I think that overstates the case. "Deathly afraid"?
I do think that Google is concerned about the Fire; a Fire customer is likely a customer lost to the actual Android ecosystem. Google would rather have actual Android devices hold the largest share of the market.
The new Google/Asus tablet will hit the same price point as the Kindle Fire, but will have real Android (4.x, Ice Cream Sandwich) and better hardware. That will pull some of the customers potentially lost to Amazon back to Android.
It's good for us when companies compete for our affections like this. I'm looking forward to getting a tablet with good hardware, a good price, and a clean Android environment (not some vendor-hacked version like what I have on my phone).
I enjoy reading the articles posted on SemiAccurate.com about AMD, nVidia, Intel, etc. Most of the articles are by two writers, and the most entertainingly acerbic ones are by Charlie Demerjian (I'll call him "CD").
Five months ago, CD thought nVidia was going to crush AMD on the high end:
However, nVidia seemingly can't produce their high-end chips in any useful quantity. So, CD snipes at nVidia about that in his comments about the new Radeon HD 7970:
Executive summary (aka TL;DR): AMD has production of high-end chips running smoothly and they are now able to produce, in quantity, chips that are reliable at higher clock rates. AMD is actually shipping a graphics card that performs better than what nVidia is actually shipping.
I think Apple's patents fall into two major categories: "design patents" that cover appearance, and UI innovations that come about because Apple has done a good job of pushing the frontiers of the user experience.
"Rounded corners" and such are an example of the design patents, and that is a whole different category from technology patents like the ones on Java. Does not apply; moving on.
Apple's UI innovation patents, as far as it seems to me (a non-lawyer), are mostly about doing something that hasn't been done before and trying to patent as much of it as possible. Some of these patents are bogus (IMHO the pinch-zoom gesture is an obvious thing to do if you have a multitouch display, so shouldn't be patentable) but some of these might not be bogus.
On the other hand, the Java patents were really weak. The Java Virtual Machine (JVM) was hardly the first VM ever; the UCSD "p-System" VM is over three decades old, so Sun couldn't patent the basic idea of a VM to let programs run anywhere. So they patented a few aspects of Java, and then Oracle claimed in court that the Dalvik VM infringed those patents. But I've read several analyses of these patents, and they pretty much agreed that the patents were weak. It seems the court agreed.
Finally, why should Google pay Oracle? Google is using a different VM, all new and all original code. Google isn't using the Java trademark, and doesn't have any agreement with Oracle. As people have observed here on Slashdot: If you want to argue that Oracle "owns" Java so completely that nobody may copy it, then maybe the creators of the C programming language and the C standard library could collect staggering royalties from pretty much the whole world.
Google making Dalvik over the objections of Oracle is just like Dodge making a car over the objections of Ford. You can see why there might be objections, but society shouldn't interfere.
The original series laid down the example of what Star Trek should be. At its best, TOS rocked.
The ideal episode (which does not exist) would somehow contain all these elements: exploration; dealing with aliens; banter among Kirk, Spock, and McCoy; Scotty performing miracles; the Enterprise doing cool stuff; actual science fiction ideas; and a certain amount of comedy. Since you can't get all that in one episode, here are a few to consider. I'll try to keep these free of essential spoilers, for those of you who haven't gotten around to watching TOS yet.
"The Doomsday Machine" -- my all-time favorite epsiode. A bit of exploration, tense drama with Commodore Decker, actually plausible science behind the SF ideas, and a tight plot that makes sense from beginning to end. Plus the best cliffhanger ever: "Gentlemen, I suggest you beam me aboard..."
"The City on the Edge of Forever" -- loses points for not being typical Trek, but it's awesome SF, with a dramatic love story. Also, if I had to pick one episode that best qualifies as a "chick flick" it would be this one!
"The Trouble with Tribbles" -- pure fun. Despite being almost completely a comedy episode, there is actually a plot and it is resolved in a satisfactory manner. Uhura, Chekov, and Scotty all get character moments.
"Balance of Terror" -- a great drama with some great surprises, and the moral themes it presents are naturally in the story and don't feel forced. Mark Lenard is great as the Commander.
"The Naked Time" -- a strange affliction causes the crew of the Enterprise to start acting oddly. Lots of character moments.
"Spectre of the Gun" -- a really unusual episode and an example of making a virtue of necessity. Kirk and an away team seemingly find themselves in the "Wild West" of the late 19th century; there wasn't enough money to film this realistically, so instead it was filmed in a surreal, fantastic fashion. The sets are incomplete, with missing walls, and instead of trying to hide this they left it right in the open. It's actually pretty cool.
This was the first science fiction book I ever read, and it still has an important place in my heart. It's out of print, but trust me, it's worth finding a used copy and buying it. It's probably worth it to buy a hardcover; the mass-market paperback (from "Scholastic Book Services") was printed on very cheap paper that is turning brown these days.
The story: a family of three (father, mother, son) has been living on Ganymede, but will now move to Earth. The son grew up on Ganymede with a robot, Rex, as a nanny/companion. Since shipping is expensive, they plan to sell Rex and leave him behind. Hating to leave Rex, the boy gets off the space liner just before it departs, and runs away; the boy and the robot have adventures as they try to get to Earth together. It's a tale of adventure and loyalty and love. The story is narrated by Rex in the first person.
