File this one under Left Field. LG just announced a deal to acquire the remnants of webOS. If you’ll recall, webOS sprung to life as Palm’s next generation smartphone platform and answer to the iPhone… before being snapped up by HP. While HP had grand intentions of webOS powering mobile devices, printers, and PCs (!), they abruptly shuttered commercial operations mere weeks their TouchPad launched to lackluster reviews and sales. Fast forward a year or so ahead, past an open source initiative that hasn’t caught on:
To support its next-generation Smart TV technology, LG has entered into a definitive agreement with HP to acquire the source code, associated documentation, engineering talent and related websites associated with webOS. As part of the transaction, LG also will receive licenses under HP’s intellectual property (IP) for use with its webOS products, including patents acquired from Palm covering fundamental operating system and user interface technologies now in broad use across the industry.
Our ongoing fantasy has been that Amazon might pick up webOS to power it’s Kindle Fire line of tablets (and perhaps smartphones). But with Google continuing to develop and freely distribute Android, at least until the Microsoft patent tax gets you, it’s probably most efficient and cost effective to continue skinning as opposed to rolling their own platform. In regards to LG’s hopes, we doubt “webOS” is something that would move the needle in sales, but so-called “smart” televisions could certainly benefit from a more stable and ad-free presentation. Will they deliver?
They’ll put in tv’s, dvr’s, and a tablet to control the home entertainment systems. Don’t be surprised if they turn it into smart home systems.
“Huh.”
What Joel said…
The underlying OS for WebOS, as it is for Android, is Linux. They could have got that for free. So the reason to do this is either the UI, which is meant for a touch interface, the apps (ha!) or the patents.
The patents are worth something of course if they allowed LG to avoid paying the Microsoft tax on Android. But of course they agreed to a deal on that early last year. So they’re going to continue paying Microsoft regardless (I think).
Which leaves the User Interface Layer. If you listen to the Engadget HD podcast you know that Ben Drawbaugh thinks Windows 8 for example, while a reasonable interface for tablets, is a bad interface for a home theater PC. I haven’t used WebOS on a TV obviously but I suspect that WebOS on a TV won’t be a good idea either, without changing the interaction model etc. All those gestures that worked so well on a touch screen? Are they really appropriate for an IR-based remote with a tiny transmitter and tiny receiver?
I suspect LG didn’t pay much and isn’t going to use the OS whole-hog. I think they’re just going to take pieces of it.
With the price of SSDs now in free-fall I wonder how far we are from a basic DVR inside the TV itself.
Even w/o guide data an SSD could be used to buffer live TV, offering instant replay or manual record (for fixed time periods) for different channels.
Glenn, Not sure it’s so simple. TiVo is built on top of linux too, but their play/pause action is exponentially better than most.
Bill, are there SSDs suitable for the continual read/write access a DVR in the traditional sense would require? Hm.
SSDs are great at I/O operations per second and bad at cost per GB and lifespan. HDDs are terrible at IOPS and great at cost per GB and lifespan. Putting a SSD in a DVR doesn’t make use of a SSDs strong points and exposes its weak points. Sticking with HDDs in DVRs is a far better option.
There are some DVR functions (indexing, databases, etc.) that benefit from SSDs and perhaps a dual drive approach with the SSD doing the database work and the HDD doing the recording would be great, but nobody is going to want to pay that much money for a DVR.
DogsOfWar,
All true but eventually all waves come over the wall. SSDs are going to kill off rotating media at some point. Once they’re big enough (most people don’t need 2TB), once they are cheap enough, rotating media is going to go on the decline pretty fast, even without the pressure from ever-thinner laptop designs.
I mean look at newegg.com. Seriously, go there and search for 128GB SSDs. You can find lots and lots of them for around $100. 80GB-150GB laptop drives cost MORE than that, like $129-$149. And the cheapest laptop drives (320GB typically, presumably the sweet spot for the moment) are $49. So an SSD is a $50 adder for a typical laptop TODAY for a typical user.
Sure there are DVRs and desktops and… but laptops now are 75-80% of PC sales and climbing. Sure that means there’s a lot of boxes to put spinning media into still and for a couple more years maybe, but the writing is on the wall.
The Samsung 500GB 840 TLC-based SSD should last over a decade writing 10GB/day, according to reviewers.
It’s under $300 retail now and will likely be under $200 by Xmas (under $100 by Xmas 2014)
I’d think something like the above would work in a “smart” TV, given most would still only offer a 1-year standard warranty.
I’m sure there are ways to tweak provisioning on the above to extend the rated lifespan (at the expense of speed and some capacity)
Glenn,
You are absolutely right about SSDs eventually replacing HDDs. I don’t expect many decent laptops are sold these days with HDDs and even fewer in the coming years.
But the question at hand is DVR functionality in a TV. In that use case price per GB and longevity are paramount.
The cost per GB curve will favor SSDs in a fairly short time as the process size for the etching of SSD chips shrinks and the economies of scale tip in their favor.
Bill in NC,
10GBytes/day is not a realistic duty cycle for a storage device in a DVR.
If a DVR implements buffering (like the TiVo 30 minute buffer) then the throughput is somewhere in the neighborhood of 8GBytes/hour per tuner if all tuners are on a HD channel. On a four tuner DVR that requires 32GBytes/hour or 768GBytes/day of data written to disk. Even a single level cell flash chip will fall over and give up with that kind of throughput. That triple level cell Samsung drive would only last about 1.3 years in this scenario. Keep in mind that in flash chips the fewer levels in a cell the more reliable it is.
For a consumer computer (laptop, desktop, whatever) 10GB/day is great. But the question was about DVRs in televisions.
Again, this would be much more basic than a traditional DVR.
No need for 24/7 operation to buffer live TV or to wake from sleep to do some simple (manual timed) recordings (using the TV’s single tuner)
Would be nice if it could use OTA guide data, but not a deal breaker.
Assuming a duty cycle of 6-8 hours/day that SSD’s expected lifespan would suffice for a standard 1 or even 3 year TV warranty.