Author Archive
The death of video/game renting and the raise of P2P file sharing
02/23/2011 | 
Author Don’t think of this as the end of the road for Blockbuster, think of it as a new beginning. After filing forBankruptcy back in September, big blue and yellow has been working to get its affairs in order — and trying desperately to fend off both Netflix and Redbox with offerings like Blockbuster on Demand andBlockbuster Express. Things apparently haven’t gone entirely to plan. The next stop is For Salesville, with bids for the company starting at $290 million. CEO Jim Keys put a positive spin on the news, saying that this will “allow for the consolidation of ownership of the company to those with a clear and focused vision for Blockbuster’s future.” It might also finally let that guy take a vacation — a potentially very long one.
Item-level RFIDs
Toilets, cows, and Germans have all been tagged by RFIDs, but according to a new study, it’s footwear and fashion that top the demand for radio-enabled tracking. In a report released yesterday, ABI Research said more than three-quarters of a billion RFID tags will be used in global apparel markets in 2011, with retailers like Walmart, Macy’s, and JC Penney leading the way. Item-level tracking isn’t new — in fact we saw something similar in 2006 — but with the likes of Walmart on board, the system is expected to grow as much as 60 percent in the next three years. The study suggests inventory and security as driving factors in the adoption of RFIDs, but we’ve got our suspicions. And anyway, we don’t want anyone to know how much we spent on that Material Girl leopard print shrug — not even a machine.
This idea worth us to look into and try to propose this implementation in China Telecom, thus we need somebody mastering Signal and Wave area in ECE.
Inside the OnLive Console
When Marvell was still teasing its Armada processor company co-founder Sehat Sutardja said it would be showing up in “a new gaming platform” — but he kinda left us hanging after that. Four months later it’s looking like we might have found it. Reader David Fisher was kind enough to tear his OnLive Microconsole down to its requisite bits, spreading them upon his kitchen countertop and exposing the Marvell Armada lurking within. Other specs include 512MB of RAM, an unknown quantity of Samsung ROM, and networking chips also from Marvell. There you have it: another mystery of the world solved thanks to your friend the screwdriver.
This is a game streaming hardware similar to what I’ve talked about before – OnLive microconsole has HDMI, Speakers and Ethernet ports, just the necessary for performance and to narrow down power consumption. Now it is still hard to say whether the architecture works. And I wonder if it is true cloud or just streaming from the OnLive servers? As I searched on the Onlive Games, it seems most of the games have low video quality. Some of them are relatively high, but I’m unsure whether they degrade the FPS, or allow huge graphical delays during the game (e_x_p: 720p, the delay is barely noticeable on fast networks). Still, the network speed is a huge gap.
When I have a chance, I will buy one and unscrew that myself.
OnLive is a good concept similar to the “Single Purpose Hardware” we are interested in. We will develop more in the future as there is no need to use a high power consumption architecture for 5% of the utility consuming 300W of power. I believe this idea will be more concerned in the next five years. Somehow most people only need a portable device with one (maybe a couple) purpose at a time.