At a rumored 1280x800 res at 7", that's 215dpi, which certainly sounds close enough to "retina".
That's highly interesting! Apple is now claiming that 220 DPI is a "retina display" for their new MacBook Pro model, so maybe the Nexus 7 will have a "retina display" after all.
I think it is quite clear that Google is on a mission here, and that mission is: ship a compelling tablet at the exact same price as a Kindle Fire.
Would an 8" screen make the tablet more compelling? IMHO, not really.
Would an 8" screen make it harder to hit the $199 price point? Yes. Not only are larger screens more expensive as a general rule, but the massive number of 7" devices already on the market mean that there should be multiple possible sources for a 7" screen, and volumes should help keep the cost down.
Will the Nexus 7 have a retina display? No, absolutely not, because there is no way they could hit the $199 price point.
Will the Nexus 7 have a GPS? Only if it can have a GPS and still hit the $199 price point.
Will the Nexus 7 have a camera? According to one of the linked articles, it will have a forward-facing camera for video conferencing, and will not have a rearward-facing camera to save on costs. As I already have a phone with a rearward-facing camera, a rearward-facing camera isn't that compelling IMHO, and I think Google made exactly the correct call here.
How much flash will it have? As much as it can have while still costing $199. I predict 8 GB.
My apologies; I don't know why, but I thought you said "twice as much as a netbook". You didn't say that, it's my fault for mis-reading, and I'm sorry about the confusion.
There are some netbooks for $200, but the ones I could find (using Google Shopping) all have moving-parts hard drives. This is a direct result of the campaign to ship Windows on netbooks instead of shipping a custom Linux build with just web browser and email and such. When netbooks were new, the cheapest ones all had flash storage; now you need to spend more for flash storage (but at least you will get more capacity also).
If you want an SSD, you would either need a vendor that will custom-build, you will need the school to swap out the hard drive, or you will need to spend a lot more than $200. I think realistically, if a school goes with the $200 netbook, they are going to leave the hard drive alone. Maybe it will be okay.
I once had to swap out a hard drive from a friend's computer, because the friend's small children had played too aggressively in the same room and head-crashed the hard drive. (The kids didn't bash the computer; as far as we could figure out, they just made the floor boards flex, and bounced the computer!) But a modern netbook's hard drive should be more robust; maybe it will be okay.
At least one good point about netbooks for kids: kids have small hands, and the small keyboard won't bother them.
Well, the #1 problem with that is it almost certainly won't run edubuntu without a fair degree of hassle.
Agreed. Also, netbooks almost certainly won't run kids' apps from the Android Market (or Google Play or whatever it is now).
Although, I think Android will run an X server, and you might be able to run Edubuntu remotely... but I'm not seriously suggesting that.
I am suggesting that Android apps on an Android tablet might be a better solution than Edubuntu on cheap laptops. Not because I hate Edubuntu, but because I love durable hardware.
A lightweight, one-piece tablet with no moving parts is a better match for elementary school kids than a laptop with a cooling fan that can suck dirt inside, than a moving-parts hard drive that can head crash if the laptop is dropped, and than a clamshell design with hinges as a weak point.
For kids, I would suggest putting the tablet inside an inexpensive protective case, so that if the device is dropped, the corners are padded and the screen might not make contact with the ground.
So now you've got something where the base model costs as much as a netbook
Really? I'm expecting the Google tablet to cost $200 for the base (and only) model. Are you telling me that schools can buy a netbook for $100?
you have to buy extra, detachable, losable (this is a school) hardware bumbing the price up further,
That word "bumbing" is interesting. It's new to me.
But yes, essentially correct.
and some dodgy app for word processing. How reliable is that app going to be?
Sorry, but you are grasping at straws here. How reliable do you think a business app is going to be? If it's "dodgy" it won't sell well. And how hard is it to support the word processing needs of elementary school students?
And aside from the dodgy app, there is Google Docs or whatever via web browser.
Would you suggest that the GP deploy a bunch of rooted custom-rom nooks to a bunch of schoolkids?
Actually, no. Which is why I didn't suggest any such thing.
Of course, for a sufficiently low-budget school, such an approach wouldn't be completely crazy. You can buy a brand-new Nook Color right now for $140, and the rooting process is pretty easy. But it doesn't really make sense because the new Google tablet is supposed to come out soon, and it will be $200 for a much better device and no need for dodgy "rooting" and such.
I think that an Android tablet might be a good way to go: very compact and lightweight, durable (no moving parts such as a hard disk or cooling fan), and very long battery life. Less expensive than a laptop, and you could buy accessories and software with the left-over money: get some sort of keyboard and Android software for word processing and such.
Asus and Google are going to announce a low-cost Android tablet. The rumored specs are: 7" screen, Tegra 3 processor at 1.3 GHz (that's 4 general-purpose cores), probably 1 GB of RAM and probably 8 GB of flash storage. Expected price will be $250 or $200.
