Wikipedia:Reference desk/Archives/Computing/2015 September 9

Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

September 9

What software can you use for producing Japanese style animations (anime)?

What programs do we use for animation movies like: Love, Chunibyo & Other Delusions? --Scicurious (talk) 00:24, 9 September 2015 (UTC)

I don't know about that movie, but it appears that a number of Japanese studios use RETAS (partial list here). Studio Ghibli uses Toonz. -- BenRG (talk) 04:23, 9 September 2015 (UTC)

Why are the apps in the Amazon play store the cheapest, in Microsoft/Android Store middle in price and on iOS the most expensive?

The Windows Phone User and the Amazon Phone user are maybe 0,5% of all phones so it would be more intelligent to make this price as expensive as posible because creating the app under BluescreenOS and FireOS is harder than creating under AndroidOS or iOS. And because the market for Andoird and Ipod/Iphone/Ipad is the biggest, it wouldn´t be necessary to make an app overpriced because there are more customers --Japanischindonesien (talk) 00:54, 9 September 2015 (UTC)

Because people pay for it? Costs do not matter. You pay the market price no matter it takes to produce something. It might also be that MS or Amazon are subsidizing the development of apps to their platforms, to make them more attractive to the users.--3dcaddy (talk) 01:09, 9 September 2015 (UTC)

In the past at least, Amazon's contract with developers allowed them to match the Google Play store price including discounts. I don't know if this term do exist, but if developers started regularly setting their prices significantly higher than Google Play store, you can be sure Amazon would do something about it. Also, I wouldn't be surprised if there are at least just many users of Amazon's store who don't have an Amazon phone. (Although probably most of them just got some of the free apps from the now defunct app of the day programme rather than buying anything.) Note also, despite the Amazon phone and tablet offerings small market share, it doesn't mean it's that hard to get stuff working from them. As I understand it, the Amazon alternatives APIs are generally quite comparable to the Google ones so you generally only have to do a little work. Meanwhile, there are only a few Amazon devices and 0.5-2% is [1] surely far larger a market share than than many of the million different Google Play Android devices out there have. Nil Einne (talk) 01:40, 9 September 2015 (UTC)

Looking more carefully at the agreement [2], it sounds like they can still change the price to match Google Play (and probably other stores) at will

Royalty. For each sale of an App, the responsible Amazon Party will pay you a royalty (“Royalty”) calculated as follows (where List Price is defined in and subject to Section 5a of this Schedule):

5a is as below:

The “List Price” for an App for an Amazon Marketplace or a country is an amount that does not exceed, at any time, the lowest list price, suggested retail price, or actual price you set for that App (including any similar edition, version or release) for any Similar Service in a country served by that Amazon Marketplace (for Mobile Apps) or country (for PC Games and PC Software).

....

You will update the List Price for each App as necessary to ensure that it meets the requirements of this section. We have sole discretion to set the retail price and other terms on which we sell Apps.

Taken together, these seem to say to my non lawyer reading that they can set the retail price whatever they want. They're still supposed to pay you based on the list price (for Android stuff, not for PC software). However the list price is not simply what you set it, but at worst the same as the lowest list price/actual price, on other stores. So if they change the retail price because it's cheaper on Google Play or wherever, they will surely only pay you on this "correct" list price, not the invalid list price you set.

Nil Einne (talk) 02:12, 9 September 2015 (UTC)

-> "Costs do not matter." <- for me it would matter. If I have to make an app for an OS what is used by less than 50 million persons it sounds to be more expensive to pay someone who develop the app and I would make this app more expensive and not cheaper than the App for an OS with over 500 million persons ..--Japanischindonesien (talk) 04:01, 9 September 2015 (UTC)

Then you're likely approaching this from the wrong way. Most people with some degree of business sense will consider the price elasticity of demand far more important than costs and hope to set the optimal price (although in reality I suspect most don't really analyse the market that well and just go by feeling). There's no point setting you app at a higher price to make up for the higher costs, if what actually happens is fewer people it and you make less money overall. That's just cutting off the nose to spite the face. If you think the costs/potential profit from an app on a certain store isn't worth it, your best bet is simply to not develop for it rather than worrying so much about the cost that you set an excessively high price that you make less money. Nil Einne (talk) 04:13, 9 September 2015 (UTC)

BTW, the marketshare of the device will generally matter less to you than the number of real paying customers you get. For example it's commonly recognised that while there are far more Android phones than there are Apple ones, developers can often still make good money perhaps even more on the iOS stores because more people are willing to pay for apps. It doesn't help as I mentioned earlier that cost wise, developing for an Android phone is generally more due to device fragmentation. Net result of this and other factors is that many developers do release on iOS first and some even take quite a while to reach Android even with Android having the far larger device marketshare.

Also I didn't mention this earlier but I'm not sure why you think it costs more or is harder to develop for the Amazon or Windows devices. It's possible it will, particularly for Windows devices due to fewer capable programmers, presuming your hiring someone for the purpose, or because you have less experience. Although as mentioned device fragmentation on Android can be a big issue. However it's also easily possible it doesn't especially considering Amazon devices vs non Amazon Android devices. (Although as mentioned earlier, if you're planning on releasing to the Amazon store most people won't just target Amazon devices.) The average cost per sale may very well be higher due to fewer sales, but that's distinct from the overall cost and difficulty.

