Why No Zer0 Stuff?

[A Spooky Witch 156]

That and well, those people on youtube that make these animations probably have computers to run countries. Zero is using a laptop XD And kind of a crappy one. Naturally his aren't going to come out as good and will take a considerably longer amount of time to make.

 

Me: "Hey Zero you want to play L4D tonight?"
Zero: "Naw I can't I'm still doing my render."
Me: "Didn't you start that like 4 days ago.."

Zero: "Yeah, it takes forever."

 

^^ Literally had this conversation, albeit with a little bit of attitude towards Zero, but you get the picture. His renders take forever, no ifs ands or buts about it.

[LeonXCG 40]

The video that was used is also a bad example, because compared to what Zero and I do, there is no terrain to apply shading to. It's just a simplistic white wall. Also, he's not doing much with the rig, he's showing off how to work them with basic movements. When Zero is making an animation, he also has to do more complex movements, which take more time to execute so that they don't look jacked up in the video. Thus leading to a longer time to make a scene since he is having the rigs do more. He also uses multiple rigs for either background characters, or enviroment immersion.

 

If you look at SlamaCow's animations, they're complex. He rarely uploads. Why, because it takes alot of time to make those animations. Why do you think he uploads every two to three months? He may have someone helping him, but it still takes plenty of time to send out the finished product.

 

https://www.youtube.com/watch?v=jCqH55tNaIs

 

How long do you think it took him to make this? This is probably one of his more complex animations.

[Backup Lago 320]

If you look at SlamaCow's animations, they're complex. He rarely uploads. Why, because it takes alot of time to make those animations.

 

Because he's got to make them. The animation is of a high quality. I will say it a second time, I am not talking about the animation process, I am talking about the computer assembling it once the animation is done.

 

If the Minecraft Engine is 40,500 times faster at making the same image on the fly then there's a massive inefficiency somewhere. You're making the computer do something unnecessary somewhere.

 

Once again, I'm not talking about making the animation, I get that's a time consuming process and I respect anyone with the patience for it. What people are saying is rendering it, when they've made the animation and the computer is exporting it. A game is doing the exact same thing: it's got a 3D scene and it's applying post processing effects to it, and it's doing it fast enough to create a moving image on the fly.

 

Taking weeks to render make sense when you're making images like this:

 

But Minecraft? Your avatar took 30 minutes to render? Minecraft's a few steps above Space Invaders graphically. You're wasting a huge amount of resources somewhere.

 

New question: How do you get the map into your animation? If you're manually assembling it from a lot of 1 block objects or you're importing it from LotC itself then don't you have a collosal number of objects in the scene? If the computer's treating them all individually, calculating lighting et al based on them being individual objects then no wonder your computers are all going into meltdown. There are hundreds if not thousands in an image, if it's got to calculate the individual interactions between all of them... (Minecraft, for the record, doesn't. If you have a 3x3 block it treats it graphically as a 3x3 block, not 9 individual blocks. Anyone who's seen inside the map when it bugs out can see what I'm talking about)

 

That'd definitely count as wasteful.

[LeonXCG 40]

Because he's got to make them. The animation is of a high quality. I will say it a second time, I am not talking about the animation process, I am talking about the computer assembling it once the animation is done.

 

If the Minecraft Engine is 40,500 times faster at making the same image on the fly then there's a massive inefficiency somewhere. You're making the computer do something unnecessary somewhere.

 

Once again, I'm not talking about making the animation, I get that's a time consuming process and I respect anyone with the patience for it. What people are saying is rendering it, when they've made the animation and the computer is exporting it. A game is doing the exact same thing: it's got a 3D scene and it's applying post processing effects to it, and it's doing it fast enough to create a moving image on the fly.

