Once you’ve learned everything you need to know about ultraportable laptops, check out our list of the best ultraportable laptops.
The average Ultrabook is an amazing piece of hardware. It combines the portability of a tablet with the power of your typical notebook. Just ask Intel.
Of course, Intel isn’t an unbiased source. To cut through the hype about Ultrabooks and their performance we’ve decided to use the benchmark data we gather from our laptop reviews to perform a by-the-numbers comparison of Ultrabooks with your average laptop.
We’ve taken a look at the hardware before, but from a broad perspective. Since then we’ve reviewed several production Ultrabooks equipped with both Core i5 and Core i7 low-voltage processors. Now that we have these results we can compare them to normal Intel Core i5 and i7 processors to gauge the real performance difference.
One of the benchmarks we use to gauge processor performance is SiSoft Sandra’s Processor Arithmetic benchmark. It is optimized for multiple threads, which is good, because all of Intel’s mobile processors support Hyper-Threading.
So, how does the low-voltage processor found in an Ultrabook compare to a standard Intel part? Let’s have at the graph below.
These are the real numbers pulled straight from the benchmark. Normally we only report the Combined (GOPS) figure for simplicity’s sake, but here we’ve presented the whole enchilada. Intel’s low-voltage processors are designated by their odd model numbers, while standard processors have even model numbers.
As you can see, the fastest low-voltage model, the Intel Core i7-2677M, is slightly slower than the slowest standard model, the Intel Core i5-2410M. But while our benchmark picks up on this difference, it’s not perceptible in real-world use.
However, the Core i7-2677M is a rare processor that’s only available in some extremely expensive Ultrabook configurations. A more realistic battle is the Core i5-2467M against the Core i5-2430M.
Here we see that the performance difference is much more noticeable. The Core i5-2430M offers a combined score that’s about 30 percent higher than the low-voltage model. This performance gap will be noticeable in any demanding application (such as Photoshop or Windows Live Movie Maker).
All processors built by Intel now have the integrated graphics processor (IGP) on the same die. This means that every single Intel processor currently produced automatically comes with an IGP. They’re inseparable.
Intel uses the name “Intel HD 3000” to refer to the integrated graphics processor in both its low-voltage and standard processors, but the base clock speed of the IGP in a low-voltage processor is almost half that of the base clock speed in a standard processor (350MHz vs. 650MHz).
The IGP can overclock itself automatically to make up for that difference, but is this truly effective? Let’s find out by having a look at performance in 3DMark 06.
Here we see results that are more varied than those in the processor performance graph. As it turns out, laptop design seems to have some impact on Intel’s HD 3000 IGP. We’ve found that laptops with the same processor almost always offer identical performance, but the IGP is not as predictable.
This makes the results a bit difficult to parse, so let’s use averages. The four Ultrabooks we’ve tested offer an average 3DMark 06 score of 3,703, while the four standard laptops shown above reached an average score of 4,027.
Though this is another win for standard laptops, the overall difference is only 10 percent, which isn’t much to brag about. We also see that some Ultrabooks are quicker than some standard laptops. The Toshiba Z835 and Asus UX31 offer better scores than the Lenovo ThinkPad T420s and HP EliteBook 2560p.
We’ve now demonstrated that Ultrabooks are indeed slower than standard laptops. However, slower performance may not be an issue – and in fact may even be desirable – if it results in better battery life. Manufacturers have been claiming some serious battery life figures for Ultrabooks, with some (like the HP Folio 13) shooting for eight or nine hours. Are these results achievable, or are they exaggerations?
Let’s have a look. Ultrabooks are in the top group, while ultraportables with standard processors are in the bottom group.
Results here are a bit mixed. The blue bar represents the “Standard Test” which is an all-out attack on the battery. This provides an idea of what endurance might be if you tried to run demanding applications, like games, without being plugged in to a wall socket. It’s clear that the Ultrabooks have an advantage there.
When we look at the “Reader’s Test,” however, the standard laptops take the day. The Reader’s Test is a much less demanding workload that replicates endurance in near-idle situations (such as light Web browsing or document editing). The normal Core i5 processors never have to ramp up performance, giving room for the batteries to stretch their legs.
Your exact battery life will always depend on your workload, but our benchmarking and our experience indicates that Ultrabooks have no advantage here. This may seem strange, but the results make sense. Slim-and-sexy Ultrabooks have less space to devote to the battery. This negates the power savings provided by the use of low-voltage processors.
It’s clear that you should not buy an Ultrabook solely for its endurance. They have no advantage over standard ultraportables. In fact, they are at a disadvantage – while battery life is about the same overall, their batteries are not meant to be user serviceable. That means you will have difficulty replacing the battery yourself. It also means that extended life batteries are rarely available.
Beyond the numbers
By the numbers, we see that standard laptops do offer better performance than Ultrabooks. In a worst-case scenario (a processor intensive application that can use all available threads) the low-voltage processor is significantly behind the standard part.
Most of us don’t constantly use demanding applications, however. The performance numbers above must be tempered by the fact that your use will impact perceived performance greatly. No modern Intel processor is daunted by Microsoft Office, Google Chrome or Windows Media Player.
On the other hand, the advantage held by a standard processor can provide an extremely noticeable improvement in the correct software. For example, let’s say that you want to save a movie to 1080p in Windows Live Movie Maker, a free program that does a good job of using multiple threads. In this situation the standard processor could save you five or six minutes if you’re working with a long recording.
The impact of the solid-state hard drives should not be forgotten. An SSD-equipped Ultrabook will boot in under twenty seconds, while laptops equipped with a mechanical hard drive usually take over forty.
Gaming performance will generally be better on a standard processor, but the difference will not be noticeable unless a game relies heavily on the processor rather than the IGP. This is true for some strategy and MMO titles, but not for most other games. All laptops lacking a discrete graphics processor from AMD or Nvidia will have difficulty with demanding new 3D games, such as Battlefield 3.
While Intel’s bluster about Ultrabook performance perhaps represents an enhanced version of reality, the general idea – that consumers will be as happy with an Ultrabook as they are with a standard laptop – is correct.
Only demanding applications reveal the advantage of a standard laptop. As a consumer, you have to ask yourself – “How do I use my laptop?” If you surf the web, watch streaming video and use Microsoft Office, an Ultrabook can serve you well. On the other hand, consumers who produce YouTube videos, edit images and develop websites will enjoy the extra power provided by a larger laptop.