Anand Shimpi is widely recognized as one of the most authoritative voices covering the PC industry today. Having started the eponymous AnandTech.com as a teenager, he has now been informing the computer-buying world about CPUs, GPUs, SoCs, SSDs, and other acronym-friendly technology for over 14 years. Today, he was kind enough to chat with us for a few minutes — to discuss trends, technology, and trends within technology.
You started your website way back in the 90s. What has been the most significant change you've seen on the web in that time?
When I started AnandTech, the web was useful but still underutilized by the vast majority of the world. If there's one significant change I could point to, it would have to be just how ubiquitous the internet has become in the past 14 years. Telling someone to search the web for an answer to their question in 1997 was a bit odd, whereas today it's common vernacular.
What about the rate of change, do you think it's accelerating or slowing down?
That's the beauty of exponential growth, an ever increasing rate of change right?
In hardware I believe the rate of change has accelerated a bit compared to the early days, but that's simply because a lot of the companies who ultimately drive change are now more efficient at their own product execution. E.g. in the late 90s we might see a new microprocessor architecture every five years, now we get major design changes every two. We're still ultimately gated (on the hardware side) by Moore's Law.
The smartphone/tablet industry appears to be under the influence of a limitless rate of change, but that's due to the fact that the technology there is still operating in a hyper-Moore's-Law phase. Die size and process technology haven't really limited progress in ultra mobile devices yet. Now that there's significant demand for more performance in these devices, the roadmap accelerates. You'll see things calm down there, but we still have another couple of years of this pace left, I believe.
Software is the wild card, of course. Improvements in software efficiency can do wonders for progress in this industry. The advent of ultra mobile OSes and web browser releases in the past few years are both examples of this.
What do you use Twitter for?
I use Twitter mostly to interact with the readers, typically in the form of answering questions. While I do respond to questions via email, it's far easier (and less rude) to send out short responses via Twitter. In email there's a lot more structure that I feel compelled to include, not to mention that the answer lengths can be considerably longer. I have noticed that most of the questions I get via Twitter are those that can be answered in a short sentence (e.g. What SSD should I buy for my Mac?).
Are you ever tempted to just introduce yourself as Anand and expect people to know who you are?
I always introduce myself as Anand, it's my name!
Now if you mean at industry events or meetings, I still introduce myself as just Anand. I didn't come from a journalism background, when I first started attending industry events I noticed everyone always introduced themselves as Name from Publication. It always rubbed me the wrong way. If I'm asking a question, and my question is appropriately phrased, the publication I write for is irrelevant information.
The obvious counterargument is knowing the publication one writes for helps put their question into context, and as a side benefit, tells the audience whether or not this person should be respected (e.g. I'm Vlad from The Verge vs. I'm Vlad from Hookers Weekly). Again I'd argue that context can easily be delivered in the question itself, and respect is more significantly gained or lost with actions — not with the historical value of a brand.
What current tech development has you most excited?
Not much has changed for me here in the past decade: I'm still most excited about the progress in the silicon space. It's what makes the world go 'round.
Huge growth in the PC space was encouraged by the marriage of high speed internet and a lot of compute. In smartphones we're on the verge (see what I did there?) of a similar era with widespread LTE and more modern looking microprocessors (think Cortex A15 or later). We're not too far away from being able to deliver an Xbox 360 level of gaming experience in a smartphone, including things like network latency. And yes, I do believe the carrier problem goes away eventually as well. There was a time when I had to pay for internet usage by the minute on my desktop, mobile data caps are a solvable problem.
Which trend makes you gnash your teeth in frustration?
Something I call the cable-TV-ification of the internet. For the past several years it seems as if there has been a trend away from ultimate understanding in content online and towards the tenets of modern mainstream media (sensationalism and the general silliness you see on US cable TV news). The transition isn't anywhere near complete, but I feel like that's the direction things are headed. We have to learn from the mistakes of our predecessors, not repeat them with sweeter technology.
Which mobile OS do you use today? What do you expect you'll be using 18 months from now?
OS X on my notebook, iOS on my phone and Android on my tablet (technically also webOS for my tablet but the TouchPad is basically my kitchen computer).
