Archive for April, 2013

PostHeaderIcon AMD Radeon HD 7990 First Look

Radeon HD 7990

Nvidia’s GTX 690 finally has some serious competition

Today the embargo is lifting on the AMD Radeon HD 7990 that was teased back at GDC, so here’s the TLDR version; yes it’s just as fast and a tiny bit quieter than the Nvidia GTX 690, and it includes a mega bad ass eight-AAA-game bundle and costs the same price as its nemesis, making it quite a tempting package for those with the budget for it. Whether or not that will be enough to convince anyone to actually buy it remains to be seen of course, but at least AMD can no longer be knocked for conceding the $1,000 GPU market to Nvidia. It also signifies somewhat of a resurgence for AMD, who first came off the bench late last year and early this year with its totally righteous Never Settle game bundles, then attacked the midrange recently with the surprisingly powerful and quiet Radeon HD 7790 card, and is now going for the jugular with the dual-slot and triple-fan HD 7990. Whether AMD wins or loses that battle is slightly less important than the overall significance of this introduction, as in our minds its designed to not only beat Nvidia’s offering, but also to send a very clear signal to hardcore PC enthusiasts everywhere — AMD is still in the game, and doesn’t intend to give an inch of ground to Nvidia any time soon.

Radeon HD 7970

Spec Speak

Let’s start with the card’s basic specs: It features dual HD 7970 Tahiti GPUs clocked at 1,000MHz, which is a bit higher than we expected them to be, and higher than the clocks on a stock HD 7970 (925MHz) and the GTX 690 (915Mhz). Each GPU sports 3GB of GDDR5 RAM and has 2,048 Stream Processors, and the two connect to each other and the motherboard via a PCIe Gen 3 PLX bridge chip. The card takes up two slots, is 12 inches long, has a triple-fan cooler with copper heatpipes, and sports a semi-high TDP of 375w. For comparison, the GTX 690’s TDP is 300w. Power is provided by dual eight-pin PCIe connectors, which is the same requirement of the GTX 690, and one less than what was required on the dual-Tahiti PowerColor Devil 13 board. The board supports five displays with its four Mini DisplayPort connectors and single Dual-Link DVI port. The card will cost $999 and includes an eight-game bundle featuring: Bioshock Infinite, Crysis 3, Far Cry 3, Far Cry 3 Blood Dragon, Tomb Raider, Hitman Absolution, Sleeping Dogs, and Deux Ex: Human Revolution. We thought Battlefield 4 would be included as well since the game was demo’d at GDC on the HD 7990, but that deal appears to have fallen through. The card should be available two weeks from the day this is published from all the standard add-in board partners such as Sapphire, HIS, Gigabyte, MSI, etc. 


When it comes to performance, the card performs just as well as you would imagine, if you were imagining that AMD would only bring this to market when it was sure it could beat Nvidia’s card. To do otherwise would be a fool’s errand, so naturally the HD 7990 is faster than the GTX 690 in most tests that we ran, but not in all of them (see benchmark chart below). Overall performance with both the GTX 690 is very close though, enough to effectively call it a draw. It’s not as fast as the overclocked and water-cooled Asus Ares II, however, but nobody can buy those cards so the point is somewhat moot. When compared to a CrossFireX setup, the results are in favor of the HD 7990, but the CrossFireX results were achieved last month with an older driver, so it’s quite possible that two cards would be a tiny bit faster than the HD 7990 if using updated drivers. Even more surprising is the card’s lack of noise and heat, as the other dual Tahiti boards we’ve seen so far have either been water-cooled (Ares II) or so loud and hot as to require earplugs and heat shields (Devil 13), but not the HD 7990. It got up to about 85C in gaming and was a tiny bit audible, but not “loud” at all, which is an amazing development for AMD. We’d say all in all it’s probably just a bit quieter than the GTX 690, and gets about as hot. 