移动设备实现运行高配置游戏
如果你是一位拥有一台Android智能手机PC游戏迷，那么你一定想在手机上玩魔兽世界（World of Warcraft）。但是，手机设备的硬件相比PC机来说CPU运算能力太低了，特别是缺少3D硬件。
不过最近，云计算的思想给人们带来了手机上的WOW，虽然是Alpha版。
GameString是一个云游戏平台，简单来说，它能让某个程序（游戏程序）在某一台计算机（宿主）上执行，并通过网络接受其它设备（如PC，手机设备等）传输来的操作指令，从而控制程序。也就是说，宿主机像一台服务器一样提供服务，而手机作为输入面板。多么巧妙的设计！
从技术上讲，这种思想能普遍看到，比如Windows远程桌面，Linux远程桌面，SSH，等等，只要你看到或者听到“远程控制”这个词，它就和GameString有相同的思想。
回到Android手机上的WOW话题上来，除了GameString平台的软件，你还需要如下硬件才能在手机上玩WOW：
- HTC Desire智能手机
- 速度超过2Mbps的网络速度！
我猜测GameString的原理是，在手机上播放宿主机上的WOW视频，并在视频画面上接受用户手指点击输入传回宿主机。可想而知，且不说移动通信的网速能否达到需求，其所需要的网络流量和带宽将是惊人！估计一个游戏下来, 你就倾家荡产了！
这个想法我在之前就详细的考虑过，实现并不是不可能，然而现在的网络延时及不稳定性依然是一个巨大的发展瓶颈。云计算是必然要运用起来的，我构想中的移动设备是不需要内嵌高性能的：
1. CPU/GPU
2. 高敏感电容触屏
3. 主板：主要是一些bus（其实10年后就貌似不用bus了）和组合逻辑，还有闪存（存放简易BIOS，操作系统内核，和文件系统）
4. 无线/有线网卡
足以。事实上可以将网卡这部分结构和缓存级别内存整合起来配成一块新的硬件，在大量的数据传输的同时，需要足够大的缓冲来支持。然而考虑到功耗问题，得设计一个合适的大小。
所有运算通过网络直接把用户输入传至多个云服务器，多层平行运算出结果以后直接送到本地。
做到以上这些不是不可能，就目前市场看来，难点是，网络速度和稳定性。就拿现在3G的速度来说，在xPad上玩个WOW是必须不够给力的。网络传输延时和传输出错也有相当大的阻碍，解码之后还要证。我相信信号和无线科技方面的同仁们必然还在这方面努力研究，研发更优秀的传输算法。我对于信号放面不是非常了解，但是我以为，如果要实现上述概念，恐怕需要网络实际传输速度（不是理论速度）最佳情况是要达到和如今bus的速度一样（我们这里不谈主板bus，主板上的bus用MHz来衡量；这里直观的来说我们现在用的USB2.0的是实际值大约是24MB/s左右，事实上USB的速度很大程度上被主板bus给限制了，所以通过USB我们也能看出我们的电脑配置如何)。如果实际网络速度能够靠近这个值，运行高配置游戏就有可能基本不卡。实际这里大部分的数据都是图像的传送，主要为了显示用。比如我们希望游戏达到30FPS的画质（大约看在线视频那清晰度），在一个960X640（IPhone4的分辨率）上用24色呈现，那大约每秒需要30*（960*640*3）byte = 39MB左右的数据。当然，这个绝对属于高清的了，相当奢侈的了。如果普通情况下，降低一下FPS和分辨率，将这个值控制在20以下，是目前看来比较理想的一个尝试。
文中提及“从技术上讲，这种思想能普遍看到，比如Windows远程桌面，Linux远程桌面，SSH，等等，只要你看到或者听到“远程控制”这个词，它就和GameString有相同的思想。” 其实并不完全正确，至少在我们的观点来看，我们所构想的并不是单纯的一对一传输，而是通过云来指导传输源，去向，和分支策略。
这里的“云”需要涉及“共享部分硬件”的概念。在ChoppA Phase #1中，我们在主力研发新的P2P file-sharing strategy，之后将开发P2P hardware-sharing protocol。 我们主力研发与云，低耗能高效率系统相关的科技。
我们并不因新科技而诞生，我们创造之。
Posted in ChoppA, Cloud Computing, Discussion, Games, Google, Hardware, Ideas, Innovation, Inspiration, Mobile, Research, RoBoard
Money and File-sharing Motivation
Piracy is so difficult to battle because file-sharers are motivated by altruism and not financial gain, according to one academic.
Joe Cox, an economist at the Portsmouth Business School, believes file-sharers who post content online see themselves as the “Robin Hoods of the digital age,” according to a study he’s published in the journal Information Economics and Policy.