I have a Nook Color that I rooted, and installed "PhireMod 7.2" (a particular build of CyanogenMod 7). I am very pleased with my 7" tablet. It's big enough to be useful and small enough to carry around, and I love the battery life.
I have an Acer laptop with an 11.6" screen and I am very fond of it. The size and weight are great.
The model I have is no longer made, but the Acer Aspire One series is still made.
Most of those seem to have an Atom chip. I also have an Acer with an Atom and I pretty much hate the Atom... very slow. It's possible that newer Atom chips suck less.
I haven't tested the AMD "E" chips yet, but here is an Acer Aspire One with a dual-core "E" chip.
The only thing is that for kids, it might be better to have an SSD rather than a spinning-metal hard disk, but that model has a 320 GB spinning-metal hard disk.
Oh, on at least my Acer laptop, modern Linux distros like Ubuntu or Mint just work. All hardware detected correctly, WiFi works out of the box, etc. As I said, I haven't tried the newest one so I can't promise anything for sure.
RasPi: $25. Monitor: About 100. Mouse/keyboard: 20 or so. Power supply: 5. Speakers: 5. SD card: we'll say about 20. So we're talking about $175, total for a 700 Mhz machine. I'll bet you could do as well on Craigslist looking for used laptops.
Or you could get the monitor on Craig's List and save money, if Craig's List is an option. And in a year or two, a kid might be able to pick up a whole RPi setup including monitor on Craig's List.
In short, it is not fair to compare the straight-up full cost of an RPi with the cost of a random used laptop, possibly old. Heck, I once got a computer free because the owner got a new one. Should I complain that the RPi isn't free?
A school will be able to equip a whole classroom lab with a bunch of RPi computers. Each student will have his/her own to use during class, and at $25, any student who wants to buy one for home use realistically can.
So, students will show up to class, each insert their own personal SD card, and boot up the RPi. Then do development in Python, or Scratch, or Logo, or whatever teaching language the school prefers. Many of the students can have an RPi at home and can play with it there.
The idea of "buy a cheap used laptop on Craig's List" doesn't scale for a whole classroom. With the RPi, each device is identical to each other device, and it is really cheap to have a bunch of hot spares put away.
This is a device to inspire future geeks, not teach the masses how to program. I think that's the idea.
Well, you could always try going to the official web site and see if there is anything here:
Oh, they want to enable computer classes to focus on computer stuff (as opposed to training in how to use Microsoft Office), and they want every family to be able to afford a computer that can access the Internet.
I categorize music into two major classes: "work music" and "non-work music". "Work music" is music that is suitable for work. It must not call too much attention to itself; most of my work music is instrumental, and most of it is familiar to me from having listened to it many times.
Some of my music has crazy sound effects or other avant-garde stuff that makes it unsuitable to be used as work music. Some of it is great for waking you up when you are sleepy, but far too distracting to be work music. (You might be different from me; maybe you can concentrate while rocking out to loud, hard music. I can't.)
Some of my favorite work music is "jazz fusion", jazzy music with a strong beat (the name means "fusion of jazz with rock-and-roll").
Yeah, maybe it is theoretically best to be in an acoustically quiet environment with no distractions. But my familiar work music is much less distracting than all the sounds of the people around me. And I'm in a relatively quiet office environment; I'm one of the lucky ones now.
Back when I spent a year as a temp, and I had to share a single office with one, later two, and eventually three other people? Only my music and some Sennheiser HD-280 pro headphones saved my productivity and my sanity. (Sennheiser claims 30 dB of acoustic isolation from the HD-280 pro headphones. That might be high, but they do a great job overall of blocking out background noise.)
I remember when the CEO of Handspring announced that smart phones were the future for Handspring, and sales of the Visor PDA went almost to zero immediately, and sales of Visor accessories (Springboard cards, etc.) also went almost to zero immediately.
The Cius Tablet has been shipping for less than a year, and the CEO just announced that no further development will ever be done. The chances of anyone getting interested in this now: 0%
And wow, Android 2.2 on a $750 device with a 7 inch screen and a funky Intel chip? 680 grams (about 1.5 pounds)? The review didn't say anything about an ARM emulator so I assume any Android apps with native code for ARM just won't run on this thing. I'd sooner put CyanogenMod 7 on an old Nook Color. 448 grams (just under a pound) by the way.
(Oh wait, I already did that. A Nook Color makes a surprisingly nice Android tablet! It is a lot faster with CM7 than with the factory Nook software.)
The fact that the CEO was willing to Handspring this device probably means that the sales were already close to zero, so he didn't feel there were any sales left to discourage.
I don't read German, but Google Translate does. Looks like energy costs have gone up by 57% in the past decade; taxes on energy have gone up 1000% in the last 15 years.
"The de-industrialization has already begun," Energy Commissioner Guenther Oettinger warned in an interview with the Handelsblatt.
This ad is reproduced in the book To Catch a Mouse, Make a Noise Like a Cheese by Lewis Kornfeld. In that book, Kornfeld shares his thoughts on marketing, and as part of that, he shows a few ads and then makes comments about them.