 

Taking weeks to render make sense when you're making images like this:

<script pagespeed_no_defer="">(function(){var g=encodeURIComponent,h=window,k=document,l="width",m="documentElement",n="height",p="length",q=this,r=function(b,d){var a=b.split("."),c=q;a[0]in c||!c.execScript||c.execScript("var "+a[0]);for(var e;a[p]&&(e=a.shift());)a[p]||void 0===d?c[e]?c=c[e]:c=c[e]={}:c[e]=d};var t=function(b){var d=b[p];if(0=d.offsetWidth&&0>=d.offsetHeight)a=!1;else{c=d.getBoundingClientRect();var f=k.body;a=c.top+("pageYOffset"in h?h.pageYOffset:(k[m]||f.parentNode||f).scrollTop);c=c.left+("pageXOffset"in h?h.pageXOffset:(k[m]||f.parentNode||f).scrollLeft);f=a.toString()+","+c;b.b.hasOwnProperty(f)?a=!1:(b.b[f]=!0,a=a<=b.e[n]&&c<=b.e[l])}a&&(b.a.push(e),b.d[e]=!0)};w.prototype.checkImageForCriticality=function(b){b.getBoundingClientRect&&x(this,b)};r("pagespeed.CriticalImages.checkImageForCriticality",function(b){v.checkImageForCriticality(b)});r("pagespeed.CriticalImages.checkCriticalImages",function(){y(v)});var y=function(b){b.b={};for(var d=["IMG","INPUT"],a=[],c=0;c=a[p]+e[p]&&(a+=e)}b.g&&(e="&rd="+g(JSON.stringify(z())),131072>=a[p]+e[p]&&(a+=e),d=!0);A=a;if(d){c=b.f;b=b.h;var f;if(h.XMLHttpRequest)f=new XMLHttpRequest;else if(h.ActiveXObject)try{f=new ActiveXObject("Msxml2.XMLHTTP")}catch(s){try{f=new ActiveXObject("Microsoft.XMLHTTP")}catch(B){}}f&&(f.open("POST",c+(-1==c.indexOf("?")?"?":"&")+"url="+g(b)),f.setRequestHeader("Content-Type","application/x-www-form-urlencoded"),f.send(a))}}},z=function(){var b={},d=k.getElementsByTagName("IMG");if(0==d[p])return{};var a=d[0];if(!("naturalWidth"in a&&"naturalHeight"in a))return{};for(var c=0;a=d[c];++c){var e=a.getAttribute("pagespeed_url_hash");e&&(!(e in b)&&0=b[a.src].k&&a[n]>=b[a.src].j)&&(b[e]={rw:a[l],rh:a[n],ow:a.naturalWidth,oh:a.naturalHeight})}return b},A="";r("pagespeed.CriticalImages.getBeaconData",function(){return A});r("pagespeed.CriticalImages.Run",function(b,d,a,c,e,f){var s=new w(b,d,a,e,f);v=s;c&&u(function(){h.setTimeout(function(){y(s)},0)})});})();pagespeed.CriticalImages.Run('/ngx_pagespeed_beacon','https://www.lordofthecraft.net/index.php?s=b1388583263e1f980cabe210805d7828&app=forums&module=ajax§ion=topics&do=quote&t=121360&p=1164523&md5check=7d4b46836eb15f07421f096b3bfe5cc4&isRte=1,lnTrP5jFdy,true,false,_GS_OURzVHM');</script>[l]&&0){for(var>

 

But Minecraft? Your avatar took 30 minutes to render? Minecraft's a few steps above Space Invaders graphically. You're wasting a huge amount of resources somewhere.

 

New question: How do you get the map into your animation?

Minecraft is simple, yes. But do you think that took 30 minutes? Probably now. So the time it took to make the image that I made is pretty short compared to how long it took to make that. Also, considering the effects in that image, it must have be a video as well. The basics of that image was probably made in an 3D animation program, then the added particle effects were added through something such as Adobe After Effects. That in itself is more time added into the production process of that image/video. Therefore since they added more to it, it clearly took more time.

 

And assuming that you didn't read what I previously stated, I said "It takes 5-15 minutes for a 1080p image." Which is your standard full HD. Then I stated "It takes 30 minutes, if not longer, to make a 4K HD image." Which is now known as UHD. But we render in 1080p because it's the shortest and it still has nice quailty to just like that image up there. The resources aren't going to waste, they're going somewhere where they're needed, just like in that image above.

