Search Results

Search found 134 results on 6 pages for 'prediction'.

Page 4/6 | < Previous Page | 1 2 3 4 5 6  | Next Page >

• A Look Back at 2010 Predictions

  - by David Dorf

  Now is the time of year people make their predictions for next year, but before I start thinking about 2011 it's worth a look back to see how my predictions for 2010 fared. 1. Borders and Blockbuster bite the dust. I would have never predicted a strong brand such as Circuit City could die, but now I know it can happen to anyone. Borders has lost the battle with Barnes & Noble and Blockbuster has lost to Netflix. And just to be sure, Amazon put an extra nail in each coffin. Borders received additional investment from Bennett LeBow to keep it afloat, but the stock is down around $1.25 with no profits in sight. Blockbuster filed for bankruptcy back in September. 2. Every retailer finally has a page on Facebook... but very few figure out how to keep fans engaged. Retailer postings become noise, and fans start to unsubscribe. Twitter goes in the same direction. A few standout retailers will figure out how to use social media, and the rest will remain dumbfounded. Most retailers are on the Facebook bandwagon, and their fan bases seem to be increasing thanks to promotions like The Gap's logo redesign, Lowes' black Friday sneak peak, and Walmart's Crowd Savers. There are several examples of f-commerce advancements, including some interesting integrations from Amazon.3. Smartphones consolidate and grow. More and more people will step-up to smartphones, most of which will choose iPhone, Blackberry, and Android phones. Other smartphones will vanish, and networks will start to strain. But retailers will finally embrace mobile as the next big channel. Retail marketing departments will build mobile apps without the help of their IT department, and eventually they will get into a bind. Android has been on a tear lately stealing market share from Blackberry. Palm and Microsoft are trending down, and Apple is holding steady. Smartphone sales are up 15% and expected to continue. Retailers understand the importance of mobile, and some innovative applications have been produced this year. 4. Google helps the little guys. Google will push its Favorite Places project to help give exposure to small retailers and restaurants. They will enable small retailers to act like big ones by providing storefronts, detailed product information, and coupons for consumers. Google will find a way to bring augmented reality to the masses. I can't say I've seen much new from Google regarding Favorite Places, but they've continued to push local product search. From the PC or smartphone, consumers can search for products and see which nearby stores have it stock. Oracle Retail even productized an integration to Google to support this effort. I suppose if Google ever buys Groupon then it will bring them even closer to local shopping. Google talked about augmented humanity, but that has nothing to do with augmented reality. 5. Steve Jobs Is Bugs Bunny and Steve Ballmer is Elmer Fudd. (OK, I stole that headline from an InformationWeek article. I couldn't resist.) Both Apple and Microsoft will continue to open new stores, but only Apple will show real growth. POSReady 2009 (formerly WEPOS) will continue to share the POS market with Linux. The iPhone and iPod will continue to capture market share, but there won't be an Apple tablet. There won't be an Apple tablet? What was I thinking? While Apple has well over 300 stores, there are less than 10 Microsoft stores. Initial impressions show that even though Microsoft is locating its store near Apple Stores, they are not converting customers, with shoppers citing a lack of assortment and high prices. 6. Consolidation of e-commerce software providers. Software vendors in the areas of search, reviews, online call-centers, payments, and e-commerce will consolidate, partly driven by the success of m-commerce and SaaS. Amazon will find someone else to buy, and eBay will continue to lose momentum. Consolidation of e-commerce providers continued with IBM acquiring Sterling Commerce and CoreMetrics, and Oracle recently announcing the acquisition of ATG. Amazon grabbed Zappos, Woot, and Diapers.com to continue its dominance of online selling. While eBay's Marketplace growth may have slowed, its PayPal division is doing quite well, fueled in part by demand for mobile payments. 7. Book publishers mirror music labels. Just as the iPod brought digital downloads to the masses, the Kindle and Nook will power the e-book revolution. Books will continue to use DRM for a few more years before following the path of music. Publishers will try to preserve the margins of hardbacks by associating e-book releases with paperbacks. Amazon has done a good job providing e-reader clients for smartphones, PCs, and tablets. Competition from Barnes & Noble has forced Amazon to support book loaning, and both companies are making it easier for people to publish ebooks (with or without DRM). Progress is slow but steady. 8. NFC makes inroads, RFID treads water. Near Field Communications start to appear in mobile phones, and retailers beta test its use for payments and loyalty programs. RFID tag costs come down a bit, but not enough to spur accelerated adoption.Nokia announced plans to offer NFC-enabled phones in 2011, and rumors are swirling about NFC in the upcoming iPhone.  I think NFC is heading in the right direction, and I've heard more interest from retailers about specialized uses for RFID.9. Digital Signage goes the way of augmented reality. People use their camera phones to leave geo-tagged notes all over cities, rating stores and restaurants, and "painting" graffiti. But people get tired of holding their phones in front of their faces, so AR glasses are offered in much the same way bluetooth headsets emerged. Retailers experiement with in-store advertising using AR. Several retailers like Pizza Hut, Benetton, and Target have experimented with AR but its still somewhat of a gimmick used by marketing.  I think this prediction is a year or two too early. 10. JDA flip-flops again. After announcing their embracing of the .Net architecture, then switching to J2EE after the Manugistics acquisition, JDA will finally decide to standardize on Apple's Objective C. Everything will be ported to the iPhone and be available on the AppStore. After all, there's not much left to try. This was, of course, a joke but the sentiment is still valid.  JDA seems more supply-chain focused than retail focused, which is a an outcrop if their i2 acquisition.  Of the 10 predictions, I'm going to say I got 6 somewhat correct.  (Don't you just love grading your own paper?)  Soon I'll post my predictions for 2011 so be on the lookout.  Until then here's one more prediction:  Va Tech beats Stanford in the Orange Bowl -- count on it!

  Read the article

• Search predictions broken in Chrome - reinstall didn't solve the problem

  - by Shimmy

  I recently changed the default search engine to a custom google search URL (using baseUrl) with some additional parameters and removed all the rest of the search engines, and since then, the search predictions stopped working. I even tried to reinstall Chrome but as soon as I resync, the problem is back! Search predictions are just gone without option to fix!! In IE changing the search provider allows specifying a prediction (suggestion) provider, In chrome, once you change the default search engine, you'll never be able to have predictions again!! This is a terrible bug, I mean WTF!!! Is there any workaround to that? I posted a bug report a while ago but it seems no one looks at it. I'm about to give up on Chrome and go back to IE, the only good thing about Chrome is the Extension market and the AdBlocker (which I can find in IE as well). The perfrormance changes don't matter to me too much. Thanks

  Read the article

• Google maps later traffic

  - by bobobobo

  Google maps has this feature of showing you how long it will take to get somewhere based on current traffic. But what about history, or asking for a prediction of how long it will take to get somewhere, during rush hour, for example? So, how long will it take to get there on this route if I leave at 5, vs how long it will take if I leave at 3. (A difference of 30 minutes on the road sometimes!) If google maps doesn't provide this service, iis there anywhere that does?