He was impressed (in a negative way) that Steve Jobs had his picture inserted in the ad many times. (This was pre-Mac so Steve Jobs wasn't famous yet; Kornfeld was a pioneer in the field of calling Steve Jobs on his ego.)
Also, he was displeased by the tagline: "When we invented the personal computer, we invented a new kind of bicycle" He had something to do with the TRS-80, which predated the Apple ][, so he thought it was rather cheeky for Apple/Jobs to claim to have invented the personal computer.
By the way, Lew Kornfeld was a big part of the project to get a special Superman comic book to try to get kids interested in computers. It's cheesy but perhaps not fully deserving of all the vitriol it gets: http://io9.com/5835188/that-stupid-time-superman-shilled-computers-for-radio-shack He talks about how that comic book happened in the book.
To Catch a Mouse, Make a Noise Like a Cheese is out of print, but I think it's a classic and worth finding a used copy.
All I know is what Google knows. I don't know anything new.
Have you seen these? If this rumors site is to be believed, nVidia is slipping its schedule, which would explain why they are not talking much right now.
I don't know much about this rumors site, but this guy sure seems to like criticizing nVidia.
It just makes me think of Kodak, who wouldn't go after the digital market because they were so profitable in the traditional film market...
That's a good point.
Maybe Intel has learned their lesson, but remember how Intel got in trouble two ways and had to play catch-up: Intel tried to push Itanium for 64-bit computing (since nobody but Intel could make an Itanium; no cross-licensing deals), and Intel went full-tilt for Pentium 4. AMD shipped a 64-bit x86 chip and Intel had to scramble, and AMD was shipping chips with much more power-per-clock than a Pentium 4, and Intel had to scramble.
With the Pentium 4, Intel was saved by a small group (in Israel, if I recall correctly) that had been trying to make a good laptop chip; it turned into the Core architecture, much better than the Pentium 4 design. So Intel has already seen the value of having an alternative ready-to-go in-house. Maybe they will take ARM seriously and make a top-secret in-house ARM super-chip just in case?
You want Intel to make an ARM chip that is competitive with x86. Intel will never, ever do that if they can possibly avoid it.
Intel dominates in x86, and they make good profits on x86 chips. As noted in TFA, Intel would be just another ARM source out of many, and they would make less on an ARM than on x86. nVidia, on the other hand, is no longer friendly with Intel and has no reason not to build a super ARM as you would like; and in fact they seem to be working on it. Google search for "Project Denver".
The first Project Denver silicon is rumored to be 8 ARM cores, 64-bit, with a 256 CUDA core GPU on the same die. I would love a smartbook with that chip, but I think they will be able to sell that as a blade server platform as well.
some people with engineering or physical science backgrounds seem to underestimate these human problems.
He didn't say we need a government of philosopher-kings who can command us to do the right thing; he said politicians need to let the actual facts influence their plans, which I think is hard to argue against. And America is a republic, where the people have the ability to get rid of politicians they don't like, which is not the case in China.
Massie recalls Sununu saying, "We need more engineers and fewer lawyers" in politics. As Massie explains, "Lawyers are taught to take a position, whether it's right or wrong ideologically, and defend it—to go collect facts to support it. Whereas engineers are taught the inverse of that, they're taught to collect facts and then come up with an answer based on the facts. He said, 'That's the kind of thought process we need more of in government.' On the stump, that's what I'm trying to convey, that we need more problem solvers in Washington, DC."
Somebody mod above post up, please. TheRaven64 knows compiler stuff.
It seems possible that the reason my code is faster is due to the autovectorization stuff. DSP code generally is a good candidate for vectorizing, and my code is straight C with no explicit use of SIMD features.
The quality of the generated code, however, is significantly worse, at least at this time.
At work, I use Clang to compile my DSP code, and the result is faster than if I use GCC. I haven't rigorously analyzed the situation but the speedup is on the order of 5% to 10% for my code; not dramatic, but nothing to sneer at either, and all I had to do was edit my makefile to specify Clang.
So I find the blanket statement that Clang generates "significantly worse" code rather surprising. My own experience says otherwise.
Slashdot Mirror
Google is deathly afraid of the Kindle Fire.
I think that overstates the case. "Deathly afraid"?
I do think that Google is concerned about the Fire; a Fire customer is likely a customer lost to the actual Android ecosystem. Google would rather have actual Android devices hold the largest share of the market.
The new Google/Asus tablet will hit the same price point as the Kindle Fire, but will have real Android (4.x, Ice Cream Sandwich) and better hardware. That will pull some of the customers potentially lost to Amazon back to Android.
It's good for us when companies compete for our affections like this. I'm looking forward to getting a tablet with good hardware, a good price, and a clean Android environment (not some vendor-hacked version like what I have on my phone).
steveha
I enjoy reading the articles posted on SemiAccurate.com about AMD, nVidia, Intel, etc. Most of the articles are by two writers, and the most entertainingly acerbic ones are by Charlie Demerjian (I'll call him "CD").