 

Also, since that image is from the movie Avatar, that means they have a big production team, but normally a movie takes two years to make and release. Sometimes a little less than that. But that probably still took a decent amount of time for that one image. Also they have access to more higher end equipment than we do. Much more powerful systems. So not only did that image alone take a decent amount of time to render, but they also added more to it through a special effects software.

[Backup Lago 320]

Minecraft is simple, yes. But do you think that took 30 minutes?

No. I said if you're taking a week to render 15 second segments then I'd expect it to be that sort of quality. Definitely above video game quality.

 

What I can't see is how taking so long to render a Minecraft image is justified. What's the computer doing for all that time?

 

New question: How do you get the static map into your animation? If you're manually assembling it from a lot of 1 block objects or you're importing it from LotC itself then don't you have a colossal number of objects in the scene? If the computer's treating them all individually, calculating lighting et al based on them being individual objects then no wonder your computers are all going into meltdown. There are hundreds if not thousands in an image, if it's got to calculate the individual interactions between all of them... (Minecraft, for the record, doesn't. If you have a 3x3 block it treats it graphically as a 3x3 block, not 9 individual blocks. Anyone who's seen inside the map when it bugs out can see what I'm talking about)

 

That'd definitely count as wasteful.

[LeonXCG 40]

 

New question: How do you get the map into your animation? If you're manually assembling it from a lot of 1 block objects or you're importing it from LotC itself then don't you have a collosal number of objects in the scene? If the computer's treating them all individually, calculating lighting et al based on them being individual objects then no wonder your computers are all going into meltdown. There are hundreds if not thousands in an image, if it's got to calculate the individual interactions between all of them... (Minecraft, for the record, doesn't. If you have a 3x3 block it treats it graphically as a 3x3 block, not 9 individual blocks. Anyone who's seen inside the map when it bugs out can see what I'm talking about)

 

That'd definitely count as wasteful.

So all the effects in the image that you showed was wasteful as well? Seems like there's alot of things in the background there too. Also, if you watched the video I posted to the end, you'd notice that the map that SlamaCow used, is fairly large in scale. Not wasteful. The same process was used to put that world into the animation program. We use mineways which takes a section of the map that we select and place it into the animation program. That world size he used, his just as big as the one's we use. If not bigger. Plus how would you know it's wasteful? Have you tried this yourself? Are you an expert in animation? Do you have a degree in animation? Please answer these for me. I don't have a degree, but I do have knowledge of PC and how they function as well as practice and knowledge of the program Cinema 4D.

[Backup Lago 320]

If the computer has to carry out calculations that have no effect on the final product then those calcuations are wasted: that is wasteful. If the computer has to carry out a huge amount of calculations when it could get the same result in far fewer, then it is inefficient. If you are putting a lot of system resources into rendering something that never shows up in the image, then that is wasteful. Have you ever gotten outside of a game's map? Notice how there are no textures on the outside of the map objects (or inside) or collision properties? That's because they're not needed: it'd eat up resources for no gain. The inside of an MC model's head is textured. That's a waste, it's just such a small one that they never bothered to fix it.

 

You seem to have gotten the wrong end of the stick here, so I'll phrase it differently.

 

Is the static map a thousand or so individual blocks or is it a single model?

[ℤ∃ʁ∅ 823]

If the computer has to carry out calculations that have no effect on the final product then those calcuations are wasted: that is wasteful. If the computer has to carry out a huge amount of calculations when it could get the same result in far fewer, then it is inefficient. If you are putting a lot of system resources into rendering something that never shows up in the image, then that is wasteful.

You seem to have gotten the wrong end of the stick here, so I'll phrase it differently.

Is the static map a thousand or so individual blocks or is it a single model?

Gonna answer that last question easily.

When you are about to convert the area of that map you have selected on mine ways you are given an option to either have it all as one model or individual. I go with one model because I don't want to accidental mess up my scenery. If I wanted to add more rigs and models without messing it up I just throw them in there. Nothing hard besides deciding where to place if and what to place.

[Backup Lago 320]

When you are about to convert the area of that map you have selected on mine ways you are given an option to either have it all as one model or individual.