  Read the article

• book about psychology of decision and psychology of human

  - by boos

  I'm a unix developer and i want to make career in project/people management as first step. I think sometimes is better to have good communication skill and in general more human skill to make career more fast. Almost in Italy, a lot of people made career development more fast for his human skill and not for his technical skill. Anyone have read some book about psychology to better manage how people and personality work and to exploit decision making situation in the right way? I have found some interesting book about people personality and psychology of decision, but i am in doubt about the usefulness about reading such book. anyone have some experience in this path ? Anyone have found useful to read similar book about how people work, to manage career development in a more fast way and handle people and decision in a more useful way? i have already read peopleware. The table of content of one of this book have: 1 - Judicment and decision 2 - Euristics and sistematics error 3 - Estimating probability and frequency prediction 4 - Risk and decision 5 - rappresentation and decision 6 - Memory, attention and decision. Etc. what do you think about ?

  Read the article

• Creating Java Neural Networks

  - by Tori Wieldt

  A new article on OTN/Java, titled “Neural Networks on the NetBeans Platform,” by Zoran Sevarac, reports on Neuroph Studio, an open source Java neural network development environment built on top of the NetBeans Platform. This article shows how to create Java neural networks for classification.From the article:“Neural networks are artificial intelligence (machine learning technology) suitable for ill-defined problems, such as recognition, prediction, classification, and control. This article shows how to create some Java neural networks for classification. Note that Neuroph Studio also has support for image recognition, text character recognition, and handwritten letter recognition...”“Neuroph Studio is a Java neural network development environment built on top of the NetBeans Platform and Neuroph Framework. It is an IDE-like environment customized for neural network development. Neuroph Studio is a GUI that sits on top of Neuroph Framework. Neuroph Framework is a full-featured Java framework that provides classes for building neural networks…”The author, Zoran Sevarac, is a teaching assistant at Belgrade University, Department for Software Engineering, and a researcher at the Laboratory for Artificial Intelligence at Belgrade University. He is also a member of GOAI Research Network. Through his research, he has been working on the development of a Java neural network framework, which was released as the open source project Neuroph.Brainy stuff. Read the article here.

