SillyCow

Member
  • Content count

    386
  • Joined

  • Last visited

Community Reputation

1472 Excellent

About SillyCow

  • Rank
    Member

Personal Information

  • Interests
    Programming
  1. Why A.I is impossible

    This is actually what is at the center of the recent A.I. or "Deep Learning" revolution. You are not longer triggering code from specific events. Rather: A modern neural network is something that is not programmed (in the traditional sense). You supply it with some inputs, and you give it a goal. It then goes on to develop (or program) all the skills that it needs by itself. You are not in control for how it achieves it's goal. An example (that has not really happened yet): You give a swarm of neural network robots a useful goal like: "Clean the streets". This is a very simple goal... Their inputs are: A camera, and some form of radio communication to coordinate with eachother. You don't tell them what dirt looks like, and you don't supply a communications protocol. You do grade them by how clean the streets are in the morning and how much fuel they've wasted. Years pass, and the robots gradually get better. The thing to be expected, is that they learned what dirty things look like . However, you find that the communications protocol they are using is very similar to a spoken language. These robots actually learned to speak, and further more invented their own language. Now it's true, intelligence of such magnitude out of reach of today's neural networks. However the principle is the same: Machines are responding to stimuli in ways which they were never programmed to do. This is why "Alpha Go" and "Alpha Zero" are so special: Unlike deep blue, and previous A.I: Nobody ever taught them how to play "chess" or "go". One could claim that this is the start of "creativity" in machines. And putting "consciousness" aside for a minute. This is the big deal with the latest developments. We have machines which can be "creative". As a side note what I find even more threatening is: Unlike "Alpha Go" which was playing a human, "Alpha Zero" was playing the best computer program that many humans have studied and researched. I don't think there exists a game that is more studied than chess. And the shear amount of computer science skill directed at building chess engines dwarfs any other gaming AI research. And alpha zero beat all of that human effort, without building on top of it. (It learned how to play chess from nothing). Basically it means: There will soon come a time when computers will be able to write better computer programs than humans. Even worse, these programs will be so complex that they will be beyond human understanding (you will not even be able to help debug them...). I don't know about you, but as a "traditional" software dev, this makes me nervous about my job security.
  2. Unreal Memory Allocation

    This was actually my next step. Does unreal have a realtime (live) video codec for android? Or should I go about compiling FFMPEG? (Is there a mobile FFMPEG plugin allready available?)
  3. Unreal Memory Allocation

    The memory is released when ImageWrapper is destroyed. That is not my current problem. However, I do not see how can gain control of when ImageWrapper allocates this memory. I downloading mage, The I call IImageWrapperPtr->SetCompressed with some jpeg data. IImageWrapperPtr does some magic and decompresses this some buffer that is allocated internally. This is by far the biggest memory allocation, because the uncompressed image data is the largest data I get a pointer to this data using GetRaw(). What I would like to do is prevent IImageWrapperPtr from allocating this uncompressed data buffer in the first place. If possible, I would like to feed it a pre-allocated work area, and reuse this work area for every subsequent image that I need to decompress.
  4. I am pretty new to Unreal development. I am trying to fix some memory allocation issues that I have. I am streaming jpegs to a texture at ~5FPS on an Android app. (the texture updates at 5FPS) To unpack the texture jpeg, I am using this code: const TArray<uint8>* UncompressedBGRA = NULL; if (!ImageWrapper->GetRaw(ERGBFormat::BGRA, 8, UncompressedBGRA)) However, as far as I can tell this allocates a new bit buffer every time. And my app eventually grinds to a halt and crashes. This is "partially fixed" if I call: GetWorld()->ForceGarbageCollection(true); "Partially fixed": Meaning the app doesn't crash anymore, but the app still stutters. Is there some way to get Unreal to unpack a jpeg in place? Or do I need to compile some other C++ jpeg decoder?
  5. Unreal Learning Unreal C++ development