Nil Einne (talk) 04:35, 9 September 2015 (UTC)

I am pretty sure the bit "costs do not matter" means, 1 - the buyer does not care about costs, he cares about what he gets for his buck. And 2 - the producer can not price his products higher just because it took more to produce, he is aware of point 1. As Nil Einne says above "no point setting you app at a higher price to make up for the higher costs", given point 1. --Scicurious (talk) 16:14, 9 September 2015 (UTC)

Why can't Sony just produce more PS3 dongles for guitars?

Quiet every 3rd guitar for sale in eBay is without dongle because the owner lost his one or other problems..--Japanischindonesien (talk) 00:57, 9 September 2015 (UTC)

I presume you're talking about the "Guitar Hero" guitars? I suspect they can make more money selling entire replacement guitars? (Also, the guitars aren't a Sony product - I think they come from Activision, although there may well be some knock-offs out there). SteveBaker (talk) 01:12, 9 September 2015 (UTC)

You probably can get a replacement dongle, consumer law in numerous places (e.g. NZ [3]) requires replacement parts are available and if these are clearly separate parts manufactured separately there's no reason why you can't produce a few more of one than the other. But you'll probably find the price of a replacement dongle means you might as well just buy a new guitar. (Consumer law may also require replacement parts are at a reasonable price, but since that's such an iffy thing, for this sort of thing it's basically meaningless.) Nil Einne (talk) 02:17, 9 September 2015 (UTC)

When writing the above, I was assuming that the guitars are still produced. From some quick searches earlier, I'm not sure if this is the case. If it isn't and they haven't been for quite a while, by now it may be difficult or impossible to get just dongles from Activision or whoever sold them. Nil Einne (talk) 11:59, 9 September 2015 (UTC)

Different types of 3d drawings

It's not the same to draw a cube or a perspective, than to draw a cube or a perspective where you can move around to see it from a different angle. How do you call this difference in 3d? The first can be done with any type of drawing program (rightly called 2d) like GIMP or on paper. The second would need 3d (wrongly called) programs like Bender.--3dcaddy (talk) 16:45, 9 September 2015 (UTC)

2.5D seems to be what you are searching for.--Scicurious (talk) 23:09, 9 September 2015 (UTC)

3dcaddy, I want to make sure I understand you. Here are two examples that I think match what you're talking about:

• Imagine you kept your eye in a fixed position, looked through a glass pane at a box, and traced the image of the box on the glass pane using a marker. You could take the glass pane with you, but it might be hard to tell the exact size of the original box from just the marker lines. And it might be hard to create a differently rotated view of the box from just that image.

• If you measured the exact dimensions of the box and wrote them down, it would be easier to create differently rotated views. Or at least, it would be easy for a computer to create any rotated or zoomed view of the box. When the computer has the exact dimensions of the box, it can use perspective math to create any view you want.

3dcaddy, you asked "How do you call this difference in 3d?" I'm unclear what kind of answer you need. I'd call the image in the first example something like a "perspective image", and the dimensions data in the second example something like "3D data". Does that answer your question? --Bavi H (talk) 01:48, 10 September 2015 (UTC)

• Yes, your insight helps understand the difference.--3dcaddy (talk) 03:34, 10 September 2015 (UTC)

There is really no difference in the process of DRAWING the cube in blender and GIMP. In each case, the three-dimensional coordinates of each vertex is reduced from three numbers (x,y,z) into two (x,y) using some process or other. Perhaps an isometric projection (where z is basically just discarded) or a perspective projection (where x and y are divided by z). Either way, the 3D cube is reduced to a 2D figure.

The significant distance is that blender remembers x, y and z and repeatedly performs the projection into 2D as you move the object to give the illusion of a three-dimensional object on the screen. GIMP does the conversion once - and then you're stuck with an unchanging 2D projection. Every step of the 3D to 2D projection is likely to be the same in both cases - the only distinction is that GIMP only retains the 2D projection where blender does the same calculations over and over to produce motion - and therefore retains all three dimensions of the original data.

The concept of two-and-a-half dimensions isn't a real mathematical concept. It's used in some situations where (typically a video game) does the projection from 3D to 2D just once using an isometric projection - and then composes a 3D-looking scene by layering those 2D images - possibly moving them relative to each other to create the illusion of 3D motion.

To say that blender isn't "really" a 3D program is incorrect. All of the calculations and geometry that it performs are done with three dimensions - only the final display is reduced to two dimensions because we (mostly) only have a 2D computer screen. But if you invest in a true 3D display (an Oculus Rift headset, for example) - then the object will appear to be in true three-dimensional form.