I'm not much of a fortune teller, so predicting the success of any one mobile OS 18 months from now is a tall order for me. I can share some data and insight though.
Nearly every manufacturer I talk to is more interested in Windows 8 for tablets than Android. Either through incredible insight or great convincing by Microsoft, many of the hardware vendors are expecting big things from Windows 8 tablets. Whether or not this pans out and the impact it will have on Android is unclear at this point. Apple obviously has a huge following and I don't expect that to significantly change in the short term.
Big markets tend to like consolidating and I don't expect the mobile OS space will be any different. We only have to look at how things played out in the PC space in the mid-90s for a hint of what's to come. There was a time when people viewed OS/2 as a viable competitor in the PC OS space. I don't have a good feel for who is going to ultimately emerge victorious though. Apple alone is an example of how silly you can look if you try to predict the future (e.g. Mac OS in the late 90s vs now, iOS).
Is the PC, as embodied by a tower chassis hidden away under a desk, dead?
I don't believe so. The PC in a tower is still the most affordable way to get a ton of compute. With Sandy Bridge, I finally could get enough performance in a 15-inch notebook to ditch my Mac Pro. Lately I've been testing Intel's Sandy Bridge E and man does it tempt me back to a desktop as my primary workhorse. If I didn't travel so much for work I'd definitely use a desktop as my primary machine (note that I still have multiple desktops that I use, but most of my work is done on a notebook thanks to its portability).
We may see the desktop turn into something a bit more modular (notebook + external attached graphics and storage), but we're still a bit away from making that happen. I'd also add that the technology world is very cyclical. When I was a kid, folks were talking about the mainframe + dumb terminal model. A couple decades later, we pretty much have a version of that with the cloud + smartphone model.
Moore's Law has given us a new category of "fast enough" computing devices (see first page of my original article on the Atom back in 2008). However to completely abandon the PC and embrace the smartphone/tablet does mean you are giving up around 5 years of performance improvements to do so. For many users that's fine, but not for all. This industry is built around progress, forcing everyone to take a step back in the compute department doesn't make sense.
Within the next 5 years, many very smart people believe that mainstream computing will move to wirelessly dockable smartphones. I believe this is a valid vision. It's not for everyone, but it can surely happen.
What do you think of Google's Chromebook?
It's surprisingly usable. I think it's definitely sufficient for a lot of folks, the question really boils down to price and competitive alternatives. I don't know that Google has delivered a substantial enough experience to take off in the sense of the iPad, but it's a decent platform.
Star Trek or Star Wars?
One thing about me: I don't have favorites. No favorite color, no favorite food, no favorite book, no favorite movie — just a lot of things that I like. In high school I probably would've said Star Wars, but these days I feel a bit closer to Star Trek (likely due to Lucas' handiwork in the prequels). But, like I said, no favorites — I'll take both.
Do you think extraterrestrial life will ever be intelligent enough to understand Carl Sagan's theories about it?
Aren't you presuming that it isn't already? The sheer scale of the universe is a difficult concept to realize by itself. Attempting to quantify the intelligence of extraterrestrial life somewhere in that vastness is no easier. If you want my completely uninformed opinion, I'd say yes, but I lack sufficient evidence (other than a couple of statistics classes) to prove it.
What was the last book you read?
The fifth Harry Potter book. It turns out I'd never read any of the Harry Potters or seen any of the movies. I realized this is something I should probably change, so I started doing weekly viewings of the Harry Potter movies in my home theater with some of my friends. After making it through most of them and getting the feeling that I was seeing a relatively incomplete story I decided to give the books a try. I like them better than the movies, particularly later on in the series.
I regret that I don't get a chance to read a lot anymore outside of the really technical stuff. It's nice to read a good story as a change of pace. Reading about state machines and voltages all day makes Anand a dull boy.
What's missing in technology that nobody seems to be working on?
Your broad question deserves a similarly broad answer.
Home automation, automobile electronics, home audio and video (receivers, TVs, and to a lesser extent, projectors) and the medical field are all in dire need of help from some truly passionate individuals from our industry. There's an incredible amount of business opportunity in all four of these areas, and perhaps the greatest upside in the latter. We just need folks to solve the unique problems each poses.