  Read the article

• The future for Microsoft

  - by Scott Dorman

  Originally posted on: http://geekswithblogs.net/sdorman/archive/2013/10/16/the-future-for-microsoft.aspxMicrosoft is in the process of reinventing itself. While some may argue that it’s “too little, too late” or that their growing consumer-focused strategy is wrong, the truth of the situation is that Microsoft is reinventing itself into a new company. While Microsoft is now calling themselves a “devices and services” company, that’s not entirely accurate. Let’s look at some facts: Microsoft will always (for the long-term foreseeable future) be financially split into the following divisions: Windows/Operating Systems, which for FY13 made up approximately 24% of overall revenue. Server and Tools, which for FY13 made up approximately 26% of overall revenue. Enterprise/Business Products, which for FY13 made up approximately 32% of overall revenue. Entertainment and Devices, which for FY13 made up approximately 13% of overall revenue. Online Services, which for FY13 made up approximately 4% of overall revenue. It is important to realize that hardware products like the Surface fall under the Windows/Operating Systems division while products like the Xbox 360 fall under the Entertainment and Devices division. (Presumably other hardware, such as mice, keyboards, and cameras, also fall under the Entertainment and Devices division.) It’s also unclear where Microsoft’s recent acquisition of Nokia’s handset division will fall, but let’s assume that it will be under Entertainment and Devices as well. Now, for the sake of argument, let’s assume a slightly different structure that I think is more in line with how Microsoft presents itself and how the general public sees it: Consumer Products and Devices, which would probably make up approximately 9% of overall revenue. Developer Tools, which would probably make up approximately 13% of overall revenue. Enterprise Products and Devices, which would probably make up approximately 47% of overall revenue. Entertainment, which would probably make up approximately 13% of overall revenue. Online Services, which would probably make up approximately 17% of overall revenue. (Just so we’re clear, in this structure hardware products like the Surface, a portion of Windows sales, and other hardware fall under the Consumer Products and Devices division. I’m assuming that more of the income for the Windows division is coming from enterprise/volume licenses so 15% of that income went to the Enterprise Products and Devices division. Most of the enterprise services, like Azure, fall under the Online Services division so half of the Server and Tools income went there as well.) No matter how you look at it, the bulk of Microsoft’s income still comes from not just the enterprise but also software sales, and this really shouldn’t surprise anyone. So, now that the stage is set…what’s the future for Microsoft? The future I see for Microsoft (again, this is just my prediction based on my own instinct, gut-feel and publicly available information) is this: Microsoft is becoming a consumer-focused enterprise company. Let’s look at it a different way. Microsoft is an enterprise-focused company trying to create a larger consumer presence.  To a large extent, this is the exact opposite of Apple, who is really a consumer-focused company trying to create a larger enterprise presence. The major reason consumer-focused companies (like Apple) have started making in-roads into the enterprise is the “bring your own device” phenomenon. Yes, Apple has created some “game-changing” products but their enterprise influence is still relatively small. Unfortunately (for this blog post at least), Apple provides revenue in terms of hardware products rather than business divisions, so it’s not possible to do a direct comparison. However, in the interest of transparency, from Apple’s Quarterly Report (filed 24 July 2013), their revenue breakdown is: iPhone, which for the 3 months ending 29 June 2013 made up approximately 51% of revenue. iPad, which for the 3 months ending 29 June 2013 made up approximately 18% of revenue. Mac, which for the 3 months ending 29 June 2013 made up approximately 14% of revenue. iPod, which for the 3 months ending 29 June 2013 made up approximately 2% of revenue. iTunes, Software, and Services, which for the 3 months ending 29 June 2013 made up approximately 11% of revenue. Accessories, which for the 3 months ending 29 July 2013 made up approximately 3% of revenue. From this, it’s pretty clear that Apple is a consumer-and-hardware-focused company. At this point, you may be asking yourself “Where is all of this going?” The answer to that lies in Microsoft’s shift in company focus. They are becoming more consumer focused, but what exactly does that mean? The biggest change (at least that’s been in the news lately) is the pending purchase of Nokia’s handset division. This, in combination with their Surface line of tablets and the Xbox, will put Microsoft squarely in the realm of a hardware-focused company in addition to being a software-focused company. That can (and most likely will) shift the revenue split to looking at revenue based on software sales (both consumer and enterprise) and also hardware sales (mostly on the consumer side). If we look at things strictly from a Windows perspective, Microsoft clearly has a lot of irons in the fire at the moment. Discounting the various product SKUs available and painting the picture with broader strokes, there are currently 5 different Windows-based operating systems: Windows Phone Windows Phone 7.x, which runs on top of the Windows CE kernel Windows Phone 8.x+, which runs on top of the Windows 8 kernel Windows RT The ARM-based version of Windows 8, which runs on top of the Windows 8 kernel Windows (Pro) The Intel-based version of Windows 8, which runs on top of the Windows 8 kernel Xbox The Xbox 360, which runs it’s own proprietary OS. The Xbox One, which runs it’s own proprietary OS, a version of Windows running on top of the Windows 8 kernel and a proprietary “manager” OS which manages the other two. Over time, Windows Phone 7.x devices will fade so that really leaves 4 different versions. Looking at Windows RT and Windows Phone 8.x paints an interesting story. Right now, all mobile phone devices run on some sort of ARM chip and that doesn’t look like it will change any time soon. That means Microsoft has two different Windows based operating systems for the ARM platform. Long term, it doesn’t make sense for Microsoft to continue supporting that arrangement. I have long suspected (since the Surface was first announced) that Microsoft will unify these two variants of Windows and recent speculation from some of the leading Microsoft watchers lends credence to this suspicion. It is rumored that upcoming Windows Phone releases will include support for larger screen sizes, relax the requirement to have a hardware-based back button and will continue to improve API parity between Windows Phone and Windows RT. At the same time, Windows RT will include support for smaller screen sizes. Since both of these operating systems are based on the same core Windows kernel, it makes sense (both from a financial and development resource perspective) for Microsoft to unify them. The user interfaces are already very similar. So similar in fact, that visually it’s difficult to tell them apart. To illustrate this, here are two screen captures: Other than a few variations (the Bing News app, the picture shown in the Pictures tile and the spacing between the tiles) these are identical. The one on the left is from my Windows 8.1 laptop (which looks the same as on my Surface RT) and the one on the right is from my Windows Phone 8 Lumia 925. This pretty clearly shows that from a consumer perspective, there really is no practical difference between how these two operating systems look and how you interact with them. For the consumer, your entertainment device (Xbox One), phone (Windows Phone) and mobile computing device (Surface [or some other vendors tablet], laptop, netbook or ultrabook) and your desktop computing device (desktop) will all look and feel the same. While many people will denounce this consistency of user experience, I think this will be a good thing in the long term, especially for the upcoming generations. For example, my 5-year old son knows how to use my tablet, phone and Xbox because they all feature nearly identical user experiences. When Windows 8 was released, Microsoft allowed a Windows Store app to be purchased once and installed on as many as 5 devices. With Windows 8.1, this limit has been increased to over 50. Why is that important? If you consider that your phone, computing devices, and entertainment device will be running the same operating system (with minor differences related to physical hardware chipset), that means that I could potentially purchase my sons favorite Angry Birds game once and be able to install it on all of the devices I own. (And for those of you wondering, it’s only 7 [at the moment].) From an app developer perspective, the story becomes even more compelling. Right now there are differences between the different operating systems, but those differences are shrinking. The user interface technology for both is XAML but there are different controls available and different user experience concepts. Some of the APIs available are the same while some are not. You can’t develop a Windows Phone app that can also run on Windows (either Windows Pro or RT). With each release of Windows Phone and Windows RT, those difference become smaller and smaller. Add to this mix the Xbox One, which will also feature a Windows-based operating system and the same “modern” (tile-based) user interface and the visible distinctions between the operating systems will become even smaller. Unifying the operating systems means one set of APIs and one code base to maintain for an app that can run on multiple devices. One code base means it’s easier to add features and fix bugs and that those changes become available on all devices at the same time. It also means a single app store, which will increase the discoverability and reach of your app and consolidate revenue and app profile management. Now, the choice of what devices an app is available on becomes a simple checkbox decision rather than a technical limitation. Ultimately, this means more apps available to consumers, which is always good for the app ecosystem. Is all of this just rumor, speculation and conjecture? Of course, but it’s not unfounded. As I mentioned earlier, some of the prominent Microsoft watchers are also reporting similar rumors. However, Microsoft itself has even hinted at this future with their recent organizational changes and by telling developers “if you want to develop for Xbox One, start developing for Windows 8 now.” I think this pretty clearly paints the following picture: Microsoft is committed to the “modern” user interface paradigm. Microsoft is changing their release cadence (for all products, not just operating systems) to be faster and more modular. Microsoft is going to continue to unify their OS platforms both from a consumer perspective and a developer perspective. While this direction will certainly concern some people it will excite many others. Microsoft’s biggest failing has always been following through with a strong and sustained marketing strategy that presents a consistent view point and highlights what this unified and connected experience looks like and how it benefits consumers and enterprises. We’ve started to see some of this over the last few years, but it needs to continue and become more aggressive and consistent. In the long run, I think Microsoft will be able to pull all of these technologies and devices together into one seamless ecosystem. It isn’t going to happen overnight, but my prediction is that we will be there by the end of 2016. As both a consumer and a developer, I, for one, am excited about the future of Microsoft.

  Read the article

• MMO Data Persistence Question

  - by JasonG

  I wanted to ask a question regarding data persistence strategies for an MMO. I have some experience in the games industry with social synchronous games. At Zynga, we stored static proto data in XML on both the client and the server and stored instance/runtime data in membase. For clarity sake, proto data for a Potion would be PotionName or MaxCharges, while runtime/instance data would be something like ChargesRemaining. So basically, if a player picks up a potion the instance is (via prediction) created from XML data on the client, the request gets sent to the server where the instance is created from XML and then added to membase. Is the same strategy that would be used for soemthing like an MMO? Would it be feasible to have static proto data in some kind of in-memory no-sql database on both client and server with instance data being stored on the server in a more enterprise level database? Or should all data (proto/instance) be stored on the server and the client gets everything from server? I know a lot of this might on certain game requirements, however, i'm basically looking for some general opinion/best practices here if there are any.