(Actually, GIMP is a bad example here because it doesn't even retain knowledge of lines, vertices and faces - it retains only the pixel data. A program like Inkscape does retain these properties of the original object - just as blender does - but only in two dimensions. So in effect, every video frame, blender converts the 3D data into an Inkscape-like 2D representation and then that 2D representation is rendered into a GIMP-like collection of pixels. Inkscape does the 3D to 2D conversion just once (if you use the cube-drawing tool) - but does the conversion from lines and faces into GIMP-like pixels every frame. GIMP does 3D to 2D to pixels just once and retains only the pixel data.)

(And to drill down one level further, blender feeds true 3D data to your graphics card, which does the 3D->2D->pixels conversion automatically - inkscape retains the 2D information as an SVG hierarchy, which is handed off the the graphics card to do 2D->pixels automatically. GIMP does the entire process just once, in software.)

SteveBaker (talk) 02:22, 10 September 2015 (UTC)

• Thanks for this thorough explanation. --3dcaddy (talk) 03:34, 10 September 2015 (UTC)

Some associated references cover most of this - raster graphics (e.g. GIMP) vector graphics (e.g. Inkscape), 3D graphics (e.g. Blender). It's almost like we have a whole encyclopedic collection of relevant information and sources here ;) SemanticMantis (talk) 06:02, 10 September 2015 (UTC)

Also though the term can be used sloppily in computer graphics applications, fractional dimension AKA fractal dimension does have a well-defined meaning in mathematics. Or rather several related meanings, e.g. Hausdorff dimension, Minkowski dimension, and a handful of others. In other words, saying a mathematical object has dimension=2.5 is perfectly cromulent in formal mathematics, since at least 1918. SemanticMantis (talk) 06:07, 10 September 2015 (UTC)

While you're technically correct that some kinds of weird mathematical objects might happen to have precisely 2.5 fractal dimensions - that's definitely NOT what our User:Scicurious is referring to here. Our article 2.5D explains exactly what is meant here - and the meaning is not a mathematical meaning - it's a description of a technique that more or less looks like orthographic/isometric 3D graphics using nothing but 2D rendering techniques. What it emphatically isn't is some kind of intermediate between two and three spatial dimensions as Scicurious seems to be implying. A classic example of a poorly chosen name. SteveBaker (talk) 18:59, 11 September 2015 (UTC)

It is possible to run 2 os at the same time without emulator?

Assuming you dont care about this entire situation being a cooperative multitasking-like thing (needed thing or this question answer would be obviously a no), it is possible to run 2 os at the same time without emulator?
PS: This is not about dual booting, on my idea both os would be loaded but (for the sake of this idea being maybe possible) one will be active and runing, and the other(s) would not be active and be freezed, until you decide to make it active (making the current active os not active to do it).201.78.128.120 (talk) 18:29, 9 September 2015 (UTC)

You'll need per definition a virtual machine of some kind. Look at hypervisor, which is a virtual machine, but of a different type. --3dcaddy (talk) 19:03, 9 September 2015 (UTC)

The problem is, which OS controls the devices? coLinux can do this, but it delegates device control to MS-Windows. LongHairedFop (talk) 19:38, 9 September 2015 (UTC)

Remember this would be a cooperative multitasking-like thing as being Preemptive multitasking like thing would nullfy my question, making the answer a obvious, no.201.78.128.120 (talk) 20:07, 9 September 2015 (UTC)

coLinux does not support multi-processor, nor 64-bit Windows or Linux. I wonder this is of any use. --Scicurious (talk) 22:59, 9 September 2015 (UTC)

NTVDM, Wine, etc. Also, Windows 9x was basically MS-DOS underneath, so opening a DOS prompt would be like running another OS. - TheChampionMan1234 05:27, 10 September 2015 (UTC)

If you want to run two OSes in parallel with each owning the hardware while it's active, they would need a way to partition the RAM amongst themselves, and there would need to be some shared code for switching between them, and all drivers (including third-party drivers) would have to support handing off control of the hardware. This isn't necessarily impossible because it's not too different from what you have to do when suspending and resuming. But the major OS vendors would have to cooperate in defining and implementing a spec, and realistically I don't think that'll happen.

One thing you probably can do today is alternately hibernate and de-hibernate two OSes. It's a kind of dual booting, but it preserves running processes and open documents across switches. -- BenRG (talk) 05:59, 10 September 2015 (UTC)

"...OS vendors would have to cooperate in defining and implementing....."Thanks for the answer,this means what I means its impossible since I wouldnt be able to get some random OS (for my architecture of course) I found on net and use it.201.78.128.120 (talk) 12:29, 10 September 2015 (UTC)

See the comparison of platform virtualization software. Such software might be an minimized host-OS itself, running various virtual machines. The CPU switches from task to task by changing all it registers what last extensions of CPUs do. These CPUs have multible registers loading switching them on demand or waiting for cache data transfer. The GPU needs also to process on the video memory, but bringing the picture to the screen is still the RAMDAC or grabbing it compressed over network. The guest-OS just needs its drives, stored in an image on a physical disk oder raid. If you rent a server for more than a webpage only, it is a virtual machine anywhere in a server farm. The differences are the performance and accessibility of your machine at home, booting the user friendly host-OS first. --Hans Haase (有问题吗) 14:09, 11 September 2015 (UTC)