn’t a radical departure from Phenom. It has the same basic core and still features an integrated memory controller and HyperTransport connections for chip-to-chip connections. The update does include a few substantial changes, however. The biggest is the move to a 45nm process, which significantly shrinks the size of the chip and results in better yields; additionally, the 45nm-based Phenom II has 758 million transistors but is only 258mm2. The original 65nm Phenom has 450 million transistors and measures 285mm2.

By shrinking the die, AMD is able to use some of the freed up real estate for more cache. While the L1 and L2 remain unchanged, the L3 goes from 2MB in Phenom to 6MB in Phenom II. This larger cache is also slightly faster than the 65nm Phenom’s.

In other good news for enthusiasts, the new chip includes both a DDR2 and a DDR3 integrated memory controller. The bad news is that the first two Phenom II chips will support only DDR2; both DDR2 and DDR3 will be supported with its AM3 revision of Phenom II, which will be released in the next few months.

So why release a version of Phenom II that is limited to DDR2? AMD didn’t want to wait the additional months it would have taken to validate the CPUs for both newer DDR3 boards and DDR2 boards. The company felt that to have a Phenom II that runs at decent clock speeds, overclocks like crazy, and drops into existing boards is just a better way to prove its back on track.

More importantly, AMD doesn’t think people are that hot for DDR3 right now due to its premium price. To some extent, AMD is right: Two 2GB modules of DDR2/800 will set you back just $28, while a pair of 2GB DDR3/1333 modules costs about $100. However, true sticker shock sets in at the highest speeds: 4GB of DDR3/1600 costs about $300 and 4GB of DDR3/2000 will set you back about $400.

We would have preferred it if AMD had introduced one CPU that would work with both types of memory, but we understand that due to its position in the market, it simply doesn’t have the luxury of waiting three months to get Phenom II to work with both new DDR3 boards and the older DDR2 infrastructure.

But all you really want to know is whether Phenom II will work with your board, right? Minus the missteps with the original Socket 940 and Socket 754 nonsense (well, and QuadFX), AMD has worked hard to ensure that CPU swapouts won’t cause havoc. Phenom II will work in almost every board that supports the original Phenom CPU, with the only caveat being boards not designed to handle CPUs hotter than 95 watts. Since both Phenom II CPUs are 125 TDP chips, they likely will not work with those boards.

Cooler than Ever

While new manufacturing doesn’t always lead to more efficient parts, this die shrink certainly seems to have helped AMD with thermals. For example, the 65nm-based 2.6GHz Phenom X4 9950 BE had a thermal design power rating of 140 watts, while the 45nm-based 3GHz Phenom II X4 940 has a TDP of 125 watts.

AMD also seems to have finally shed the “cold bug” that frustrated extreme overclockers. The original Phenom would overclock to a certain level on air, but when extreme cooling techniques were applied, it wouldn’t overclock any further. While cold temperatures aren’t a cure-all, most CPUs offer additional headroom at -150 F. But the original Phenom simply hit a wall and no amount of cooling would allow for additional overclocking. AMD set out to prove it fixed this issue in Phenom II by hosting private demos for a group of extremes overclockers. Apparently, no one left the demo unhappy.

Platform Shmatform

Every PC is essentially a CPU, a chipset, a GPU, and storage, so you may be confused when you hear the word “platform” thrown around like it’s some new type of technology. It’s not. It’s an artificial way Intel and AMD brand a set of components. For Intel, Centrino is simply the combination of the CPU, chipset, and a Wi-Fi chip. Laptops sold without those three key Intel ingredients are not allowed to use the Centrino sticker. Since Intel advertises the hell out of Centrino, not Core 2 Duo Mobile, most OEMs feel compelled to buy all three parts from Intel.

AMD is not being as Machiavellian with its platform (at least not today), but it is doing some branding around a Dragon theme. Dragon is a combination of the Phenom II, an ATI 790GX or ATI 790FX chipset, and a 4000-series Radeon HD GPU. Does this mean that you can’t use a GeForce GTX 295 with Phenom II? No. Everything is as it was before—you can probably even use the Phenom II 940 in some older AM2 boards with the original Nvidia 590 SLI chipset.

So why bother to push all this platform hooey? Today, it’s just a marketing gimmick, but tomorrow it may be far more meaningful. With the functionality of the chipset, CPU, and GPU morphing together, this collection of hardware may indeed be a platform that you buy in a few years. That’s one thing AMD likes to toot its horn about: Intel has CPUs and chipsets and Nvidia has GPUs and chipsets, but only AMD has all three ingredients.