  Read the article

• SSIS Dashboard v0.4

  - by Davide Mauri

  Following the post on SSISDB script on Gist, I’ve been working on a HTML5 SSIS Dashboard, in order to have a nice looking, user friendly and, most of all, useful, SSIS Dashboard. Since this is a “spare-time” project, I’ve decided to develop it using Python since it’s THE data language (R aside), it’s a beautiful & powerful, well established and well documented and with a rich ecosystem around. Plus it has full support in Visual Studio, through the amazing Python Tools For Visual Studio plugin, I decided also to use Flask, a very good micro-framework to create websites, and use the SB Admin 2.0 Bootstrap admin template, since I’m all but a Web Designer. The result is here: https://github.com/yorek/ssis-dashboard and I can say I’m pretty satisfied with the work done so far (I’ve worked on it for probably less than 24 hours). Though there’s some features I’d like to add in t future (historical execution time, some charts, connection with AzureML to do prediction on expected execution times) it’s already usable. Of course I’ve tested it only on my development machine, so check twice before putting it in production but, give the fact that, virtually, there is not installation needed (you only need to install Python), and that all queries are based on standard SSISDB objects, I expect no big problems. If you want to test, contribute and/or give feedback please fell free to do it…I would really love to see this little project become a community project! Enjoy!

  Read the article

• Can Near Field Communications (NFC) Benefit your Supply Chain?

  - by Stephen Slade

  Leading firms continue to leverage the latest tools and technologies to drive performance especially around minimizing transaction costs. With razor thin margins in manufacturing and distribution, the leading producers are resorting to Near Field Communications to gain efficiencies.  In this week’s CIO magazine (Apr1, 2012, pg.30, see http://www.cio.com)  Lauren Brousell talks of the things you need to know to make a more informed decision with NFC.  Sandy Shen of Gartner says NFC appeals because "it supports any services that requires data transfer and authentication' 1. NFC is Cheap and Easy - short range transmitting technology connecting smartphones to data transfer. 2. Adoption Seems Inevitable - more merchants will use NCF for payments in the futures. Wallets are becoming obsolete. 3. It's a Hot Potato for Enterprise - Business with credit card companies and cell phone providers are debating who handles the billing process. 4. It's in use Overseas. Japan uses FeliCa to pay by smartphone. In the US, billing agreements are causing territorial conflict. 5. Security Risks Come Standard. As people lose HH devices, security will be an ongoing concern. Credentials and timeout features and alleviate to some extent. My prediction: In 5 years, we won't have wallets in our pockets.  Our secure and all-powerful smart phones will be our electronic portable banks and execute the transaction for us based on our preferences and propensities and electronically execute the transaction for the supply chain.

  Read the article

• Sending changes to a terrain heightmap over UDP

  - by Floomi

  This is a more conceptual, thinking-out-loud question than a technical one. I have a 3D heightmapped terrain as part of a multiplayer RTS that I would like to terraform over a network. The terraforming will be done by units within the gameworld; the player will paint a "target heightmap" that they'd like the current terrain to resemble and units will deform towards that on their own (a la Perimeter). Given my terrain is 257x257 vertices, the naive approach of sending heights when they change will flood the bandwidth very quickly - updating a quarter of the terrain every second will hit ~66kB/s. This is clearly way too much. My next thought was to move to a brush-based system, where you send e.g. the centre of a circle, its radius, and some function defining the influence of the brush from the centre going outwards. But even with reliable UDP the "start" and "stop" messages could still be delayed. I guess I could compare timestamps and compensate for this, although it'd likely mean that clients would deform verts too much on their local simulations and then have to smooth them back to the correct heights. I could also send absolute vert heights in the "start" and "stop" messages to guarantee correct data on the clients. Alternatively I could treat brushes in a similar way to units, and do the standard position + velocity + client-side prediction jazz on them, with the added stipulation that they deform terrain within a certain radius around them. The server could then intermittently do a pass and send (a subset of) recently updated verts to clients as and when there's bandwidth to spare. Any other suggestions, or indications that I'm on the right (or wrong!) track with any of these ideas would be greatly appreciated.

  Read the article

• Weekly Cloud Roundup 2012-15

  - by Alan Smith

  Filtering the informative, insightful and quirky from the fire hose of cloud-based hype. Irving Wladawsky-Berger provides some great insight into The Complex Transition to the Cloud, sharing his views on the slow adoption of cloud computing in organizations. “…a prediction by the research firm Gartner that while cloud computing will continue to grow at almost 20 percent a year, it will account for less than 5 percent of totally IT spending in 2015.” With a more positive mindset, Balaji Viswanathan highlights 7 Salient Trends and Directions in Cloud Computing that could be shaping the industry over the next few years. Cloud computing also looks to save energy “A small business with 100 users that moved the Microsoft applications to the cloud could cut energy use and carbon emissions by 90%. Large organizations with 10,000 users saw a 30% reduction.” More on that story here. The expansion of Windows Azure has been in the news with the announcement of “East US” and “West US” datacenters; this was covered by Visual Studio Magazine and Mary-Jo, and according to thenextweb.com Microsoft are also building $112 million data center in Wyoming. The cloud price war is still in full swing with Joe Panettieri discussing the pricing of Windows Azure and Office 365 and asking How Low Can It Go?

  Read the article

• Physics not synchronizing correctly over the network when using Bullet

  - by Lucas

  I'm trying to implement a client/server physics system using Bullet however I'm having problems getting things to sync up. I've implemented a custom motion state which reads and write the transform from my game objects and it works locally but I've tried two different approaches for networked games: Dynamic objects on the client that are also on the server (eg not random debris and other unimportant stuff) are made kinematic. This works correctly but the objects don't move very smoothly Objects are dynamic on both but after each message from the server that the object has moved I set the linear and angular velocity to the values from the server and call btRigidBody::proceedToTransform with the transform on the server. I also call btCollisionObject::activate(true); to force the object to update. My intent with method 2 was to basically do method 1 but hijacking Bullet to do a poor-man's prediction instead of doing my own to smooth out method 1, but this doesn't seem to work (for reasons that are not 100% clear to me even stepping through Bullet) and the objects sometimes end up in different places. Am I heading in the right direction? Bullet seems to have it's own interpolation code built-in. Can that help me make method 1 work better? Or is my method 2 code not working because I am accidentally stomping that?

