On November 3, 2014, I published our review of the Nexus 9. It wasn't especially pretty, if I'm honest. But as with all things Nexus, time and software updates (mostly software updates) can smooth out rough edges and straighten up quirks, so a revisit seemed necessary. Now, three months on, have things really changed with Google's flagship tablet? Or is it still the HTC-made misfit I wanted to love, but just couldn't?

The end of a review is nothing to spoil, so I'll just be out with it: the Nexus 9 feels like basically the same kind-of-OK-but-not-great tablet today as it did the morning it arrived on my doorstep. There have been improvements thanks to software updates, but for the most part, I find myself still largely disillusioned of the Nexus 9's initial proposition: that of a high-end Android tablet with few compromises, as envisioned by Google.

Hardware

DSC06790

There have been rumors of HTC revising the Nexus 9 to remedy some of its early hardware woes, but there's never been anything definitive to prove it (examples of variation in manufacturing quality is what I'd call them). Google is notoriously tight-lipped about this sort of thing, so I doubt we ever will get much in the way of confirmation unless we take a week-one review unit and one bought today and then tear them apart iFixIt style. I don't know about the rest of the team, but that's not something I'm going to be doing.

So, the question really is whether or not my Nexus 9 has gotten any physically worse in 90 days, give or take.

The crackling noise you hear when pressing around the center of the back of the tablet has worsened a bit - it is more audible - and I do feel like the Nexus 9 has developed a couple of new, unrelated noises, too. But it's not falling apart - the volume rocker and power button are essentially as abysmal as the day I received it, and the rear cover is still an absolute grease magnet. Otherwise, it feels basically the same, and has developed no noteworthy physical faults since November. That is to say, from a quality standpoint, the Nexus 9 still seems quite mediocre. Well, except in one regard: light bleed.

I have never seen light bleed so bad as this on a tablet costing anything remotely close to $400, and it just keeps getting worse. The top-right corner's bleed is now visible in pretty much any situation. It's kind of embarrassing. I really hope HTC and Google have addressed this.

I freely concede that it's possible some of these issues may be remedied on newer tablets. But I don't have one. Minor manufacturing changes can certainly be implemented to improve a product, but that's not without a cost, and the retail lead-in time for such changes can easily reach into the months. I think it's more likely there is simply variability in the quality of units out there - I just happened to have gotten a crappy one, as did some other unfortunate folks.

The Nexus 9's front-facing speakers still don't win much love from me, as the fidelity and dynamic range they provide is categorically inferior to that of my iPad - I'm 99% certain these are just the One M8's speakers shoved into a tablet body, which would make sense, because HTC.

The display itself is, I think, pretty good. It's not amazing, but aside from the awful light bleed, I still like it. It gets reasonably bright, it has solid viewing angles, and the high resolution makes it look nice and crisp. Could it be a bit more colorful? Yes. I think it has a slightly washed-out quality, but I don't think Google dropped the ball or anything.

Oh, and it still gets weirdly hot even doing something simple like web browsing. Confidence-inspiring!

Battery life

If there's one area I've seen improvement on the Nexus 9 since receiving it 3 months ago, it's standby battery life. The Nexus 9 can reliably go seven days sitting untouched on a single charge, whereas before I was averaging more like three, maybe four. Power consumption under some usages also seems reduced - I noticed I could binge on Hearthstone without hitting the battery quite as hard, though the difference is still firmly in the "marginal" range.

Web browsing still seems to decimate it, though, and in my particular case I think about 4-5 hours of screen-on time (when used in a 2-day period) is what I'm getting these days. I might be eeking out, on average, about 20-30 minutes more than I did on some of the older builds, but it's in the same ballpark - we're talking gains of 10-15%, tops.

This still puts the Nexus 9 somewhere in the lower half of typical tablet battery life. It's not abysmal, especially with the improved standby life, but it's not exactly good, either.