    I'm not clear about how to copy paste blueprints from online tutorials. From what I see they are just JPEGs of blueprints. As such, I cannot even copy paste the names of the fields. It's just a picture... As mentioned I have already had a look at many of the official tutorials, but they seem to be very poor. Take this one for example: https://docs.unrealengine.com/latest/INT/Programming/UnrealArchitecture/Actors/index.html. It is very short, and barely contains any code samples. Some of the links that you follow only contain small fragments of code. In C++ this is doubly problematic, because questions such as "which *.h file should I #include for this class/function" are not trivial in C++ . Beginner tutorials should contain entire files/classes. Prefereably, they should also contain the "click heirarchy" when you have to click through very complicated submenu heirarchies. The only "complete" tutorial that I found on their site was: https://docs.unrealengine.com/latest/INT/Programming/QuickStart/index.html . But this one is really short, and does not teach anything beyond "hello world". When ever I want to learn a new framework/programming language online, I usually work by example. I follow someone along as they slowly build a sample project (as in the Quickstart tutorial). However the official Unreal site seems like it is trying to provide something more like a reference guide. This leads me to believe that the official site is not suited for my learning style. That's why I am looking for some more "by example" programming tutorials.
  6. Hi, I am trying to build a small POC in Unreal. I consider myself a pretty good C++ dev. I have a pretty good knowledge of Directx/Opengl, and have programmed my own game engines before. I have also developed projects of various sizes using Unity. I would gladly keep using Unity, but cannot due to licensing issues. So I am trying to learn Unreal to overcome the licensing issue. However, I cannot find decent docs and tutorials for programmers. Most tutorials use Blueprints, which as a "coder" I find very hard to read (not to mention no copy+paste). Also the editor UI evolves rather rapidly, thus making alot of tutorials outdated. ( I often find myself searching for submenus that no longer exist ) While my project is very small, I am trying to do something which is at an intermediate level: I want to download images from the internet and dynamically load them onto 3d meshes. I have only found some partial solutions on Unreal forums. And have only found 2 very basic C++ tutorials on the official unreal site. I guess what I am asking is: Regardless of my specific project, can you recommend a comprehensive "coder" oriented tutorial or book ? I am not looking to create complicated scenes or effects. I just want to instantiate a mesh, and play around with it's attributes at runtime.
  7. How far away do you think that will be? I haven't seen any mainstream reports of trying to capture full body VR. The one thing I've heard which made sense was the wireless room tracking proposed with Daydream Standalone. But that got cancelled by HTC, with no release date from lenovo from over a year. Not to mention that google Daydream isn't exactly taking off. Oculus said they'd have their own wireless headset. But then they also said it would cost $199 next summer. This, again, sounds illogical to me. How are they expecting to make any money? This isn't a Kindle Fire... They don't have a steady stream of people purchasing stuff through the device. A decent mobile VR set is going to need: 1. A res IPS screen (hi res for phones) 2. High powered mobile CPU 3. A BIG battery All of which are very expensive. (Not to mention the regular stuff: Headphones, Lenses, Plastic, handheld control...). We can estimate that other stuff at ~100$ because that's what the GearVR is selling for, And I have a feeling the GearVR is a promotional for Samsung , a device only meant to sell more Samsung flagship phones. Which makes the 199$ price sound unreal. Either that, or they're planning on losing money. Which makes little sense to me, because they don't have the same ulterior motives as Samsung, Amazon, and Playstation, which can all sell at a loss or at cost value.