  Read the article

• Architecture of an action multiplayer game from scratch

  - by lcf

  Not sure whether it's a good place to ask (do point me to a better one if it's not), but since what we're developing is a game - here it goes. So this is a "real-time" action multiplayer game. I have familiarized myself with concepts like lag compensation, view interpolation, input prediction and pretty much everything that I need for this. I have also prepared a set of prototypes to confirm that I understood everything correctly. My question is about the situation when game engine must be rewind to the past to find out whether there was a "hit" (sometimes it may involve the whole 'recomputation' of the world from that moment in the past up to the present moment. I already have a piece of code that does it, but it's not as neat as I need it to be. The domain logic of the app (the physics of the game) must be separated from the presentation (render) and infrastructure tools (e.g. the remote server interaction specifics). How do I organize all this? :) Is there any worthy implementation with open sources I can take a look at? What I'm thinking is something like this: -> Render / User Input -> Game Engine (this is the so called service layer) -> Processing User Commands & Remote Server -> Domain (Physics) How would you add into this scheme the concept of "ticks" or "interactions" with the possibility to rewind and recalculate "the game"? Remember, I cannot change the Domain/Physics but only the Game Engine. Should I store an array of "World's States"? Should they be just some representations of the world, optimized for this purpose somehow (how?) or should they be actual instances of the world (i.e. including behavior and all that). Has anybody had similar experience? (never worked on a game before if that matters)

  Read the article

• Multiplayer approach for tablets on wi-fi (FPS/TPS)? Server authority, etc

  - by Fraggle

  Looking for some guidance or what has worked well for others in implementing a multiplayer FPS/TPS type game on tablets (probably just 2-6 players at a time). The main issue being that tablets/phones are typically "less" connected than say a console or pc might be. And therefore, my thought is that to have complete Server authority of everything is not going to work. But maybe I'm off base on that. So I guess I'm struggling with what (if anything) should happen on a central server and what should happen locally. Or is centralized approach even needed? Some approaches I might do: Player movement : my thought is to control this locally (player-owner) and update server with positon (which then sends out to other clients). Use client side prediction for opponent players so that connection loss will not show a plane for example stop in mid air. Server will send update and try to smoothly correct an opponent player position to server updated one.But don't update owners position on owners device from server. Powerups (health kit/ammo/coins/etc) : need to see them disappear immediately, so do it locally. Add the health locally, but perhaps allow for server correction. If server doesn't see player near that powerup, reject the powerup and adjust server health for player. Fire weapons: Have to see it happen right away, so fire locally, create local bullet and send on its way. Send rpc to server so that this player on other clients also fires. Hit detection: Get's trickier. Make bullet/projectile disappear locally, and perhaps perform local hit animations (shaking, whatever). non-authoritative approach= take the damage locally and send rpc to server or others to update health and inform of hit. Authoritative approach-Don't take the damage, or adjust health. Server will do that if it detects a hit. Anyway that's my current thought stream. Let me know what you think of the above or what has worked for you.

  Read the article

• How to perform game object smoothing in multiplayer games

  - by spaceOwl

  We're developing an infrastructure to support multiplayer games for our game engine. In simple terms, each client (player) engine sends some pieces of data regarding the relevant game objects at a given time interval. On the receiving end, we step the incoming data to current time (to compensate for latency), followed by a smoothing step (which is the subject of this question). I was wondering how smoothing should be performed ? Currently the algorithm is similar to this: Receive incoming state for an object (position, velocity, acceleration, rotation, custom data like visual properties, etc). Calculate a diff between local object position and the position we have after previous prediction steps. If diff doesn't exceed some threshold value, start a smoothing step: Mark the object's CURRENT POSITION and the TARGET POSITION. Linear interpolate between these values for 0.3 seconds. I wonder if this scheme is any good, or if there is any other common implementation or algorithm that should be used? (For example - should i only smooth out the position? or other values, such as speed, etc) any help will be appreciated.

  Read the article

• About output of vga_switcher

  - by zhangjie

  When IGD and DIS both exist in my pc,and I want to disable DIS,so I create a service to switch on and off the DIS.It works.Finally,I decide to add the service command into /etc/rc.local so that DIS will be powered off automatically.Unfortunately,it fails.There's only one command added by myself in the file /etc/rc.local,so I can affirm failure is caused by that added command. Before,I directly added the command "echo OFF /sys/kernel/debug..." into /etc/rc.local,and when I restarted,the system startup fails.So I thought maybe when the command is executed,the DIS hasn't been powered on or ready for work.So conflict occurs!It's just my prediction.Then I added one line command "sleep 1s" before the "echo OFF ...",it works nearly everytime when I start or restart pc,while fails sometimes. The output result of "cat /sys/kernel/debug..." is as following: 0:IGD:+:Pwr:0000:00:02.0 1:DIS: :Pwr:0000:01:00.0 I want know 0000:00:02.0 means what?Time of first power on? If it was really time,I can set the command "sleep 2s" to wait for DIS powered on then "echo OFF ..." Thanks for your advice!