Radeon HD 7990 benchmarks

Best scores are bolded. Our test bed is a 3.33GHz Core i7 3960X Extreme Edition in an Asus P9X79 motherboard with 16GB of DDR3/1600 and a Thermaltake ToughPower 1,050W PSU. The OS is 64-bit Windows Ultimate. All games are run at 2560×1600 with 4X AA except for the 3DMark tests.

Frame Latency and micro-stutter

There’s been a lot of discussion lately about ditching frames per second as a standard for how “smooth” a game runs on given hardware, and instead examining frame latency or frame times instead. The reason why is that sometimes frames per second can be misleading, as we all equate anything over 30fps to be “smooth” when in fact a game can be run at even 50fps and still be somewhat choppy if the frames aren’t being delivered on a consistent schedule. For example, if the GPU alternatively sends frames to the system at 20ms and 100ms or higher throughout a test, it’s possible to still have an average framerate above 30fps, but to have it feel like you’re dropping frames. This is not usually something that you notice when running a single GPU, but it’s more prominent in multi-card setups since both cards are sending individual frames to the system, hopefully in concert with one another. This has been a bee in AMD’s bonnet lately, and when we sat down for the briefing on this card in particular AMD mentioned it, and acknowledged it. They called it “micro stutter,” which is fitting, as it’s an almost imperceptible stutter/lag that you can feel but barely see when playing certain games on specific configurations of GPUs and CPUs. 

As far as the HD 7990 goes, here is the situation. Yes this card and the latest drivers we used (13.5 beta 2) suffered from micro stutter, but it was imperceptible in most games with one major exception — Far Cry 3. We’re not sure what it is about Far Cry 3, but it runs like crap on this setup at 2560×1600 with 4xAA. We mean it feels like you are running through molasses the whole time, and when examining the frame time output from FRAPS it is easy to see why, as the latency between frames fluctutates wildly. When examining the frame time charts from other games the delivery times look pretty consistent, with only Crysis 3 showing some anomalies. AMD has told us it is working on a fix for this situation, but the remedy will most likely be a software implementation rather than a hardware fix. We have yet to receive and test a final software fix for this issue, and will sure to update you when we have tested it. 

Final Thoughts

On the one hand, we’re pretty pumped to see the HD 7990 finally arrive, as it’s more than a year late to market, and its absence made a lot of people question AMD’s commitment to the uber high-end PC gaming market. Since AMD has sort of abandoned the high-end CPU market, having them leave the same GPU market would mean nothing but bad things for a lot of consumers, so we’re stoked to see them join the fight. The card itself presents a very strong option for buyers looking to drop a large bill on a GPU given the fact that it includes over $500 of Kick Ass games. If the bundle wasn’t included, we’d say you could flip a coin over whether or not to get this card or the GTX 690, but the gaming bundle clearly gives AMD a significant advantage against its competitors.

Now that the HD 7990 has launched, we have to wonder whether or not Nvidia will counter AMD’s launch with the rumored GTX Titan Ultra, but it sure seems likely the way the two have been going at it lately. This effectively means the GPU cold war that has existed throughout 2012 is now largely over, with the two companies fired up and ready to wage war this summer. All we can say to that is “bring it on!” 

Similar Posts:

Article Source: The Only Yard For The Internet Junkie
If you like all this stuff here then you can buy me a pack of cigarettes.

PostHeaderIcon Measuring and Increasing the ROI of Your Content Resources

Posted by Mike Pantoliano

Let me cut right to the chase. Do you want to know the value of your content marketing efforts? Want this report?

the resulting assisted conversion report

Read on and I'll tell ya!

Calculating the real ROI

With so much emphasis often put on the traffic generation potential of a good content marketing strategy, I want to focus this post on measuring and increasing the return on the (sometimes sneakily large) investment. Some common goals you'll hear surrounding a content marketing strategy include generating traffic for generic terms, increasing social shares, and developing the brand's authority (measured by increases in branded traffic, or some other indicator). In the right circumstances, all of these are nice metrics for the relevant stakeholders in the organization, but they're all just proxies for measuring the growth of a business. They're measurements of the means, not the end.