Five months ago, CD thought nVidia was going to crush AMD on the high end:
http://semiaccurate.com/2012/01/19/nvidia-kepler-vs-amd-gcn-has-a-clear-winner/
However, nVidia seemingly can't produce their high-end chips in any useful quantity. So, CD snipes at nVidia about that in his comments about the new Radeon HD 7970:
http://semiaccurate.com/2012/06/21/amd-launches-tahiti-2-aka-hd7970ghz-edition/
Executive summary (aka TL;DR): AMD has production of high-end chips running smoothly and they are now able to produce, in quantity, chips that are reliable at higher clock rates. AMD is actually shipping a graphics card that performs better than what nVidia is actually shipping.
steveha
I think Apple's patents fall into two major categories: "design patents" that cover appearance, and UI innovations that come about because Apple has done a good job of pushing the frontiers of the user experience.
"Rounded corners" and such are an example of the design patents, and that is a whole different category from technology patents like the ones on Java. Does not apply; moving on.
Apple's UI innovation patents, as far as it seems to me (a non-lawyer), are mostly about doing something that hasn't been done before and trying to patent as much of it as possible. Some of these patents are bogus (IMHO the pinch-zoom gesture is an obvious thing to do if you have a multitouch display, so shouldn't be patentable) but some of these might not be bogus.
On the other hand, the Java patents were really weak. The Java Virtual Machine (JVM) was hardly the first VM ever; the UCSD "p-System" VM is over three decades old, so Sun couldn't patent the basic idea of a VM to let programs run anywhere. So they patented a few aspects of Java, and then Oracle claimed in court that the Dalvik VM infringed those patents. But I've read several analyses of these patents, and they pretty much agreed that the patents were weak. It seems the court agreed.
Finally, why should Google pay Oracle? Google is using a different VM, all new and all original code. Google isn't using the Java trademark, and doesn't have any agreement with Oracle. As people have observed here on Slashdot: If you want to argue that Oracle "owns" Java so completely that nobody may copy it, then maybe the creators of the C programming language and the C standard library could collect staggering royalties from pretty much the whole world.
Google making Dalvik over the objections of Oracle is just like Dodge making a car over the objections of Ford. You can see why there might be objections, but society shouldn't interfere.
steveha
The original series laid down the example of what Star Trek should be. At its best, TOS rocked.
The ideal episode (which does not exist) would somehow contain all these elements: exploration; dealing with aliens; banter among Kirk, Spock, and McCoy; Scotty performing miracles; the Enterprise doing cool stuff; actual science fiction ideas; and a certain amount of comedy. Since you can't get all that in one episode, here are a few to consider. I'll try to keep these free of essential spoilers, for those of you who haven't gotten around to watching TOS yet.
"The Doomsday Machine" -- my all-time favorite epsiode. A bit of exploration, tense drama with Commodore Decker, actually plausible science behind the SF ideas, and a tight plot that makes sense from beginning to end. Plus the best cliffhanger ever: "Gentlemen, I suggest you beam me aboard..."
"The City on the Edge of Forever" -- loses points for not being typical Trek, but it's awesome SF, with a dramatic love story. Also, if I had to pick one episode that best qualifies as a "chick flick" it would be this one!
"The Trouble with Tribbles" -- pure fun. Despite being almost completely a comedy episode, there is actually a plot and it is resolved in a satisfactory manner. Uhura, Chekov, and Scotty all get character moments.
"Balance of Terror" -- a great drama with some great surprises, and the moral themes it presents are naturally in the story and don't feel forced. Mark Lenard is great as the Commander.
"The Naked Time" -- a strange affliction causes the crew of the Enterprise to start acting oddly. Lots of character moments.
"Spectre of the Gun" -- a really unusual episode and an example of making a virtue of necessity. Kirk and an away team seemingly find themselves in the "Wild West" of the late 19th century; there wasn't enough money to film this realistically, so instead it was filmed in a surreal, fantastic fashion. The sets are incomplete, with missing walls, and instead of trying to hide this they left it right in the open. It's actually pretty cool.
steveha
This was the first science fiction book I ever read, and it still has an important place in my heart. It's out of print, but trust me, it's worth finding a used copy and buying it. It's probably worth it to buy a hardcover; the mass-market paperback (from "Scholastic Book Services") was printed on very cheap paper that is turning brown these days.
The story: a family of three (father, mother, son) has been living on Ganymede, but will now move to Earth. The son grew up on Ganymede with a robot, Rex, as a nanny/companion. Since shipping is expensive, they plan to sell Rex and leave him behind. Hating to leave Rex, the boy gets off the space liner just before it departs, and runs away; the boy and the robot have adventures as they try to get to Earth together. It's a tale of adventure and loyalty and love. The story is narrated by Rex in the first person.
http://www.amazon.com/The-Runaway-Robot-Lester-Del/dp/B000DZDQD0
I believe I was 8 when I read this. I loved it then and I still love it today.
steveha
At a rumored 1280x800 res at 7", that's 215dpi, which certainly sounds close enough to "retina".
That's highly interesting! Apple is now claiming that 220 DPI is a "retina display" for their new MacBook Pro model, so maybe the Nexus 7 will have a "retina display" after all.
http://en.wikipedia.org/wiki/Retina_display
But it turns out that Apple is trying to trademark "retina display" so maybe we will not be allowed to use that term on the Nexus 7.