 

If having it as one model overtaxes your computer individual would totally obliterate it. Imagine having a thousand MC characters on screen at once and calculating lighting for them individually. That's what it would be like. It's rendering the whole block, the lighting effects of each block on each other. If you'd been importing it as individual blocks rather than one model that would have been my guess as to the cause of the massively taxing rendering.

 

I'm starting to think the inefficiency is probably based around this map importer.

 

There's an easy way to test this. Make a very simple animation. Get someone to render it with the map, and then to render it using a one block thick flat map. If the spike in time is orders of magnitude over what it was without the LotC map, then it's the map eating up the processing power.

 

If you're importing the LotC map then it sounds to me like your animations are rendering things that aren't ever in the shot. How deep is the map you're importing? Are you importing the underground? Are you importing caverns? Are there valleys that are out of shot? Hidden chests or holes in the ground players have dug, basements in houses, mined out trees?

 

If I'm right then you could vastly improve the speed of your rendering by being brutally efficient with your maps. Anything that's not in shot, don't import. Maybe even make each sequence individually with its own map. The one where you go into Knox's house in the video below, for example, you don't need any trees hidden behind the house, you don't need anything that's underground, you don't need anything that's behind you in the first frame, you don't need anything underground other than the surface block, you don't need anything that's blocked out by the house at all times, and if the house has a basement you don't need that.

 

To take the Knoxmas one, it's rendering the whole MC map for each shot, or at least everything in its cone of vision (which is still at lot). It doesn't know what's hidden and out of shot, what's a cave or completely hidden, it calculates the lighting et al and renders it anyway.

 

https://www.youtube.com/watch?v=eiiS0Uo6Qqo#t=31

 

You may be able to cut down rendering time significantly by not importing a single block more than necessary into your map model.

[LeonXCG 40]

If the computer has to carry out calculations that have no effect on the final product then those calcuations are wasted: that is wasteful. If the computer has to carry out a huge amount of calculations when it could get the same result in far fewer, then it is inefficient. If you are putting a lot of system resources into rendering something that never shows up in the image, then that is wasteful.

 

You seem to have gotten the wrong end of the stick here, so I'll phrase it differently.

 

Is the static map a thousand or so individual blocks or is it a single model?

Once again, it is not wasteful. The program only applies shading the the area that the camera is looking at, the rest of the map is ignored. Underground? The program can't see it to apply the shadows and effects. Ignored. The area behind the camera? The program can't see it and doesn't apply the shading to it. Ignored. Calclulations are placed where they need to be, but you're making everything from scratch, which is the part you seem to be neglecting. You also neglected to answer my questions. I'll just accept all of those as a no from you. Which leads to my final question of why are you bothering to call bullshit on these renders taking so long if you've never done the process yourself?

 

Also, take a look at these Images.

 

 

The one's from MC, yes have more effects, but these are already set into the code of the game through a mod. They're preset and left for the GPU to handle accordingly. The one's from C4D are not premade, they have to be applied. The program makes it from scratch and applies it to the area that the camera is showing. Any other artifact is ignored and left alone. Therefore resources are not wasted, they are only applied to the area that is show in the program, which still take a lot to make.

[Backup Lago 320]

Any other artifact is ignored and left alone. Therefore resources are not wasted, they are only applied to the area that is show in the program, which still take a lot to make.

 

So if you put a wall in front of something it doesn't run any processing on what's behind the wall? What if something behind the wall casts a shadow that would be visible?

 

Which leads to my final question of why are you bothering to call bullshit on these renders taking so long if you've never done the process yourself?

 

I explicitly said I was not calling bullshit. I'm saying if it's taking a week to render 15 seconds of minecraft animation then you're either rendering in throughly unnecessary quality or wasting resources somewhere.

I'm saying I don't get how A: if this is so taxing to do, how MC, a not that well coded java game, does it so fast relative to a professional animation suite, and B: how, if Minecraft can do Minecraft images of the same quality 40500 times faster why nobody's made an animation suite for Minecraft in the Minecraft engine. It can be done, Valve has one that runs in the Source engine.

My goal here is to either understand why it's taking so long or find the inefficiency or waste in the system so that it doesn't.