Software

Whether the Nexus 9 is now faster than it was when I first reviewed it, I can't really say with measurable certainty. But it really doesn't seem like it. It still gets random hangs when multitasking and in certain applications, or appears to thermally throttle itself, and this just makes the whole experience of using it less than pleasant for me. In fact, it is decidedly unpleasant. If I were spending $400 (or $480, more likely) on a tablet, I'd expect considerably better performance in day-to-day activities like web browsing. In some situations, the Nexus 9 feels quite quick, but in too many others I find myself frustrated by its unpredictable slowdowns.

To be sure there wasn't some strange software issue, I flashed it to the most recent build and did a factory reset. It's still just as janky.

I think it's also important to explain what I mean by slowdowns, because people do have varying definitions here. Multitasking on the Nexus 9 is when you're most likely to encounter speed problems - noticeable delays switching tasks (one to three seconds, or more), UI hangs when going home from a memory-intensive application, random stutters in the launcher occasionally.

The Nexus 9 also consistently feels as though it needs to "warm up" when I pick it up after it's sat for more than an hour or so, as it seemingly enters a 'deep sleep' mode requiring five or ten seconds to get to full crank again. This might be an endearing trait in an old Italian car, but not a tablet.

Multitasking lag is by far the greatest impediment to using the Nexus 9 in anything resembling a productive fashion.

Chrome is still an utter dog in terms of performance - browsing on my iPad feels like a Crisco-lubed slip'n slide on a warm summer day; the Nexus 9 is akin to covering your body in velcro and running blindfolded through Bob's Carpet Emporium. While both would make for amusing Olympic events, the latter sounds substantially less pleasant and much more tedious.

It's subjective, but for me, web browsing is a huge part of what I do on a tablet. The Nexus 9 makes that experience kind of sort of not enjoyable. Forget it, let's not sugarcoat: it's just kind of shit. I don't even know how people put up with it.

But multitasking lag is by far the greatest impediment to using the Nexus 9 in anything resembling a productive fashion. App reloads are frustratingly common on the device to the point that it actually makes me wary at times to switch away from what I'm doing for a moment to check an IM, instead choosing to finish the task at hand and then open the incoming message. This is not an enjoyable experience, and it's one I never find myself having an issue with on my iPad - I should not actively have to be taking into account how my device's RAM will affect the UX.

I don't really care who is to blame, either, be it NVIDIA, Google, or Cool Pope. The end result is a crappy experience that makes me not want to use the tablet.

There's also the fact that Android still just doesn't quite feel like the tablet OS I want sometimes, but this is much more personal. The navigation bars while in landscape mode make it seem like Android is still rough around the edges on a larger screen, for example. While I certainly will not defend Honeycomb as an operating system (slow, buggy, ugly, overwrought), there is little denying that Google took a much more tailored approach to tablets when Andy Rubin was still in charge, even if the navigation bar was still an eyesore. I think we're going in the right direction with Android 5.0 versus 4.4, but the experience still doesn't feel finished. In all honesty, even something like Chrome OS might make more sense to aspire to in terms of visual presentation (well, minus the whole window aspect), and I think we'll see Google explore that in the future as Android and Chrome OS interact on more and deeper levels.

Many of these concerns would be alleviated if Android was the model of a thriving tablet app and content ecosystem, but for all the gains it has made, it just isn't there yet. Yes, many iPad apps now share Android counterparts, but still, all too often these apps have subpar UX because of lazy porting or, puzzlingly, Android's responsive design philosophy and development tools, which encourage developers to create a single scalable, modular experience for all form factors in their applications. In theory, it sounds great. In practice, we end up with things like the official Twitter and Dropbox apps - poorly scaled and unattractive phone apps that don't make good use of tablet screen real estate or layout. Google above all others is to blame here, too: they've often encouraged the belief that a phone app shouldn't really be all that different from a tablet app, even when that philosophy comes at the sometimes openly apparent expense of usability. There's also the huge variety of tablet display sizes and the whole portrait versus landscape debate - this makes designing an app that truly scales well to tablets more difficult.

And lest we forget: for over two years, Google pushed the 16:9 7" form factor into dominance, but now has decided maybe that wasn't the best size. Even 2" of extra diagonal and an aspect ratio change can be the difference between an app that scales well to a tablet and one that looks moderately terrible.