Price Matters

CPU companies like to use mysterious model numbers that don’t tell you a damn thing about how their chips actually perform. One quick and dirty way to see what the company thinks of a particular chip is to look at its price. AMD’s pricing of Phenom II reveals where the company thinks the CPU will compete. For example, the current king of the hill, the Core i7-965 Extreme Edition, is priced at $999. AMD has priced the Phenom II X4 940 at $275, so you can see where the company expects the CPU to fall—it’s clearly not intended to take on Intel at the high end. AMD, however, thinks there’s plenty of room to compete in the midrange against Intel’s large stable of Core 2 Duo and Core 2 Quad parts.

We put the top-end Phenom II X4 940 against Intel’s top-end Core i7 part, the still-shipping top-end Core 2 Extreme Edition part, as well as a lineup of budget Intel and AMD CPUs. The upshot is that AMD fans can take Phenom II as a sign that the company has some magic left. While Phenom was Detroit Lions bad, Phenom II is maybe Oakland Raiders or Green Bay Packers bad. Yeah, it was an ugly season, but you can tell the team is on the right track.

Our Testing Method

For our Phenom II showdown, we used a 3GHz Phenom II X4 940 BE on an MSI DKA790GX board. AMD partisans pitched a fit when we conducted our Core i7 tests with the AMD Phenom X4 9950 BE using “just” DDR2/800 RAM—they believed it was a travesty that we didn’t run DDR2/1066. Truth is, the performance difference between DDR2/800 to DDR2/1066 is minimal. In fact, after we published our Core i7 tests we spoke with AMD representatives, who agreed that the small difference in memory bandwidth had virtually no impact on the beatdown Core i7 gave Phenom.

To keep the peanut gallery happy, we tested the Phenom II X4 940 BE with 4GB of DDR2/1066. For comparison, we used a 3.2GHz Core i7-965 Extreme Edition and a 3.2GHz Core 2 Extreme Edition QX9770. We downclocked these parts to simulate the performance of a 2.66GHz Core i7-920 and a 2.83GHz Core 2 Quad Q9550, respectively. We also included the 2.6GHz Phenom 9950 X4 BE in our tests.

For all the test runs, we used the same GeForce 8800 GTX card and Western Digital Raptor 150 hard drive. The Core 2, Phenom and Phenom II rigs featured 4GB of RAM, while the Core i7 machines had just 3GB of RAM. All tests were conducted using the 64-bit version of Microsoft Windows Vista Home Premium.

Our benchmarks reflect various levels of multithread rendering, video editing, encoding, and 3D rendering. Nvidia likes to say that quad-core CPUs are unimportant, but we’re finding a very strong and fast move by application vendors to support quad core where it’s needed. We didn’t feature any dual cores in our tests because they simply can’t compete against these opponents. However, the majority of today’s games exploit two cores at best, so to eliminate graphics as a bottleneck, we ran all of the games at very low resolutions, with all the eye candy turned off. We also ran a set of synthetic memory and scientific and application workload tests to get a balanced picture of how well these quad-cores perform.

Analysis

If you’re an AMD fanboy expecting Phenom II to put its bootprint on the hind end of Core i7—any Core i7—prepare to be disappointed. The slowest 2.66GHz Core i7 920 beat the Phenom II by double digits in most of our tests. We saw differences from 11 percent to 27 percent in encoding, and in our WinRar test, the Core i7-920 was 35 percent faster. It wasn’t all bad news for Phenom II though. The chip won the ScienceMark 2.0, Quake 4, and PC Mark Vantage tests and eked out a win in the Valve map compilation test. However, we’re still calling this competition for the i7 920. Of course, the 920’s big brother, the 965 Extreme Edition, completely walked away from the Phenom II. AMD, however, isn’t concerned that its $275 chip can’t beat a $999 one—the company isn’t competing at the top end of the market. And even though the 920 is about $300, the price of a new i7 motherboard ($250) and three pieces of required DDR3 ($150) nullifies any performance benefit the i7 has, AMD claims.

AMD is far more interested in how Phenom II does against a Core 2 Quad. The Phenom II actually outscored the Core 2 Quad in our MainConcept encoding test, our ProShow Producer slideshow creation test, and Quake 4, and it just about broke even in our WinRar file compression test. The Core 2 Quad hit back in both 3DMark tests, Premiere Pro CS3, Photoshop CS3, and both of our Valve multithreading tests. Although the Phenom has a 167MHz advantage, we’d have to call this one a tie.

This again comes down to perspective. Intel fanboys can say, “Been there, done that” since AMD’s best CPU just barely pulls even with a chip family Intel introduced more than a year ago. But from AMD’s perspective, the Phenom II is a big deal. With a down economy, the company believes that people will be looking for performance on a budget, and if Phenom II supplies that without the need for a pricey new motherboard, it’s won half the battle.