  8. Playing around with the occulus touch, I am trying to understand the type of ecosystem that the VR market is trying to build. I am especially bewildered at the prices things cost: The Oculus Rift and PS VR have gone significant price cuts which would have me think they are not selling for a profit. Or at least not a reasonable one considering the risk. This is understandable with Sony, as console H/W has traditionally been sold at a loss. But what are Oculus/Facebook's motives here? It should be clear to FB by now that VR is not the next social platform. Or at least I don't see any convincing attempt by FB to write apps for VR (The videos app is worse than Youtube 360 on google cardboard...). So what's with the price cuts? Aside from that, the game authors have me bewildered: So far, I have tried only 2 "AAA" single player experiences. AAA is a relative term, because the VR market is smaller. I have tried "RoboRecall" by Epic, and "Archangel". Both games have high production values, are professionally made, and are fun to play. However both games have very repetitive content and short play times: Around 4 hours of play, and assets, and mechanics are repeated heavily. The lack of versatile mechanics and repetetive 3D models leads me to think that relatively not alot of work has been put into their game design and art department (Relative to a AAA game). However both games are selling for over 30$, and the common launch price for a AAA VR game is 40$. Yet the content is parallel to a (very good) mobile game. Knowing Epic, I wasn't expecting them to try and earn money with their game, but rather use it to showcase their engine. My personal thoughts are that it takes less effort to make a VR game. They tend to have smaller worlds (because it's hard to move), and less versatile game-play mechanics. The only thing more expensive is the risk factor. Although the current "AAA" VR library is so small, that you don't have alot of competition. So that partly makes up for the risk. VR integration comes built in with game engines with some very easy to use APIs. So the tech side of VR for a software dev is laughably easy. The challenge is in designing the game mechanics. So what's going on here? Both Facebook and Epic are big companies with well established revenue streams and business plans. What's the big idea? How are they expecting to make a profit by selling a VR helmet with 2 controllers for $400? Are they expecting the masses to buy 4 hour games for 40$? I am not writing this cynically. I am both a VR enthusiast and a hobby VR developer. I desperately want VR to succeed, even if it never becomes mainstream. I am also aware that VR has all of the chances of being a passing fad (like 3D TVs). But the way this is being approached by these big companies makes me feel that the bigger players are going to ditch this due to lack of profits. Or maybe this is the just the "launch" business model, and it will evolve... Afterthought: Multiplayer games are awesome in VR. Standing next to someone in VR and giving them "high fives", turning your head to talk to them, gives a whole new sense of presence to "multiplayer". Maybe this is really the next big thing in social networking. But I exepcted FB to have made a "FB hangouts" app like "Rec-room" by now...
  9. What do you value more? A. Having alot of people playing your game, and giving you feedback? or... B. Possibly (not likely) "making" $100 for something that probably took you months to make? (There are quicker ways to make 100$)
  10. More often than not it's either an HDD problem or Lack of RAM.
  11. Can my PC handle a VR upgrade?