I'm not saying I know how to solve any of these problems - I'm not a designer, software engineer, or developer. They're hard questions. What I do know is that while Apple seems to be successfully indoctrinating its developers in the school of thought that tablet and phone apps should, in fact, be different, Android persists with a philosophy that hasn't really provided us much in the way of benefit, aside perhaps from what we see in Google's own apps. Maybe responsive design has made it a bit easier for developers to build 'tablet' experiences, increasing the overall number of tablet apps on Android, but I'm not sure that's a wholly good thing when a lot of those experiences are crappy. I'm not trying to rag on responsive design, but there's no doubt it is sometimes abused or used as a crutch by some developers, Google included. Whether the net result in terms of UX is, I can't be sure, but it has long been a reason I've preferred the iPad experiences of many apps I use (including some Google ones) to their Android counterparts.

Conclusion

Did Google flub the launch of yet another premium tablet? For all its critical acclaim, the Nexus 10 was also a commercial failure - too big for many, it was also simply ahead of its time considering the decidedly primitive state of tablet content on Android when it launched. Still, it was generally loved by those who bought it because of its low-compromise hardware.

The Nexus 9 is proof that you can design and equip a product to be as cutting-edge as is practical and still end up with something that is unremarkable, even subpar.

Two years later, the Android tablet ecosystem has matured, but it's still very much a fluid situation - no one seems sure whether or not to bet on Android tablets in the long run, OEMs and developers alike. The Nexus 9 is a rather undisguised effort by Google to define what a good Android tablet should be, to try and corral OEMs into developing products that can properly showcase tablet content in an enjoyable, visually consistent manner. Meanwhile, Android 5.0's material design is renewing the push for responsive, beautiful applications, attempting to push developers into getting with the program.

The problem, though, is that the Nexus 9 isn't actually a great tablet. Good, sure, but only in the sense of being baseline satisfactory - the Nexus 9 is proof that you can design and equip a product to be as cutting-edge as is practical and still end up with something that is unremarkable, even subpar. This will do more to scare Android tablet OEMs away from the premium approach than encourage them - if anything, it emboldens the historically stupid and wasteful practice of putting out tablets that are as cheap and often under-equipped as possible with the assumption that consumers simply won't notice or care enough to appreciate the differences. Meanwhile, developers are less likely to commit to the form factor if it doesn't gain traction in the marketplace. It's a chicken-and-egg problem, to be sure, but Google can't even seem to finish the henhouse.

Do I recommend buying a Nexus 9, all abstractions aside? Assuming you get a good one - no light bleed, decent buttons, no weird noises - I think it's about your only reasonable option for a big, high-end Lollipop tablet. Whether Lollipop is all that important to the tablet experience, though, depends on your usage patterns - Samsung's Tab S 8.4 is lighter, much better built, and has a gorgeous screen. And respectable battery life. TouchWiz is a bit of a chore, but honestly, it might still be quicker than stock Android with the Nexus 9's temperamental Tegra K1. It's also $50 cheaper at the moment and adds an SD card slot, a non-trivial feature on a 16GB tablet. If you really want a Lollipop tablet and don't want to get stuck with the Nexus 9, buy a refurbished 2013 Nexus 7 - I quite like the one my girlfriend and I share.

Alternate conclusion: now with more words

I'm adding in a second conclusion (really, some additional thoughts) because I want to talk about tablets more generally - I think it's a discussion quite relevant to the fate of the Nexus 9.

The tablet market at large is in a state of flux, and exactly how things will end up falling out isn't clear yet. Even so, the Nexus 9 simply feels outgunned at a price point that has not generally been lucrative for those who choose to compete in it in the first place, aside from the originator of the modern tablet market: Apple. High-end tablets are a tougher sell than ever in a world where smartphones are two to three times larger in display surface area than they were just four years ago. The display surface area of a Nexus 6 is equal to that of nearly three Nexus Ones. Think about that. This trend has to be hurting sales of 7" Android tablets, previously the de facto Android tablet size as ordained by Google. This is also, I think, why we can probably consider the Nexus 7 as good as dead - I really don't believe it's coming back. Big phones killed it.