Benchmarks
Model	3GHz Phenom II X4 940 Black Edition	2.6GHzPhenom X4 9950 Black Edition	2.83GHz Core 2 Quad Q9550	2.67GHz Core i7-920	2.93GHz Core i7-940	3.2GHz Core i7-965 Extreme	3.2GHz Core 2 Extreme QX9770
MainConcept (min:sec)	1,569	1,867	1,660	1,300	1,190	958	1,489
MainConcept Pro (min:sec)	942	1124	988	741	679	608	889
ProShow Producer 3.1 (min:sec)	802	1210	918	670	616	619	772
Premiere Pro CS3 (min:sec)	841	987	771	759	701	617	686
Photoshop CS3 (min:sec)	142	168	124	125	123	110	115
Cinebench 10 32-bit	99,791	8,179	10,837	12,632	13,793	15,398	12,175
Cinebench 10 64-bit	12,049	10,431	12,288	15,217	16,651	18,963	13,849
Valve Map Compilation (min:sec)	143	167	130	152	141	125	116
ScienceMark Overall	1,903	1609	1,,716	1,710	1,885	2,091	1,920
ScienceMark Membench	9,198	7,279	7,105	12,737	13,028	13,312	8,560
PCMark Vantage x64 Overall	6,447	5,724	5,945	6,616	6,767	7,510	6,423
PCMark Vantage Overall	6,085	5,299	5,460	5,347	6,043	6,705	5,961
Sisoft Sandra RAM Bandwidth (GB/s)	11.69	9.73	6.9	18.07	18.09	18.15	7.4
Sisoft Sandra RAM Latency (ns)	97	95	81	79	78	77	79
Everest Ultimate MEM Read (MB)	7,716	6,701	8,006	14,449	14,841	15,167	8,252
Everest Ultimate MEM Write (MB)	6,085	4,856	7,075	11,627	14,788	12,041	8,490
Everest Ultimate MEM Copy (MB)	9,734	7,760	7,334	15,039	15,011	15,583	8,426
Everest Ultimate MEM Latency (ns)	59	65	66	39	37	39	67
WinRAR 3.80 (min:sec) *	882	1091	888	652	645	584	837
POV-Ray 3.7 (min:sec)	570	712	548	498	462	408	488
3DMark06 overall	12,018	11,639	12,583	12,407	12,559	12,859	12,906
3DMark06 CPU	4,116	3,532	4,276	4,620	5,035	5,638	4,717
3DMark Vantage	6,928	7,301	7,459	7,450	7,453	7,516	7,588
3DMark Vantage CPU	20,207	26,709	30,615	34,909	35,548	39,725	32,446
3DMark Vantage GPU	5,524	5,877	6,034	5,902	5,868	5,917	6,044
Quake 4 (FPS)	190	152	180	145	156	228	207
Valve Particle Test (FPS)	85	69	100	131	143	161	111
Crysis 1.2 10x7 very low CPU1(FPS)	140	112	153	151	155	164	153
Crysis 1.2 10x7 very low CPU2 (FPS)	85	70	95	113	115	106	113
World In Conflict 1.09 10x7 Very Low Graphics	170	136	188	223	232	250	220

Bold denotes winner.

AMD, Intel CPUs Compared
Model	AMD Phenom II X4 940	AMD Phenom II X4 920	AMD Phenom X4 9950 Black Edition	AMD Phenom X3 8750	Intel Core 2 Quad Q9550	Intel Core 2 QX9770 Extreme Edition	Intel Core i7-965 Extreme Edition	Intel Core i7-920
Clock Speed	3GHz	2.8GHz	2.6GHz	2.4GHz	2.83GHz	3.2GHz	3.2GHz	2.66GHz
L1 Cache (total)	512KB	512KB	512KB	284KB	256KB	256KB	256KB	256KB
L2 Cache (total)	2MB	2MB	2MB	1.5MB	12MB	12MB	1MB	1MB
L3 Cache (total)	6MB	6MB	2MB	2MB	N/A	N/A	8MB	8MB
Front-Side Bus / Interconnect Speed	3.6GHz	3.6GHz	3.6GHz	3.6GHz	1,333MHz	1,600MHz	6.4GT	4.8GT
Execution Cores	4	4	4	3	4	4	8*1	8*1
Process Technology	45nm	45nm	65nm	65nm	45nm	45nm	45nm	45nm
Transistors	758	758	450	450	820	820	731	731
Die Size	258	258	285	285	214	214	263	263
Wholesale Price	$275	$235	$174	$124	$316	$1,399	$999	$284
Interface	AM2+	AM2+	AM2+/Am2	AM2+/AM2	LGA775	LGA775	LGA1366	LGA1366
TDP*2	125	125	140	140	95	136	130	130
Memory Support	Dual- Channel DDR2/ DDR3*3	Dual- Channel DDR2/ DDR3*3	Dual- Channel DDR2	Dual- Channel DDR2	Dual- Channel DDR2/ DDR3*4	Dual- Channel DDR2/ DDR3*4	Tri- Channel DDR3	Tri- Channel DDR3