    That's good to know. I wasn't aware of that. So I'll be getting the Oculus then. SteamVR was the only thing I was worried about from a software perspective.
  12. Can my PC handle a VR upgrade?

    @Frob, thanks for the detailed response. It's really helpful @Hodgeman. Thanks. I would actually also prefer the Vive. I like HTC and Steam. I really like the Steam integration. However the Vive is tremendously expensive in Europe. HTC is selling it for 825 euros. (I can get it from other non-oficial importers for 700 euros). So it's really not in the same price range. I can get 2 Oculusses for that price. So I think the Vive isn't an option for me.
  13. Can my PC handle a VR upgrade?

    I recently got a 600 euro gift card for Amazon. And I like messing around with VR. I played some games, and I program hobby VR games in my spare time on mobile.I saw that my 5 year old CPU meets the specs for the Oculus rift. However my current graphics card does not. I see that in Europe, I can get an "Oculus touch bundle" for 400 euros. And an Nvidia 1060(3GB) for another 200 euros. (coincidentally this is the same sum of my gift certificate) The two possible problems: I only have 4GB of RAM (I don't think this will be a big deal, maybe just some more loading times) I am not sure my power supply can support a 1060 (I am not sure what the different WATTages that are printed on it mean ) Here is a photo of my power supply's spec: https://ibb.co/eaOOOR It was pretty basic even 6 years ago. Is it good enough to run a 1060? (I can only find a link for the 6GB model: https://www.geforce.co.uk/hardware/desktop-gpus/geforce-gtx-1060/specifications) I am not interested in going through the trouble of replacing my power supply, and I do not want to spend more than 600 euros on this entire endavour. And another question: Assuming I want to spend this money on VR, is there a better platform to spend this sum on? I was going to wait for the Vive focus (I don't liike cables), but it looks like it's going to be a dud (only in china, no real ecosystem). Facebook's claim of releasing a wireless rift for 200$ seem dubious to me. And since that price sounds like they are selling at a loss, there is no chance it will reach Europe for that price on official channels.
  14. Here's a thought: The cost of the game is not dependent on what what is the bare minimum needed to make it. The cost is determined based on the question: "How much money can this game make?" If the games market keeps growing, then the games budgets will keep growing. Let's presume that the best xMass FPS release will make 500M$ on playstation. If I knew you were making a good game for 10M$ which "win" that market, would I not spend 110$M for a slightly better game? Would it not be worthwhile to spend another 50M$ on marketting, etc...? That brings me to the conclusion that AAA game budgets will always be a factor of: How much money can I make vs. What is the risk? Indie games will always follow their own rules though. Just as a metaphor: Have film budgets gone down because digital cameras/prodcution are cheaper? No: James Cameron has started designing his own super expensive cameras (millions of $$$ each). Why take up this ridiculous endeavour? Because each of his films can make a billion dollars. You can see the same happening in YouTube. Even though video production started out cheap, once people started making a little money ($100k), the production budgets went up. I would guess that successful you tubers have the same production budget as a cheap "local news" tv show. You can see the same happening in the mobile app store. You will not see a lot of "flappy bird"s hitting #1. But because the market is still smaller (revenue wise), you will not see alot of AAA budgets either. So I think that new technology brings the possibility of cheaper products, but a bigger market will *necessarily* mean a bigger development budget.
  15. Does violence stem from video games

    The way I grew up (no ebooks): Books had a beginning and an end. When you were done with a book, you had to get a new one (buy/borrow). You had to contextualise and sync into a new book. Alot of computer games are neverending. For example: You can allways play another RTS skirmish. When I sink into a new book, I can cut myself off for days. But it only happens ocasionally, because once the book is over, so is my compulsion to read it. It's like comparing a movie to a series to a soap/reality/cable news. You will not spend alot of time watching movies. The narattive usually ends after several hours. You *can* spend alot of time binging a show. You most definitely will spend too much time watching a soap-opera a reality show or the daily news. In fact, they are designed with exactly that goal in mind. Many countries have a dedicated 24hr "big brother" stream which constantly broadcasts the show's participants to encourage you to never stop watching. I guess what I'm trying to say, is that TV and Video games are in another league of addictiveness. Books are not addictive enough to become a big problem for most people. However games, and facebook are. And a large percentage of the people I know (I hesitate to say "most") are addicted to TV. The one thing games have "going" for them is that they are still less popular than TV": An older parent is much more likely to encourage and participate in a TV addiction with their children than a computer game. ex: eating dinner regularly in front of a TV. I don't think people realise how much time,experiences, and personal development they lose to electronic media. I don't think the size of the problem is remotely comparable to books. I know many people in my environment who spend several hours daily watching reality shows and cable news and play computer games. I know very few people who read a book a week [although I know some :-) ]. Also parents tend to sit their *toddlers* in front of a TV to keep them quiet. Which sends a very bad subliminal message such as: "You should be watching TV, because that is what people do at home". It creates an impulse at an early age for a developing child to enter the house and hook up to a screen. It also kills any conversation/talking that the members of the household might initiate. am afraid that computer games have a similar effect. At least they are not very accessible to toddlers yet. Although mobile interfaces are changing this drastically. You don't need to sit a table enjoy a touch screen game. I have seen many two year olds happily tapping away for hours on their parents' phones. One of these times was on a camping trip. I don't think the child was even aware that there was a happy campfire meeting around them. It makes me sad... One sign of addiction is remorse: How often have you read a book over the weekend and said to yourself: "What a total waste of time, I should have gone to that party, got some work done, or attended that family meeting". While this has occasionally happened to me with books, It happens to me much more frequently with electronic media. I would say at least once a month.