But I think they're also hurting sales of all tablets - the numbers for annual tablet shipments in 2014 don't paint a rosy picture, even the iPad is down 18%. For now, the only manufacturer successfully marketing a large, powerful Android tablet is Amazon, the Fire HDX 8.9, but it doesn't publish sales numbers. I think it's fairly safe to say that most of the ones sold are the smaller devices, too, given that they're essentially marketed as a step up from a Kindle e-reader - many of Amazon's customers probably aren't especially interested in a big Kindle, as that's less practical to read on. And, of course, more expensive. So it's not even easy to call Amazon a "success story."

The conundrum Android tablets are facing, and have faced, is one of demand elasticity, and the Nexus 9 is an ideal case in point: how many people are really interested in a $400+ Android tablet? Especially one that really isn't even good value for money? And can we realistically expect Google to build something legitimately great when the return on that investment is likely to be marginal at best? I don't blame Google for trying, certainly, but I think the price and philosophy behind the Nexus 9 were DOA: too little, too late.

Intel is now the world's largest independent vendor of tablet chipsets (and its lead is growing), and the tablets those chips appear in are almost universally below the $400 mark, often well below. Add in the fact that Qualcomm appears to have seriously dropped the ball with Snapdragon 810, and I don't see that trend changing in 2015. Perhaps it's boring, predictable, and a bit dreary to look at it this way, but it seems more likely than ever that in order to really compete, Android tablets will have to have a race to the bottom. Tablets are a decadence on top of the increasingly powerful smartphone's luxury, and as the most popular way to package them affordably dies on the vine (7"), OEMs are struggling to find a balance of value and quality consumers will latch onto.

Google tried to build Android's iPad in a market where even the iPad is clearly starting to struggle.

With Intel manning the parts bin, I think there's a reasonably good chance we'll see more flexible and configurable price points for Android tablets in the future, rather than a few rigid spec reference designs. Intel has a long and controversial history of homogenizing the marketplace, but that's also allowed them to offer a diversity of intercompatible products to their customers. I think if any market is begging for a real product configuration experience, it's tablets - let me decide how much RAM and CPU power I need, along with things like storage and data connectivity. Perhaps, one day, that'll be a real possibility. For now, we're stuck with the given interpretation of the manufacturer, an interpretation that may have resonated if only it were $50 or $100 cheaper - a cost that, perhaps, could have been recouped with a more affordable chipset or less RAM. 90% of the experience at 75% of the cost is a deal a lot of people would be more than happy with.

This, I think, is more than anything the Nexus 9's fundamental problem: Google tried to build Android's iPad in a market where even the iPad is clearly starting to struggle. It didn't help that the Nexus 9 wasn't exactly a homerun to begin with. Perhaps Google will come back this year with something more interesting, something that doesn't simply seek to act as a retort to Apple - I'm honestly not even sure there's a need for Google to build a tablet if their only goal seems to be to tell us "hey, we can do that, too!" The 'iPad killer' has more than anything been the smartphone and the resurgent laptop (including Google’s highly-popular Chromebooks) - the high-end tablet market is actually shrinking as also-rans like LG, Motorola, HTC, and Sony continue to avoid or start to distance themselves from it. Even Samsung seems to be slowing its roll on tablets.

The Nexus 9 was, in theory, the right tablet, but it inarguably came at the wrong time. A year ago, the Nexus 9 might have made more sense, though that's no guarantee it would have really changed anything. Even if the Nexus 9 had the unimpeachable quality of the iPad, the real world performance K1 Denver offered on paper, and class-leading battery life, it's still not clear to me that it would have succeeded. It would have received greater critical acclaim, certainly, but I sincerely doubt the splash it made in the market would have been noticeably larger. Tablets are clearly about to undergo some significant changes to respond to market pressures, while the Nexus 9 will likely have the misfortune of being an overpriced, under-delivering product for the rest of its retail life.