Such insight could help drive policy and find ways to prevent illegal downloads, he claims. We spoke to him to find out more.
In my opinion, money is not the only motivation for seeders is true. But more importantly, what if we provide a thing called FUN to seeders? Participating in the market economics game, playing around with ChoppA market strategies, getting one’s own reputation and rewards (not just virtual money) in the community, sharing more and getting more, and etc. File-sharers really need to be awarded as they are the heroes in the P2P world. They are not quite the Robin Hoods – no villains to fight off, just a precious sharing mind. Thus we should show our respects for them.
There needs to be someone who understands them, cheers them up for never giving up on file-sharing, and awards them differently. There has to be a motivation, other than money, to push them to continue the selfish-less sharing. ChoppA is the exact idea that brings this motivation!
I am not worried that some people want to deter file-sharing. To those trying to eliminate file-sharing, you think you are the light, and file-sharing is the shadow – light always thinks it travels faster than anything, but it is wrong. No matter where and how fast light runs, the darkness has always reached the destination first and is waiting for it.
Let’s think about how big our file-sharing potential market is. The answer is simply anyone who is involved in Internet download and upload activities. Imagine that…
OnLive’s $9.99 Buffet Plan
OnLive wants to be the Netflix of video games, that much is clear, and today it’s rolling out a flat-rate monthly pricing plan called PlayPack to help seal the deal. It’ll cost $9.99 a month when it launches January 23rd, giving subscribers access to a back catalog of forty retired and indie titles, including a number of games entirely new to the OnLive service. What’s more, if you bought the company’s $99 MicroConsole, you’ll get access to that entire flat-rate catalog free until the formal launch, meaning you’ll have fourteen full games instantly ready to play the moment you boot it up. OnLive founder Steve Perlman tells us you can pay month-to-month and cancel PlayPack anytime you want, and OnLive will still store your savegames for a full year in case you decide to rejoin — or if you want to mix and match flat-rate and a la carte titles without losing your precious progress.
As if no one suggested to them on Day One…
I believe this is a mistake on their long-term view. This is understandable because the founders are mostly experienced entrepreneurs, who themselves probably do not have enough knowledge of what really turn gamers on. The technology itself is supremely innovative, but the business strategies are horrible. So far, at least for steam-lovers like me, have not shown any intention to switch to OnLive.
The new packaged plan is not exactly appealing. Only third-rate games are available – well, just like a buffet. I am not suggesting anything, but a buffeted gaming plan would not work unless the game publishers cooperate, and accept the fact that certain gamers, in certain countries, are not rich enough yet still passionate. Just like converting from piracy to Steam, until there is a compromising solution (cheap but non-refundable, digitalized library in the case of Steam), we (the hardworking gamers) would not subscribe.
Gaikai on the other hand, is taking a different approach which I believe would be more successful than OnLive. Instead of pushing out cloud computing all of a sudden (there isn’t enough bandwidth yet, let alone everyone can afford), the concept can be used to advertise existing games, and later segwayed into the next generation.
OnLive is of course aware that users are the most part of any tech startup. Thus AlphaCoding would not put greedy revenues first, but rather a stable, trusting community of happy consumers.
Fedora Booting on Intel Atom in 4 Seconds
Lineo announced its Warp 2 fast-boot technology in 2008, touting it for booting Linux in 2.97 seconds on a low-powered ARM-based system. These claims have since moved to 1.8 seconds.
Booting up an Intel architecture processor such as the Intel Atom takes longer, so the company’s new 4.06 second claim appears to be fairly significant. Lineo claims that normal boot-time on the test system is 54.72 seconds.
Warp for Atom can boot from various storage media, offers hibernation startup, and supports “multiple snapshots” handling, says Lineo. Customization to specific products can reduce the 4.06 second boot time even more, says the company.