1280x800, plus a Tegra 3, means we will be able to watch HD content in 720p. Nice!
steveha
I think it is quite clear that Google is on a mission here, and that mission is: ship a compelling tablet at the exact same price as a Kindle Fire.
Would an 8" screen make the tablet more compelling? IMHO, not really.
Would an 8" screen make it harder to hit the $199 price point? Yes. Not only are larger screens more expensive as a general rule, but the massive number of 7" devices already on the market mean that there should be multiple possible sources for a 7" screen, and volumes should help keep the cost down.
Will the Nexus 7 have a retina display? No, absolutely not, because there is no way they could hit the $199 price point.
Will the Nexus 7 have a GPS? Only if it can have a GPS and still hit the $199 price point.
Will the Nexus 7 have a camera? According to one of the linked articles, it will have a forward-facing camera for video conferencing, and will not have a rearward-facing camera to save on costs. As I already have a phone with a rearward-facing camera, a rearward-facing camera isn't that compelling IMHO, and I think Google made exactly the correct call here.
How much flash will it have? As much as it can have while still costing $199. I predict 8 GB.
I think you get the idea.
steveha
Not sure I follow: netbooks go for about $200 up
My apologies; I don't know why, but I thought you said "twice as much as a netbook". You didn't say that, it's my fault for mis-reading, and I'm sorry about the confusion.
There are some netbooks for $200, but the ones I could find (using Google Shopping) all have moving-parts hard drives. This is a direct result of the campaign to ship Windows on netbooks instead of shipping a custom Linux build with just web browser and email and such. When netbooks were new, the cheapest ones all had flash storage; now you need to spend more for flash storage (but at least you will get more capacity also).
If you want an SSD, you would either need a vendor that will custom-build, you will need the school to swap out the hard drive, or you will need to spend a lot more than $200. I think realistically, if a school goes with the $200 netbook, they are going to leave the hard drive alone. Maybe it will be okay.
I once had to swap out a hard drive from a friend's computer, because the friend's small children had played too aggressively in the same room and head-crashed the hard drive. (The kids didn't bash the computer; as far as we could figure out, they just made the floor boards flex, and bounced the computer!) But a modern netbook's hard drive should be more robust; maybe it will be okay.
At least one good point about netbooks for kids: kids have small hands, and the small keyboard won't bother them.
steveha
Well, the #1 problem with that is it almost certainly won't run edubuntu without a fair degree of hassle.
Agreed. Also, netbooks almost certainly won't run kids' apps from the Android Market (or Google Play or whatever it is now).
Although, I think Android will run an X server, and you might be able to run Edubuntu remotely... but I'm not seriously suggesting that.
I am suggesting that Android apps on an Android tablet might be a better solution than Edubuntu on cheap laptops. Not because I hate Edubuntu, but because I love durable hardware.
A lightweight, one-piece tablet with no moving parts is a better match for elementary school kids than a laptop with a cooling fan that can suck dirt inside, than a moving-parts hard drive that can head crash if the laptop is dropped, and than a clamshell design with hinges as a weak point.
For kids, I would suggest putting the tablet inside an inexpensive protective case, so that if the device is dropped, the corners are padded and the screen might not make contact with the ground.
So now you've got something where the base model costs as much as a netbook
Really? I'm expecting the Google tablet to cost $200 for the base (and only) model. Are you telling me that schools can buy a netbook for $100?
you have to buy extra, detachable, losable (this is a school) hardware bumbing the price up further,
That word "bumbing" is interesting. It's new to me.
But yes, essentially correct.
and some dodgy app for word processing. How reliable is that app going to be?
Sorry, but you are grasping at straws here. How reliable do you think a business app is going to be? If it's "dodgy" it won't sell well. And how hard is it to support the word processing needs of elementary school students?
And aside from the dodgy app, there is Google Docs or whatever via web browser.
Would you suggest that the GP deploy a bunch of rooted custom-rom nooks to a bunch of schoolkids?
Actually, no. Which is why I didn't suggest any such thing.
Of course, for a sufficiently low-budget school, such an approach wouldn't be completely crazy. You can buy a brand-new Nook Color right now for $140, and the rooting process is pretty easy. But it doesn't really make sense because the new Google tablet is supposed to come out soon, and it will be $200 for a much better device and no need for dodgy "rooting" and such.
steveha
I think that an Android tablet might be a good way to go: very compact and lightweight, durable (no moving parts such as a hard disk or cooling fan), and very long battery life. Less expensive than a laptop, and you could buy accessories and software with the left-over money: get some sort of keyboard and Android software for word processing and such.
Asus and Google are going to announce a low-cost Android tablet. The rumored specs are: 7" screen, Tegra 3 processor at 1.3 GHz (that's 4 general-purpose cores), probably 1 GB of RAM and probably 8 GB of flash storage. Expected price will be $250 or $200.
http://www.pcworld.com/article/257296/googles_sub250_nexus_7_tablet_coming_late_june_report_says.html
I have a Nook Color that I rooted, and installed "PhireMod 7.2" (a particular build of CyanogenMod 7). I am very pleased with my 7" tablet. It's big enough to be useful and small enough to carry around, and I love the battery life.
steveha
I have an Acer laptop with an 11.6" screen and I am very fond of it. The size and weight are great.