  Read the article

• Microsoft Declares the Future of ASP.NET is Web API

  - by sbwalker

  Sitting on a plane on my way home from Tech Ed 2012 in Orlando, I thought it would be a good time to jot down some key takeaways from this year’s conference. Some of these items I have known since the Microsoft MVP Summit which occurred in Redmond in late February ( but due to NDA restrictions I could not share them with the developer community at large ) and some of them are a result of insightful conversations with a wide variety of industry insiders and Microsoft employees at the conference. First, let’s travel back in time 4 years to the Microsoft MVP Summit in 2008. Microsoft was facing some heat from market newcomer Ruby on Rails and responded with a new web development framework of its own, ASP.NET MVC. At the Summit they estimated that MVC would only be applicable for ~10% of all new web development projects. Based on that prediction I questioned why they were investing such considerable resources for such a relative edge case, but my guess is that they felt it was an important edge case at the time as some of the more vocal .NET evangelists as well as some very high profile start-ups ( ie. Twitter ) had publicly announced their intent to use Rails. Microsoft made a lot of noise about MVC. In fact, they focused so much of their messaging and marketing hype around MVC that it appeared that WebForms was essentially dead. Yes, it may have been true that Microsoft continued to invest in WebForms, but from an outside perspective it really appeared that MVC was the only framework getting any real attention. As a result, MVC started to gain market share. An inside source at Microsoft told me that MVC usage has grown at a rate of about 5% per year and now sits at ~30%. Essentially by focusing so much marketing effort on MVC, Microsoft actually created a larger market demand for it.  This is because in the Microsoft ecosystem there is somewhat of a bandwagon mentality amongst developers. If Microsoft spends a lot of time talking about a specific technology, developers get the perception that it must be really important. So rather than choosing the right tool for the job, they often choose the tool with the most marketing hype and then try to sell it to the customer. In 2010, I blogged about the fact that MVC did not make any business sense for the DotNetNuke platform. This was because our ecosystem relied on third party extensions which were dependent on the WebForms model. If we migrated the core to MVC it would mean that all of the third party extensions would no longer be compatible, which would be an irresponsible business decision for us to make at the expense of our users and customers. However, this did not stop the debate from continuing to occur in our ecosystem. Clearly some developers had drunk Microsoft’s Kool-Aid about MVC and were of the mindset, to paraphrase an old Scottish saying, “If its not MVC, it’s crap”. Now, this is a rather ignorant position to take as most of the benefits of MVC can be achieved in WebForms with solid architecture and responsible coding practices. Clean separation of concerns, unit testing, and direct control over page output are all possible in the WebForms model – it just requires diligence and discipline. So over the past few years some horror stories have begun to bubble to the surface of software development projects focused on ground-up rewrites of web applications for the sole purpose of migrating from WebForms to MVC. These large scale rewrites were typically initiated by engineering teams with only a single argument driving the business decision, that Microsoft was promoting MVC as “the future”. These ill-fated rewrites offered no benefit to end users or customers and in fact resulted in a less stable, less scalable and more complicated systems – basically taking one step forward and two full steps back. A case in point is the announcement earlier this week that a popular open source .NET CMS provider has decided to pull the plug on their new MVC product which has been under active development for more than 18 months and revert back to WebForms. The availability of multiple server-side development models has deeply fragmented the Microsoft developer community. Some folks like to compare it to the age-old VB vs. C# language debate. However, the VB vs. C# language debate was ultimately more of a religious war because at least the two dominant programming languages were compatible with one another and could be used interchangeably. The issue with WebForms vs. MVC is much more challenging. This is because the messaging from Microsoft has positioned the two solutions as being incompatible with one another and as a result web developers feel like they are forced to choose one path or another. Yes, it is true that it has always been technically possible to use WebForms and MVC in the same project, but the tooling support has always made this feel “dirty”. The fragmentation has also made it difficult to attract newcomers as the perceived barrier to entry for learning ASP.NET has become higher. As a result many new software developers entering the market are gravitating to environments where the development model seems more simple and intuitive ( ie. PHP or Ruby ). At the same time that the Web Platform team was busy promoting ASP.NET MVC, the Microsoft Office team has been promoting Sharepoint as a platform for building internal enterprise web applications. Sharepoint has great penetration in the enterprise and over time has been enhanced with improved extensibility capabilities for software developers. But, like many other mature enterprise ASP.NET web applications, it is built on the WebForms development model. Similar to DotNetNuke, Sharepoint leverages a rich third party ecosystem for both generic web controls and more specialized WebParts – both of which rely on WebForms. So basically this resulted in a situation where the Web Platform group had headed off in one direction and the Office team had gone in another direction, and the end customer was stuck in the middle trying to figure out what to do with their existing investments in Microsoft technology. It really emphasized the perception that the left hand was not speaking to the right hand, as strategically speaking there did not seem to be any high level plan from Microsoft to ensure consistency and continuity across the different product lines. With the introduction of ASP.NET MVC, it also made some of the third party control vendors scratch their heads, and wonder what the heck Microsoft was thinking. The original value proposition of ASP.NET over Classic ASP was the ability for web developers to emulate the highly productive desktop development model by using abstract components for creating rich, interactive web interfaces. Web control vendors like Telerik, Infragistics, DevExpress, and ComponentArt had all built sizable businesses offering powerful user interface components to WebForms developers. And even after MVC was introduced these vendors continued to improve their products, offering greater productivity and a superior user experience via AJAX to what was possible in MVC. And since many developers were comfortable and satisfied with these third party solutions, the demand remained strong and the third party web control market continued to prosper despite the availability of MVC. While all of this was going on in the Microsoft ecosystem, there has also been a fundamental shift in the general software development industry. Driven by the explosion of Internet-enabled devices, the focus has now centered on service-oriented architecture (SOA). Service-oriented architecture is all about defining a public API for your product that any client can consume; whether it’s a native application running on a smart phone or tablet, a web browser taking advantage of HTML5 and Javascript, or a rich desktop application running on a PC. REST-based services which utilize the less verbose characteristics of JSON as a transport mechanism, have become the preferred approach over older, more bloated SOAP-based techniques. SOA also has the benefit of producing a cross-platform API, as every major technology stack is able to interact with standard REST-based web services. And for web applications, more and more developers are turning to robust Javascript libraries like JQuery and Knockout for browser-based client-side development techniques for calling web services and rendering content to end users. In fact, traditional server-side page rendering has largely fallen out of favor, resulting in decreased demand for server-side frameworks like Ruby on Rails, WebForms, and (gasp) MVC. In response to these new industry trends, Microsoft did what it always does – it immediately poured some resources into developing a solution which will ensure they remain relevant and competitive in the web space. This work culminated in a new framework which was branded as Web API. It is convention-based and designed to embrace native HTTP standards without copious layers of abstraction. This framework is designed to be the ultimate replacement for both the REST aspects of WCF and ASP.NET MVC Web Services. And since it was developed out of band with a dependency only on ASP.NET 4.0, it means that it can be used immediately in a variety of production scenarios. So at Tech Ed 2012 it was made abundantly clear in numerous sessions that Microsoft views Web API as the “Future of ASP.NET”. In fact, one Microsoft PM even went as far as to say that if we look 3-4 years into the future, that all ASP.NET web applications will be developed using the Web API approach. This is a fairly bold prediction and clearly telegraphs where Microsoft plans to allocate its resources going forward. Currently Web API is being delivered as part of the MVC4 package, but this is only temporary for the sake of convenience. It also sounds like there are still internal discussions going on in terms of how to brand the various aspects of ASP.NET going forward – perhaps the moniker of “ASP.NET Web Stack” coined a couple years ago by Scott Hanselman and utilized as part of the open source release of ASP.NET bits on Codeplex a few months back will eventually stick. Web API is being positioned as the unification of ASP.NET – the glue that is able to pull this fragmented mess back together again. The  “One ASP.NET” strategy will promote the use of all frameworks - WebForms, MVC, and Web API, even within the same web project. Basically the message is utilize the appropriate aspects of each framework to solve your business problems. Instead of navigating developers to a fork in the road, the plan is to educate them that “hybrid” applications are a great strategy for delivering solutions to customers. In addition, the service-oriented approach coupled with client-side development promoted by Web API can effectively be used in both WebForms and MVC applications. So this means it is also relevant to application platforms like DotNetNuke and Sharepoint, which means that it starts to create a unified development strategy across all ASP.NET product lines once again. And so what about MVC? There have actually been rumors floated that MVC has reached a stage of maturity where, similar to WebForms, it will be treated more as a maintenance product line going forward ( MVC4 may in fact be the last significant iteration of this framework ). This may sound alarming to some folks who have recently adopted MVC but it really shouldn’t, as both WebForms and MVC will continue to play a vital role in delivering solutions to customers. They will just not be the primary area where Microsoft is spending the majority of its R&D resources. That distinction will obviously go to Web API. And when the question comes up of why not enhance MVC to make it work with Web API, you must take a step back and look at this from the higher level to see that it really makes no sense. MVC is a server-side page compositing framework; whereas, Web API promotes client-side page compositing with a heavy focus on web services. In order to make MVC work well with Web API, would require a complete rewrite of MVC and at the end of the day, there would be no upgrade path for existing MVC applications. So it really does not make much business sense. So what does this have to do with DotNetNuke? Well, around 8-12 months ago we recognized the software industry trends towards web services and client-side development. We decided to utilize a “hybrid” model which would provide compatibility for existing modules while at the same time provide a bridge for developers who wanted to utilize more modern web techniques. Customers who like the productivity and familiarity of WebForms can continue to build custom modules using the traditional approach. However, in DotNetNuke 6.2 we also introduced a new Service Framework which is actually built on top of MVC2 ( we chose to leverage MVC because it had the most intuitive, light-weight REST implementation in the .NET stack ). The Services Framework allowed us to build some rich interactive features in DotNetNuke 6.2, including the Messaging and Notification Center and Activity Feed. But based on where we know Microsoft is heading, it makes sense for the next major version of DotNetNuke ( which is expected to be released in Q4 2012 ) to migrate from MVC2 to Web API. This will likely result in some breaking changes in the Services Framework but we feel it is the best approach for ensuring the platform remains highly modern and relevant. The fact that our development strategy is perfectly aligned with the “One ASP.NET” strategy from Microsoft means that our customers and developer community can be confident in their current and future investments in the DotNetNuke platform.