The impetus for a lot of what I'll be talking about in this post comes from Josh Braaten's post on the Google Analytics Blog a few months ago titled "How to Prove the Value of Content Marketing with Multi-Channel Funnels". Josh talks practically about how to measure the business impact of traffic that first experiences your site via a page that isn't directly selling a product or service to a consumer. Think: the "How to get into fly-fishing" article written by the outdoors retailer that sells fly-fishing poles, or even the "How to measure the effectiveness of content marketing" article written by the guy working for a company that's doing a two day kick-ass web marketing conference in Boston on May 20th & 21st :). Indeed, these content pages aren't selling a product or service, but they are selling the brand, the "purchase" made by the consumer is everlasting trust; and it has a really low conversion rate.

The necessary analysis for this gets difficult because it is so rare for a user to make the jump from discovery/informational stage to transactional stage in one sitting. Hence the need for multi-channel analysis: we need to take a conversion, look back at all of the interactions that have taken place leading up to that conversion, and assign some amount of credit to those channels that often show up toward the beginning of the conversion path. Social networks and the content that usually ranks for generic keywords are most often found in these early interactions. They are inherently 'openers' or 'exposers'.

So, now that we've covered the theory, let's look at measuring that ROI.

Expanding upon Josh Braaten's multi-content funnels

Everyone interested in what I've covered above should absolutely read Josh's post. In it, Josh walks you through how to create a report within Google Analytics' Multi-Channel Funnels that classifies users by the page type for which they first interacted (based upon landing page).

creating a channel grouping in GA's MCFs

Custom channel creation is a lot like creating an advanced segment in GA

A long conversion path in MCF

The top conversions path report – seen here displaying a pretty convoluted conversion path for one particular conversion.

I'm going to offer a slightly different direction, but they both accomplish the goal of getting value out of our visit data. Instead of comparing content sections against each other, let's instead compare it against our other channels like direct, referral, organic, and paid.

Let's do a step-by-step walkthrough

Head on down to the multi-channel funnels reports.

location for multi-channel funnels in google analytics

Make a copy of the basic channel grouping template.

make a copy of the basic channel grouping

Include traffic based on landing page URL. Hopefully you've got your resource center, blog, or content home on a neatly identifiable path in the URL. If you don't, you may have to go the route of declaring page-level custom variables.

create your channel grouping

Drag it to the top. The order at which you put these channels is important because GA will go down the line until a match is found, then stop. If we leave our Resource Center channel at the bottom, the channels above will take a ton of visitors first because our rules aren't mutually exclusive.

channel ordering is important

Though not completely related to this topic, I'd also suggest separating your organic channel into branded, unbranded, and (not provided).

break out your branded, not provided, and unbranded organic search

Because of that importance of ordering, if you put (not provided) first and branded second, the final organic group will necessarily consist of unbranded traffic.

use regex to create your branded channel segment

You can create this segment with a neatly crafted regex of your brand name and other branded terms.

Finally, let GA calculate things out, and voila!

the resulting assisted conversion report

What can we learn from the above?

Well, it should be pretty clear that under the traditional model of last click analysis, our resource center is under-valued. This much is obvious by the disparity in last click conversions and conversion value compared with assisted conversions and conversion value. Not only that, but the "Assisted/Last Click or Direct Conversions" ratio (6.62 in the screenshot) tells us that this content is acting in an assist role more than any other channel we have (the higher the number, the more likely it's an 'opener', not a 'closer' – those trend toward zero).

When we look at assisted conversion numbers, we CANNOT say that our resource center content is now directly responsible for $26k in revenue; that would not be quite fair using this model. But our content did have its hand in a lot more conversions than we may have originally assumed.