The model I have is no longer made, but the Acer Aspire One series is still made.
Most of those seem to have an Atom chip. I also have an Acer with an Atom and I pretty much hate the Atom... very slow. It's possible that newer Atom chips suck less.
I haven't tested the AMD "E" chips yet, but here is an Acer Aspire One with a dual-core "E" chip.
http://www.newegg.com/Product/Product.aspx?Item=N82E16834215340
The only thing is that for kids, it might be better to have an SSD rather than a spinning-metal hard disk, but that model has a 320 GB spinning-metal hard disk.
Oh, on at least my Acer laptop, modern Linux distros like Ubuntu or Mint just work. All hardware detected correctly, WiFi works out of the box, etc. As I said, I haven't tried the newest one so I can't promise anything for sure.
steveha
RasPi: $25. Monitor: About 100. Mouse/keyboard: 20 or so. Power supply: 5. Speakers: 5. SD card: we'll say about 20. So we're talking about $175, total for a 700 Mhz machine. I'll bet you could do as well on Craigslist looking for used laptops.
Or you could get the monitor on Craig's List and save money, if Craig's List is an option. And in a year or two, a kid might be able to pick up a whole RPi setup including monitor on Craig's List.
In short, it is not fair to compare the straight-up full cost of an RPi with the cost of a random used laptop, possibly old. Heck, I once got a computer free because the owner got a new one. Should I complain that the RPi isn't free?
A school will be able to equip a whole classroom lab with a bunch of RPi computers. Each student will have his/her own to use during class, and at $25, any student who wants to buy one for home use realistically can.
So, students will show up to class, each insert their own personal SD card, and boot up the RPi. Then do development in Python, or Scratch, or Logo, or whatever teaching language the school prefers. Many of the students can have an RPi at home and can play with it there.
The idea of "buy a cheap used laptop on Craig's List" doesn't scale for a whole classroom. With the RPi, each device is identical to each other device, and it is really cheap to have a bunch of hot spares put away.
This is a device to inspire future geeks, not teach the masses how to program. I think that's the idea.
Well, you could always try going to the official web site and see if there is anything here:
http://www.raspberrypi.org/about
Oh, they want to enable computer classes to focus on computer stuff (as opposed to training in how to use Microsoft Office), and they want every family to be able to afford a computer that can access the Internet.
steveha
I categorize music into two major classes: "work music" and "non-work music". "Work music" is music that is suitable for work. It must not call too much attention to itself; most of my work music is instrumental, and most of it is familiar to me from having listened to it many times.
Some of my music has crazy sound effects or other avant-garde stuff that makes it unsuitable to be used as work music. Some of it is great for waking you up when you are sleepy, but far too distracting to be work music. (You might be different from me; maybe you can concentrate while rocking out to loud, hard music. I can't.)
Some of my favorite work music is "jazz fusion", jazzy music with a strong beat (the name means "fusion of jazz with rock-and-roll").
Yeah, maybe it is theoretically best to be in an acoustically quiet environment with no distractions. But my familiar work music is much less distracting than all the sounds of the people around me. And I'm in a relatively quiet office environment; I'm one of the lucky ones now.
Back when I spent a year as a temp, and I had to share a single office with one, later two, and eventually three other people? Only my music and some Sennheiser HD-280 pro headphones saved my productivity and my sanity. (Sennheiser claims 30 dB of acoustic isolation from the HD-280 pro headphones. That might be high, but they do a great job overall of blocking out background noise.)
steveha
I call this technique "regeneration" and think that, in the interests of population control, people should be limited to 12 of them.
Okay, admit it: you didn't think of this yourself! Did you get this idea from some sort of Doctor? WHO?
You need to imagine Strong Bad's voice saying it for full effect:
HANDSPRING'D!!!
I remember when the CEO of Handspring announced that smart phones were the future for Handspring, and sales of the Visor PDA went almost to zero immediately, and sales of Visor accessories (Springboard cards, etc.) also went almost to zero immediately.
The Cius Tablet has been shipping for less than a year, and the CEO just announced that no further development will ever be done. The chances of anyone getting interested in this now: 0%
And wow, Android 2.2 on a $750 device with a 7 inch screen and a funky Intel chip? 680 grams (about 1.5 pounds)? The review didn't say anything about an ARM emulator so I assume any Android apps with native code for ARM just won't run on this thing. I'd sooner put CyanogenMod 7 on an old Nook Color. 448 grams (just under a pound) by the way.
(Oh wait, I already did that. A Nook Color makes a surprisingly nice Android tablet! It is a lot faster with CM7 than with the factory Nook software.)
The fact that the CEO was willing to Handspring this device probably means that the sales were already close to zero, so he didn't feel there were any sales left to discourage.