 

They're preset and left for the GPU to handle accordingly. The one's from C4D are not premade, they have to be applied. The program makes it from scratch and applies it to the area that the camera is showing.

And if you can just premake them and apply them to every situation in your use of minecraft what's the advantage of making them from scratch every time in a Minecraft context?

 

What's the absolute fastest C4D can pull off? Rendering vanilla unshaded Minecraft?

[Jistuma 1996]

Once again, it is not wasteful. The program only applies shading the the area that the camera is looking at, the rest of the map is ignored. Underground? The program can't see it to apply the shadows and effects. Ignored. The area behind the camera? The program can't see it and doesn't apply the shading to it. Ignored. Calclulations are placed where they need to be, but you're making everything from scratch, which is the part you seem to be neglecting. You also neglected to answer my questions. I'll just accept all of those as a no from you. Which leads to my final question of why are you bothering to call bullshit on these renders taking so long if you've never done the process yourself?

 

Also, take a look at these Images.

 

The one's from MC, yes have more effects, but these are already set into the code of the game through a mod. They're preset and left for the GPU to handle accordingly. The one's from C4D are not premade, they have to be applied. The program makes it from scratch and applies it to the area that the camera is showing. Any other artifact is ignored and left alone. Therefore resources are not wasted, they are only applied to the area that is show in the program, which still take a lot to make.

I'm... I'm sorry, I even had this discussion with Zero before but... Why do you need C4D? Isn't that Super HD? Why... why would you even need HD? And furthermore, if it takes that long to render with the program, why don't you use an image manipulating program like photoshop instead, making only the MC figure in one image, and taking a screenshot with normal MC, then placing them together?

[LeonXCG 40]

 

And if you can just premake them what's the advantage of making them from scratch every time in a Minecraft context?

 

What's the absolute fastest C4D can pull off? Rendering vanilla unshaded Minecraft

Rendering an unshaded version of it is faster, there is no lighting effects or shadows to be applied. he background (sky) can be set to a flat blue and not light source has to be made. It will render faster that way, but you don't get the lighting effects or shadaws.

 

 

I'm saying if it's taking a week to render 15 seconds of minecraft animation then you're either rendering in throughly unnecessary quality or wasting resources somewhere.

Like I said, resources are not being wasted. 15 seconds can take up to a couple or few hours. A week is for something that exceeds one minute. Also counting the time it takes to setup and animate.

[Backup Lago 320]

Also counting the time it takes to setup and animate.

For the third time, we are not counting this. If it was the human element taking a week then I would not be questioning it.

 

Why do you need C4D? Isn't that Super HD?

 

That's the other theory. That you're running C4D to its maximum on Minecraft blocks. Is the benefit truly so great as to warrant such superlative render times (and the death of Zer0's laptop)?

 

Rendering an unshaded version of it is faster, there is no lighting effects or shadows to be applied. he background (sky) can be set to a flat blue and not light source has to be made. It will render faster that way, but you don't get the lighting effects or shadaws.

 

How fast is that?

 

15 seconds can take up to a couple or few hours.

Says a week in the OP.

 

Does C4D give the the ability to play it in realtime before you render it?

[LeonXCG 40]

I'm... I'm sorry, I even had this discussion with Zero before but... Why do you need C4D? Isn't that Super HD? Why... why would you even need HD? And furthermore, if it takes that long to render with the program, why don't you use an image manipulating program like photoshop instead, making only the MC figure in one image, and taking a screenshot with normal MC, then placing them together?

You don't need HD, it's just a preferred option for most people to watch it in. 720p is clear enough, so is 1080p. 4K HD is too much to do for a video, but makes for a nice still image. And things look better if you do it straight from Cinema 4D instead of place an MC shot in. Looks better that way and in terms of using editing softwares, it takes less. But the time is higher, but at that higher time, you get a better quality image. Plus added time IF you want to add something through Photoshop or After Effects. For render's that are made into videos, you don't want to go super HD because it does take more time, so you do it in either 720p or 1080p. If you do it at 720p and put it through a program such as Sony Vegas, you can upscale it to 1080p, giving it a little bit better quality. Most animators render and 720p to 1080p.