Now, as for this channel's relative contribution to the bottom line compared with other channels, well, yes, it's still a lot smaller. But consider that this particular website's resource center is actually quite small, especially compared with the size of the rest of the site. Knowing how many pages are in a resource center makes it pretty easy to apply simple math to determine what each new page is roughly worth. Or you could choose to do deeper analysis into specific pages or sections within. Again, I point to Josh Braaten's post for more on that.

But at the end of the day if you know that each new page added to the resource center has an assist in $X worth of conversions per year, justifying expansion becomes a lot easier.

A bonus tip for content marketers

So that was measuring the ROI of a content marketing strategy. But I've actually got a tip for increasing ROI that I'd like to share.

Our content strategies are targeted at the generic keywords that more often than not are queries that align with the user's information-seeking intent. If we had our way, the path would go like this:

Kitten mittens purchasing decision

A user searches "my cat's too noisy" and lands on your site's blog post "10 ways to deal with a noisy cat."

The user reads and is very happy with your content. In that content, you suggest "kitten mittens," a product that you sell.

The seed is planted in the user's mind, and upon deciding that they're ready to buy, the user either searches for your brand name, that post again, or the "kitten mittens" product, all of which lead back to your site.

always sunny's kitten mittons

Nightmare scenario time: what if they searched for "kitten mittens" and you don't rank for that term? Well, your content has done all the hard work, but your high-ranking competitor swoops in and gets the purchase. This must be corrected. But how?


It doesn't matter what remarketing tool you use (this would be super easy with GA's remarketing tool – here I wrote a post on it!), put the user above in a "noisy cat owner" list, and target them with "kitten mitten" ads around the web.

creating the kitten mittes segment

Thanks for reading, I hoped you learned something!

Let me know what you think in the comments or on Twitter, @MikeCP. Don't forget that Distilled is running our search marketing conference, SearchLove, in Boston on May 20th and 21st!

Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don’t have time to hunt down but want to read!

Similar Posts:

Article Source: The Only Yard For The Internet Junkie
If you like all this stuff here then you can buy me a pack of cigarettes.

PostHeaderIcon Mobile Malware Grew 163 Percent in 2012, Almost All of It Aimed at Android

Android ZombieSecurity reports suggests mobile malware writers are almost exclusively focusing on Android.

To the victor belong the spoils, along with everything else that comes with being the most popular kid on the block. In the mobile world, Android is clearly winning in terms of market share, and while that translates into a bigger chunk of the pie, it also means there’s a big brightly lit target painted on Android’s back for malware writers to take aim at. Whether or not mobile malware is truly a problem to begin with, however, is debatable.

On the surface, a new report (PDF) by NQ Mobile paints some scary numbers. For example, NQ Mobile claims it saw a 163 percent increase in mobile malware in 2012 versus 2011, and that 94.8 percent of all malware discovered was designed to attack Android (Symbian, once the top dog in mobile market share, only found itself on the receiving end of 4 percent of mobile malware).

In terms of growth, bits of malicious code designed for mobile devices is clearly on the rise, but it’s not yet an epidemic. Though mobile malware grew 163 percent, that works out to 65,227 pieces of malware. Not only that, but avoiding mobile malware is pretty easy. According to NQ Mobile, the three primary methods for delivering dirty code to devices in 2012 included app repackaging, malicious URLs, and Smishing (phishing via SMS text messaging). In other words, you can remain relatively safe by not clicking links willy-nilly and only downloading apps from trusted sources.

As easy as that sounds, more than 32.8 million Android devices were infected in 2012, up almost three-fold versus 2011, NQ Mobile says.

Image Credit: Flickr (greyweed)

Follow Paul on Google+, Twitter, and Facebook


Similar Posts:

Article Source: The Only Yard For The Internet Junkie
If you like all this stuff here then you can buy me a pack of cigarettes.

Free premium templates and themes
Add to Technorati Favorites
Free PageRank Display
Our Partners


Related Links
Resources Link Directory Professional Web Design Template