I don't read German, but Google Translate does. Looks like energy costs have gone up by 57% in the past decade; taxes on energy have gone up 1000% in the last 15 years.
http://www.handelsblatt.com/politik/deutschland/kostenexplosion-merkels-strompreisluege-seite-all/6663536-all.html
steveha
This ad is reproduced in the book To Catch a Mouse, Make a Noise Like a Cheese by Lewis Kornfeld. In that book, Kornfeld shares his thoughts on marketing, and as part of that, he shows a few ads and then makes comments about them.
He was impressed (in a negative way) that Steve Jobs had his picture inserted in the ad many times. (This was pre-Mac so Steve Jobs wasn't famous yet; Kornfeld was a pioneer in the field of calling Steve Jobs on his ego.)
Also, he was displeased by the tagline: "When we invented the personal computer, we invented a new kind of bicycle" He had something to do with the TRS-80, which predated the Apple ][, so he thought it was rather cheeky for Apple/Jobs to claim to have invented the personal computer.
By the way, Lew Kornfeld was a big part of the project to get a special Superman comic book to try to get kids interested in computers. It's cheesy but perhaps not fully deserving of all the vitriol it gets: http://io9.com/5835188/that-stupid-time-superman-shilled-computers-for-radio-shack He talks about how that comic book happened in the book.
To Catch a Mouse, Make a Noise Like a Cheese is out of print, but I think it's a classic and worth finding a used copy.
steveha
Have you seen any updates recently?
All I know is what Google knows. I don't know anything new.
Have you seen these? If this rumors site is to be believed, nVidia is slipping its schedule, which would explain why they are not talking much right now.
I don't know much about this rumors site, but this guy sure seems to like criticizing nVidia.
http://semiaccurate.com/2011/08/05/what-is-project-denver-based-on/
http://semiaccurate.com/2011/10/19/nvidia-tegra-roadmap-slips-a-year/
steveha
It just makes me think of Kodak, who wouldn't go after the digital market because they were so profitable in the traditional film market...
That's a good point.
Maybe Intel has learned their lesson, but remember how Intel got in trouble two ways and had to play catch-up: Intel tried to push Itanium for 64-bit computing (since nobody but Intel could make an Itanium; no cross-licensing deals), and Intel went full-tilt for Pentium 4. AMD shipped a 64-bit x86 chip and Intel had to scramble, and AMD was shipping chips with much more power-per-clock than a Pentium 4, and Intel had to scramble.
With the Pentium 4, Intel was saved by a small group (in Israel, if I recall correctly) that had been trying to make a good laptop chip; it turned into the Core architecture, much better than the Pentium 4 design. So Intel has already seen the value of having an alternative ready-to-go in-house. Maybe they will take ARM seriously and make a top-secret in-house ARM super-chip just in case?
I doubt it, but it could happen, I guess.
steveha
You want Intel to make an ARM chip that is competitive with x86. Intel will never, ever do that if they can possibly avoid it.
Intel dominates in x86, and they make good profits on x86 chips. As noted in TFA, Intel would be just another ARM source out of many, and they would make less on an ARM than on x86. nVidia, on the other hand, is no longer friendly with Intel and has no reason not to build a super ARM as you would like; and in fact they seem to be working on it. Google search for "Project Denver".
The first Project Denver silicon is rumored to be 8 ARM cores, 64-bit, with a 256 CUDA core GPU on the same die. I would love a smartbook with that chip, but I think they will be able to sell that as a blade server platform as well.
steveha
some people with engineering or physical science backgrounds seem to underestimate these human problems.
He didn't say we need a government of philosopher-kings who can command us to do the right thing; he said politicians need to let the actual facts influence their plans, which I think is hard to argue against. And America is a republic, where the people have the ability to get rid of politicians they don't like, which is not the case in China.
steveha
Massie recalls Sununu saying, "We need more engineers and fewer lawyers" in politics. As Massie explains, "Lawyers are taught to take a position, whether it's right or wrong ideologically, and defend it—to go collect facts to support it. Whereas engineers are taught the inverse of that, they're taught to collect facts and then come up with an answer based on the facts. He said, 'That's the kind of thought process we need more of in government.' On the stump, that's what I'm trying to convey, that we need more problem solvers in Washington, DC."
I wholeheartedly approve of this idea.
steveha
Somebody mod above post up, please. TheRaven64 knows compiler stuff.
It seems possible that the reason my code is faster is due to the autovectorization stuff. DSP code generally is a good candidate for vectorizing, and my code is straight C with no explicit use of SIMD features.
steveha
The quality of the generated code, however, is significantly worse, at least at this time.
At work, I use Clang to compile my DSP code, and the result is faster than if I use GCC. I haven't rigorously analyzed the situation but the speedup is on the order of 5% to 10% for my code; not dramatic, but nothing to sneer at either, and all I had to do was edit my makefile to specify Clang.
So I find the blanket statement that Clang generates "significantly worse" code rather surprising. My own experience says otherwise.
steveha
Caveman0: I am draw story on cave wall!
Caveman1: No! Memorize oral tradition make brains strong! Picture story make brains weak!
[panel of cave-children staring vacantly at cave paintings, slack-jawed, drooling]
Caveman0: Me go too far! Me am play gods!
In case you don't know the meme, original source: http://dresdencodak.com/2009/09/22/caveman-science-fiction/
steveha