*1 CPU features Hyper-Threading virtual cores.
*2 AMD and Intel TDP ratings do not directly correspond
*3 DDR3 supported in core but not in any shipping motherboards presently
*4 Dependent on chipset

How the Mobos and CPUs Match up

Anyone who has tried to use a Nikon teleconverter and lens from 30 years ago with a brand-new digital SLR will tell you that compatibility doesn’t always mean easy to understand. That is, trying to figure out which 30-year-old lens works with which camera is enough to make you want to buy a modern lens.

The same can be said of AMD’s AM2, AM2+, and upcoming AM3 sockets. Physically, the various CPUs will fit in the sockets but electrically they won’t all work. For example, the 2.8GHz Phenom II X4 920 has a DDR3 and DDR2 controller in it. However, plug the CPU into an AM3 board in a few months and it won’t work. For that, you’ll need a new AM3 CPU. That new AM3 CPU, by the way, will actually work in most DDR2-based AM2+.

Thoroughly confused yet? We are. As grateful as we are that AMD isn’t forcing its customers to buy new boards, the AM2, AM2+, and AM3 thing has thrown for a loop. Here’s how it breaks down:

AMD, Intel CPUs Compared
	AM2 MOBO	AM2+ MOBO	AM3 MOBO
AM2 CPU	Yes	Yes	No
AM2+ CPU	Maybe	Yes	No
AM3 CPU	Maybe	Yes	Yes

Intel Hopes You’re Ready for Eight Cores

AMD may be gaining ground, but that’s not holding back Intel, which will waste no time this year with a new octo-core CPU, a true budget series CPU, and major changes to the world of chipsets and graphics.

Multithread enthusiasts should be pleased with Intel’s plans for Nehalem. Although nothing is set in stone, the company is reportedly bringing out a desktop CPU with eight cores. With Hyper-Threading, 16 cores would be available to applications. This thread-monster would be confined to the LGA1366 platform though and aimed at the highest end of enthusiasts.

Later this year, Intel will also introduce its budget LGA1156 processors. Previously known as LGA1160, the chip will feature four fewer pins and one big addition. While in today’s computers the PCI Express lanes are controlled by the chipset, Intel’s budget quad-core Lynnfield and dual-core Havendale will have PCI Express built directly into the CPU package. A simple Direct Media Interface link will plumb out of the CPU to connect to the lowly SATA, audio, and other low-bandwidth I/O. Both Lynnfield and Havendale will feature dual-channel DDR3 controllers in the die, but Havendale may have the most impact. Havendale should be Intel’s first attempt at integrating graphics within the CPU package on a shipping processor. With a GPU talking to the CPU at QPI speeds and with direct access to an on-die memory controller, the impact of Havendale may have far more repercussions on the PC world than an eight-core Nehalem. We’ll be happy if it just means that Intel graphics won’t suck.

How to Tell the Difference Between the Top Procs

Intel’s quad-core Penryn Core 2 Quad is the only chip here that doesn’t use a monolithic design. Instead, these are two separate dual-core CPUs connected via the front-side bus. Note the massive chunks of L2 cache at the bottom of the chip. This L2 cache has helped lessen the advantage that AMD’s previous CPUs had in memory performance.

Intel’s Core i7 is actually lower in total transistor count than the Core 2 Quad and Phenom II but is second only to the honking-big Phenom in die size. Intel’s first monolithic quad-core has been criticized for having meager L2 cache, but it hasn’t hurt the CPU from being the fastest gun in the west.

The original Phenom’s 65nm-process made it a huge chip with just 450 million transistors occupying a full 285mm2 of die space. AMD now admits that it was a mistake to push for a monolithic quad-core design using a 65nm process as this chip ran hot and had terrible yields.

Since it’s mostly a die shrink, Phenom II is actually very similar to the Phenom die. Overlay the Phenom’s die shot with the Phenom, and you’ll see a lot of familiar structures between the two with the major difference being the size of Phenom II and the additional fields of L3 cache near the bottom of the die.