The above is the architectural diagram – a very smart idea: saving/loading DRAM into a flash memory during reboot/wakeup. By injecting from flash memory to DRAM immediately, the kernel, the boot loader, and the root file system are all embedded already. The boot time is shorter because Linux leaves a smaller footprint than Windows (maybe also Mac), thus it is possible to save that amount of memory in this architecture. Even if Windows IS smaller, the complexity of loading drivers and libraries would still slow the loading down by a big factor.
However, Microsoft is actively researching on this embedded-Windows topic. As far as we have known, Tencent is also dreaming about a “QQ embedded” logo everywhere. Fast booting is an essential part of a normal work day. Picture this – while you are staring at an ever-lasting Windows logo, you could be losing your ambitions, your innovative ideas, and your flashy thoughts already…
The future of computers lies in cloud computing … which would hopefully defeat the need to worry about any of this.
Nonetheless, Linux rocks!
Posted in Cloud Computing, Hardware, Inspiration, Linux, Microsoft, Mobile, Open Source, Tencent (腾讯)
Fonera SIMPL Routers
FON (FON Wireless Ltd.) is a company that runs a system of shared wireless networks. The business was launched in November 2005.
People can become members (called “Foneros”) by agreeing to let FON share their wireless Internet connection. Members need to purchase a low-cost wireless router (called a “Fonera”), which acts as a public Internet access point (called a “FON Spot”); the device also creates a private network that can be used by the owner only. FON members can use any other FON access points free of charge. The Fonera 2.0n WiFi router has additional firmware and a USB port that allow it to also function like a networking hub.
Customers who do not share their Internet connection (called “Aliens”) can buy Internet access passes or prepaid WiFi from FON for a fee. FON members whose access points are used by a paying customer receive part of the revenue.
FON seems to be helpful to ChoppA at some point – more internet accesses and more people would join the community. Of course, this also applies to other online communities.
But, while mobile devices are getting “tabletified” (less power consumption + more capability) and 4G network is about to launch, what’s the point of having “routers” to cover the entire world? In other words, people are seeking smaller devices to replace their laptops as a better portable carrying all the time. Most mobile devices have 3G/4G network capabilities nowadays, and usually consume less power than WiFi. Also, this might just be a more expensive WiFi-Direct ripoff.
Patent Supports PSP2 Rear Touch Pad Rumor
Sony’s US gaming division, SCEA, has filed for seven USPTO patents (one, two, three, four,five, six, seven) that hint at the company’s plans for a PSP2. While some only indirectly support past leaks with reference to a touchscreen UI with multi-touch (“two-fingered touch” in the first patent) and a predictive layout (two, four and six), virtually all make reference to having a touch surface distinct from the screen that could be used to steer content without interruption. Many share the same diagrams and show both the “track back touch pad” and, in some drawings, a slider design reminiscent of the PSP2 prototype.
There are many mock-up/conceptual PSP2 product photos online. Well, it seems the story might be different. We are looking forward to how “back touchpad” works. There is also a rumor saying that the new PSP2 has only touch input, well, it might be a tablet? Or just a handheld platform? Somehow SONY does realize that they should not leave this huge cellphone and touchpad-tablet market behind.
Google TV Review
It’s always hard to review version 1.0 of a platform — it’s tempting to give concept and potential nearly as much weight as execution. And we’ll make no bones about it: viewed in that light, Google TV is a success. Features like the search bar and integrated browser are so fundamentally good on a conceptual level that they seem destined to forever change our perception of TV user experience, regardless of how well Google TV itself fares in the market, and that’s no small accomplishment. Other features, like apps and smartphone control, seem equally ambitious and worthy of praise because of promise alone.
I would love to try it when I have the chance. All software companies have the trend that they are trying to associate some part of their business with the hardware industry, by limiting hardware to only install their own products. No matter how good a 3rd party software is, if the architecture does not support it, all becomes meaningless. One of the reasons Google is the leading technology company is its will to support other exciting applications on its platform, as seen in a screenshot above.