  Read the article

• How to create a backpropagation neural network in neurondonet?

  - by Suraj Prakash

  I am doing stock market prediction using ANNs in c#.net. I am using NeuronDotNet for the neural part. I have to give eight inputs to the network, with a hidden layer consisting 8 nodes and a single node output layer. Can anybody please give me some coding ideas for this???? This project was not a AI course assignment, but my major project. I have studied about the stocks and found various factors that affected the future value of stock of a company. Now I have to use these factors as input to the neural network. I am not getting into how to implement these factors in the neural network. I have just decided to use those eight factors as eight nodes in the input layer but things are going complex. My concern is to use these factors as input and train the neural network for output as next day's stock value. What major things should I have to care about??

  Read the article

• DMX Analysis Services question

  - by user282382

  Hi, I am have two mining models, both are time series. One is [Company_Inputs] and the other is [Booking_Projections]. What I want to do is use EXTEND_MODEL_CASES to join the results of [Company_Inputs] as the extended cases. So basically something like: Select Flattened PredictTimeSeries([Bookings], 1, 6, EXTEND_MODEL_CASES) FROM [Booking_Projections] Natural Prediction Join (Select Flattened PredictTimeSeries([Metric1], 1, 6) From [Company_Inputs]) AS T This code of course doesn't work, but the idea is to use the predictions made from [Company_Inputs] as cases for predicting future values of [Booking_Projections] If anyone has an idea of how I can accomplish this I would appreciate it very much.

  Read the article

• Game network physics collision

  - by Jonas Byström

  How to simulating two client-controlled vehicles colliding (sensibly) in a typical client/server setup for a network game? I did read this eminent blog post on how to do distributed network physics in general (without traditional client prediction), but this question is specifically on how to handle collisions of owned objects. Example Say client A is 20 ms ahead of server, client B 300 ms ahead of server (counting both latency and maximum jitter). This means that when the two vehicles collide, both clients will see the other as 320 ms behind - in the opposite direction of the velocity of the other vehicle. Head-to-head on a Swedish highway means a difference of 16 meters/17.5 yards! What not to try It is virtually impossible to extrapolate the positions, since I also have very complex vehicles with joints and bodies all over, which in turn have linear and angular positions, velocities and accelerations, not to mention states from user input.

  Read the article

• "Inlining" (kind of) functions at runtime in C

  - by fortran

  Hi, I was thinking about a typical problem that is very JIT-able, but hard to approach with raw C. The scenario is setting up a series of function pointers that are going to be "composed" (as in maths function composition) once at runtime and then called lots and lots of times. Doing it the obvious way involves many virtual calls, that are expensive, and if there are enough nested functions to fill the CPU branch prediction table completely, then the performance with drop considerably. In a language like Lisp, I could probably process the code and substitute the "virtual" call by the actual contents of the functions and then call compile to have an optimized version, but that seems very hacky and error prone to do in C, and using C is a requirement for this problem ;-) So, do you know if there's a standard, portable and safe way to achieve this in C? Cheers

  Read the article

• I still think Twitter is dead &hellip; but

  - by Randy Walker

  Twitter finally hit the mainstream about 8 months ago, but I’ve been saying for a couple of years now, without a real way for the company to earn money, what’s the future fate of Twitter?  On the personal side, where is the real value for the users?  For the most part, Twitter has replaced most people’s IM (instant messaging), at least in the technology circles I run in.  It still has value for users as a communication tool.  But I see it more as a fad.  My prediction is over the next 6 months we’ll start seeing a usage drop (if we haven’t already started to see it). On the business side, how does Twitter make money?  It doesn’t.  If you use the text messaging capabilities, you see a few ads.  But most smart phone and PC users, won’t ever see them.  I still think Twitter has the best chance to make money by forcing the “collectors” to pay money.  You know what I mean by “collector”, those people that collect tons of followers or friends.  If Twitter caps the number of followers and makes you pay to have more, would you?  The normal twitter user doesn’t have that many followers, and this is where my title comes in … BUT The financial value for Twitter is really seen through businesses connecting with their customers.  I’ve seen 3 effective ways this has been accomplished. 1. Giving your customers a coupon or announcing a sale My favorite is @amazonmp3, Being a huge music lover, I get notified when they put music on sale. Various restaurants like @ruthschris_ARK will let their favorite customers know about certain specials @BluefinMemphis I was traveling through Memphis once looking for a sushi restaurant when they had %50 off if we mentioned we saw them on Twitter.  It was their first attempt at trying to encourage customers in the door, and after talking with the management, it was a huge success 2. Giveaways @namecheap Several companies have started huge marketing campaigns, but my favorite is watching companies post trivia questions, and the first person to respond wins a prize. 3. Responding to Customer Complaints I once posted a complaint about American Express (a company that I have slowly come to really dislike) but they actually had someone contact me to try and resolve the issue.  I give them credit for paying attention, but still dislike them for their horrible credit practices.

