Gateway AMD and Intel Laptops, a Platform Analysis
by Jarred Walton on August 12, 2009 2:00 AM EST- Posted in
- Laptops
The Final Word
Hopefully we've now answered and explained any lingering questions about how the AMD and Intel platforms compare when it comes to laptops. Intel has much better battery life, but that's only one aspect of the overall equation and there are definitely areas where AMD has the advantage over Intel. Intel also wins in application performance, with noticeably snappier system responsiveness (i.e. loading applications, installing programs, booting Windows, hibernating/resuming). When it comes to 3D graphics, however, the AMD solution is clearly superior to Intel's anemic IGP.
Looking at the big picture, either one of these laptops - or similar laptops from other vendors - would work very well for any student or home office user. With an entry price point of $500, the Gateway NV52 series provides an attractive package that can handle just about any task you might want to throw at it, up to and including entry-level gaming. The Intel system on the other hand is clearly faster outside of gaming tasks, but the $80 price difference represents a 16% price increase. If you're willing to sacrifice graphics performance, you do get more than a 16% increase in battery life and general application performance, so ultimately you need to decide whether you want to have better battery life or if you would prefer improved graphics.
The graphics situation is a bit muddy when we consider the full GPU market. ATI's HD 3200 may be over twice as fast on average compared to Intel's GMA 4500MHD, but that's a lot like beating a Kia Spectra with a Hyundai Accent on the racetrack. There are many faster graphics options if that's your primary concern, and truthfully you would probably be better off sticking to older games on laptops and saving your real gaming for desktops where a $500 PC doesn't fall completely flat the instant you boot up the latest 3D tour de force.
Considering the contestants, it was almost a foregone conclusion, but it's always good to have concrete numbers to back up our recommendations. As far as we're concerned, laptops - especially entry-level laptops - need to function as a mobile computer first and foremost. By that criterion, Intel has the clearly better mobile platform. Faster CPUs that draw less power and provide better battery life rate a lot higher in our book than barely adequate gaming performance. A 16% price increase for 25% more battery life and 25% faster general application performance is hard to beat. Using a higher capacity battery on an AMD platform could give you equivalent battery life, but then you're lugging around a heavier laptop and many high capacity batteries cost far more than $80. If you really want improved battery life, you'd be looking at an Intel platform with a high-capacity battery - or a MacBook.
Finally, we can't neglect the netbook platforms. The vast majority of these use Intel Atom processors, which are completely different from Intel's Core 2 chips. They use far less power, but if the AMD Athlon X2 QL-64 in the NV52 seems slow, it's only 130% faster than an Intel Atom N280… in a threaded workload where the Atom does reasonably well. In single-threaded tests and other benchmarks, the Athlon QL-64 can be 200% or even 300% faster than an Atom N280. On the other hand, with a similar size battery we've seen Atom N280 netbooks manage over twice the battery life of the NV58, so as usual it's a case of priorities. Those who want a "real" PC experience will likely appreciate notebooks like the NV52 and NV58 series a lot more than netbooks, while users that only want something small with reasonable performance and great battery life will gravitate towards netbooks.
In the end, it's all about choice. Battery life, graphics, CPU, size, features... we have plenty of options on where to spend our money and how much to spend. AMD may not be the better platform overall, but they do provide a viable alternative to Intel platforms and the lower cost is certainly an attractive aspect. Personally, I'd stick to gaming on my desktop and if I "need" to game on a laptop I'll stick to less taxing titles or buy a laptop with a discrete GPU, but however you slice it, having choice is a good thing.
Facebook
Twitter
RSS
67 Comments
View All Comments
samspqr - Wednesday, August 12, 2009 - link
(I mean the one on page 5)JarredWalton - Wednesday, August 12, 2009 - link
As mentioned in the text, the graphs on pages 5 though 8 use a different format. I chose the format because I feel it provides a quick way of looking at relative performance. The percent labels aren't particularly meaningful, since the bars are not in actual percentages (though the bar sizes are).Spacecomber - Wednesday, August 12, 2009 - link
Although I found the article interesting, comparing the advantages of intel based laptops to amd based laptops, I have to agree that the choice of how to represent the data in the charts doesn't really work for me. I understand what the notion was, representing the data as a kind of tug of war between the two platforms, but the utility used to create these charts didn't provide the means to make this clear. I'm a strong believer that charts should be intuitively obvious and that you should be able to interpret them without looking at anything other than what is contained in the chart.These charts have an additional problem when it comes to the representation of the gaming data (or any data where the same measurement unit is being used). You can no longer easily compare performance for the same platform across different games (or whatever the instances being measure are). For example, it is not easy to see in these graphs that some games don't have playable framerates, regardless of the platform.
Anyway, while I'm sympathetic with the effort to try something different, I didn't find that it helped me to comprehend the data any better, and it seemed to be more of a hinderance, instead.
JarredWalton - Wednesday, August 12, 2009 - link
The problem is two-fold: I had a bunch of data to represent, and I didn't want a ton of small charts. I debated just putting in a table, but that seemed a poor way of doing things - especially since I can show the number labels and you get the chart plus table effect. The issue is compounded by units that are not easy to chart with each other - i.e. PCMark05 has some scores in single digits, some in the tens, and some in the hundreds. Do I break PCMark05 scores into 16 different charts, or does the "100% Stacked Bar" chart option provide all the necessary detail?I'm inclined to go with the latter as a reasonable compromise between the various options, and let people spend a few more minutes looking at the data to draw conclusions. It's great when a chart can be "intuitively obvious", but there are times when complex data requires a more complex chart. The net result is that I had a choice between more aesthetically pleasing pages and a single summary chart, or I could make dozens of smaller charts and get a cluttered page.
I understand some people won't like the representation, but I've been playing with the data for the past hour and there's really no option that's as succinct. Some graphs (gaming) lend themselves to other options, but the PCMark05 chart and Application Performance charts have to be split up, put in a table, or something else.
FWIW, I'll post alternative charts in a bit for those that really hate the current charts. However, I'm still a firm believer in requiring people to READ the text. I get the impression that many of the comments have only looked at a small portion of the text and are making a lot of assumptions - assumption which are, not surprisingly, incorrect.
Wellsoul2 - Wednesday, August 12, 2009 - link
The Chart is still backwards compared to the first article onbattery life. Idle is 111 minutes and DVD 242 minutes.
It's reversed from last article.
JarredWalton - Wednesday, August 12, 2009 - link
Sorry... not sure how that happened, other than sorting charts in Excel. For some reason, the top spreadsheet cell showed up on the bottom of the chart, and so I resorted. Apparently I missed the numbers at some point. I'll correct the chart shortly....mczak - Wednesday, August 12, 2009 - link
Hmm, what about the raw numbers which should show minutes runtime? Idle has lowest runtime, DVD playback largest. That definitely looks wrong...Wellsoul2 - Wednesday, August 12, 2009 - link
Yes...looks like you reversed the numbers.whatthehey - Wednesday, August 12, 2009 - link
Reading comprehension is a great thing!"We are using a different format for the graphing than we used previously.... For this chart, equal performance is denoted by the 50% mark in the middle of the chart. A larger bar means better performance, and the numerical results (in minutes) are available for those that prefer raw numbers."
Ah, yes, the percentage marks are entirely misleading! Shame on you AnandTech for making people read! (FWIW, I had no issues reading and interpreting the graphs, but then I'm a computer programmer.)
Ben90 - Wednesday, August 12, 2009 - link
" quit looking ahead and pay attention to this paragraph! " You should have added that to the page before, my screen was already down in the chart when i read this :)