  Read the article

• Tips on ensuring Model Quality

  - by [email protected]

  Given enough data that represents well the domain and models that reflect exactly the decision being optimized, models usually provide good predictions that ensure lift. Nevertheless, sometimes the modeling situation is less than ideal. In this blog entry we explore the problems found in a few such situations and how to avoid them.1 - The Model does not reflect the problem you are trying to solveFor example, you may be trying to solve the problem: "What product should I recommend to this customer" but your model learns on the problem: "Given that a customer has acquired our products, what is the likelihood for each product". In this case the model you built may be too far of a proxy for the problem you are really trying to solve. What you could do in this case is try to build a model based on the result from recommendations of products to customers. If there is not enough data from actual recommendations, you could use a hybrid approach in which you would use the [bad] proxy model until the recommendation model converges.2 - Data is not predictive enoughIf the inputs are not correlated with the output then the models may be unable to provide good predictions. For example, if the input is the phase of the moon and the weather and the output is what car did the customer buy, there may be no correlations found. In this case you should see a low quality model.The solution in this case is to include more relevant inputs.3 - Not enough cases seenIf the data learned does not include enough cases, at least 200 positive examples for each output, then the quality of recommendations may be low. The obvious solution is to include more data records. If this is not possible, then it may be possible to build a model based on the characteristics of the output choices rather than the choices themselves. For example, instead of using products as output, use the product category, price and brand name, and then combine these models.4 - Output leaking into input giving the false impression of good quality modelsIf the input data in the training includes values that have changed or are available only because the output happened, then you will find some strong correlations between the input and the output, but these strong correlations do not reflect the data that you will have available at decision (prediction) time. For example, if you are building a model to predict whether a web site visitor will succeed in registering, and the input includes the variable DaysSinceRegistration, and you learn when this variable has already been set, you will probably see a big correlation between having a Zero (or one) in this variable and the fact that registration was successful.The solution is to remove these variables from the input or make sure they reflect the value as of the time of decision and not after the result is known. 

  Read the article

• RightNow CX Cloud Service Combined with Oracle Fusion CRM in the Cloud

  - by Richard Lefebvre

  ·        The May 2012 release of Oracle’s RightNow CX Cloud Service, the customer experience suite, is now integrated with Oracle Fusion CRM, helping organizations to achieve sustainable business growth through relevant, cross-channel customer interactions that can increase revenue opportunities and drive organizational efficiencies. Relevant Interactions Build Stronger Customer Relationships ·          Armed with a comprehensive view of all customer interactions across channels, the context and status of these interactions, and an awareness of the customer’s value to the organization, companies can now offer more relevant products and services to customers. ·         Using the combined Oracle RightNow CX Cloud Service and Oracle Fusion CRM solutions, organizations can increase customer retention, drive higher levels of customer advocacy, and increase sales conversion rates with tools designed to: - Provide a complete, cross-channel view of the customer to sales, marketing and service. - Empower sales and service departments to easily collaborate to proactively solve customer issues, using opportunities to provide purchase advice at the right time and with the right solutions. - Allow sales to easily review service history in preparation for sales calls. - Enable agents to understand customer value based upon prior buying habits and existing opportunities. Deeper Insight Enables Targeted, Personalized Opportunities ·          The combination of Oracle RightNow CX Cloud Service and Oracle Fusion CRM allows sales and marketing organizations to simultaneously leverage service interactions from RightNow CX and sales prediction and segmentation capabilities from Fusion Sales. This helps companies to: - Better match products and services to specific customer needs based on customer service history.  - Deliver targeted, personalized interactions intended to help customers derive more value from purchases and to inform future buying decisions. - Identify new opportunities to increase deal size and conversion rates. Supporting Quotes ·         “Every interaction is a relationship opportunity to grow your business. When these interactions are relevant and add value for customers, customers are more likely to trust the relationship and seek purchase advice,” said David Vap, group vice president, Oracle. “This customer trust provides an opportunity to increase customer product adoption and to reduce the cost of customer acquisition, thereby increasing company profitability.” Supporting Resources ·         Oracle Fusion CRM ·         Oracle Fusion Applications ·         Oracle RightNow CX Cloud Service ·         OracleCRM on Facebook ·         OracleCRM on YouTube

  Read the article

• Is Infiniband going to get squeezed by iWARP and external QPI?

  - by andy.grover

  The Inquirer certainly thinks so.However, I'm not so sure it makes sense to compare Infiniband to an as-yet-unannounced optical external QPI. QPI is currently a processor interconnect. CPUs, RAM, and devices connected by it are conceptually part of the same machine -- they run a single OS, for example. They are both "networks" or "fabrics" but they have very different design trade-offs.Another widely-used bus in the system is closer to Infiniband than QPI -- PCI Express. Isn't it more likely that PCIe could take on IB? There are companies already who have solutions that use external PCI Express for cluster interconnect, but these have not gained significant market share. Why would QPI, a technology whose sweet spot is even further from Infiniband's than PCIe, be able to challenge Infiniband? It's hard to speculate without much information, but right now it doesn't seem likely to me.The other prediction made in the article is that Intel's 10GbE iWARP card could squeeze IB on the low end, due to its greater compatibility and lower cost.It's definitely never a good idea to bet against Ethernet when it comes to mass-market, commodity networking. Ethernet will win. 10GbE will win. But, there are now two competing ways to implement the low-latency RDMA Verbs interface on top of Ethernet. iWARP is essentially RDMA over TCP/IP over Ethernet. The new alternative is IBoE (Infiniband over Ethernet, aka RoCEE, aka "Rocky"). This encapsulates the IB packet protocol directly in the Ethernet frame. It loses the layer 3 routability of iWARP, but better maintains software compatibility with existing apps that use IB, and is simpler to implement in both software and hardware. iWARP has a substantial head start, but I believe that IBoE silicon will eventually be cheaper, and more likely to be implemented in commodity Ethernet hardware.I think IBoE is going to take low-end market share from traditional IB, but I think this is a situation IB hardware vendors have no problem accepting. Commoditized IBoE NICs invite greater use of RDMA features, and when higher performance is needed, customers can upgrade to "real" IB, maintaining IB's justification for higher prices. (IB max interconnect speeds have historically been 2-4x higher than Ethernet, and I don't see that changing.)(ObDisclosure: My current employer now sells IB hardware. I previously also worked at Intel. My opinions are my own, duh.)

  Read the article

< Previous Page | 1 2 3 4 5 6  | Next Page >