About 11 years ago, after having written a tiny ray tracer from scratch using Java, I taught myself some ray tracing with POV-Ray. My goal was to learn a few POV-Ray features each day over 25 days and render some interesting scenes that exercise those features.
I began with simple spheres and cubes and gradually progressed to more intricate shapes and textures. Here are the results:
Cool indeed. A spin on it (also using POV-Ray), but using a photo (texture on an object) and animated (rotating that object, the texture, or the mirrors -- I forgot) in [0] and [1].
I still remember when I used to leave my 386 sx 25Mhz running all night to render very simple scenes with POV-Ray (somehow sleeping through the loud fan noise!).
And the extreme excitement the day I upgraded to a 486 dx 33/66 Mhz which, thanks to the math co-processor, rendered those same scenes in (10s of) minutes instead!!
Same here. But I took it pretty far. Wrote a primitive keyframe animation program using pascal that would compare two .pov files (generated by Midnight Modeler) and output a file with the differences replaced by a clock variable you would pass when rendering for animation. It was fragile and required you to apply all the transforms you would need to the scene before you started. The order the transformations were applied could also produce unexpected results. It worked best for camera movements.
I used it for a senior year high school project in 1998 to make an animation (that started off as a super hero story and was truncated into a funny commercial when I ran out of time.) I definitely fondly remember the feeling of waking up in the morning to see how the rendering had turned out, it also felt powerful to have my 486DX2 50 hard at work while I rested. Writing the keyframe program definitely felt good, first time I felt like I coded something useful. One frustrating aspect was that the computer could not smoothly animate the resulting videos except in lower resolutions, so the finished product changed resolution depending on the scene. It was eventually all put together in a vhs camcorder. https://youtu.be/80hp5YSp4Co?si=XQqXIdYtHssgoQz3
Are you me? :-D
Same story, had an AST 386 which I upgraded to a 486 DX33... I learned to program in the 90s by writing povray scene files, which taught me C style syntax and primed me to write actual C/C++/Java in the years that followed. I spent thousands of hours with povcad and povray on windows 3.1...
Maybe he'll get to use the "that would be impressive except if they had known what they were looking for,they would have seen it written on my dorm room window" quote later today.
I find it surprising that it would cause issues. There are as of this writing only 162 comments in the thread, which was posted 17 hours ago, and the (simple!) web page is still very slow to load. How much traffic slows down a static web page with a single image? Even if we stipulate a quite manageable 100 requests per second, that means 6 million people read this thread and decided to click that link. 6 million people and only 162 left a comment? Can that be right?
>Even if we stipulate a quite manageable 100 requests per second,
Depends on the situation, but 100 requests/seconds sound like a lot to me (depending on how heavy the processing is, of course). And every page visit generates 8 requests, so that's "just" 12 people visiting per second.
Also are you me? Except I had a 486sx and I tried in vain to persuade my parents to buy a maths co-processor for it. It would have probably saved them money in the long run from electricity bills.
Some of my first programming was writing QBasic programs to generate povray scene files.
I had a 486dx and it was such a huge upgrade over a 286. I ordered pov-ray from some shareware catalog since I didn't have internet access, and it arrived on 3.5" floppies.
What on earth this was me too!! I still remember leaving my 486 rendering all night after I had messed with the computer and disconnected the CPU fan. Several hours of sleep later there was a loud blaring alarm, because the CPU was about to overheat. It took like 4 hours to overheat! Can't remember what rendering tool it was though.
:-D I already knew (some) C, lots of Turbo Pascal and Basic by that time so I would generate povray scene files using a small C program I had developed which took various equations as starting point to plot spheres on the curves
Hah, same except my dad showed me how to write a simple search and replace using a DOS batch script to generate many files to then pass into the renderer.
I used it to do camera pans, lighting effects, etc.
What 386 owner hasn't rendered that sample POV Ray scene with a glass of wine? It made me feel that my beloved computer could do "professional" graphics.
I'm old [and stupid] enough to remember running a floating-point-only raytracer on a non-FP PC using a software-emulated math coprocessor. I think I measured output in *hours per frame*.
Pedantic nitpick: you are most likely talking about a different 'colour' of noise. White noise is really harsh, and probably not what your fan produces.
I've had 24x7 FIDO node in my room for years. I've silenced modem pulse dialing by replacing relay with expensive one. I've silenced CPU cooler by using 90mm 12v fans connected to 5v (yes, in 486DX era). But HDD was unbeatable.
On the other hand, if I wake up in the middle of the night due to HDD grinding, I was sure that I have new mail (echoareas) to read!
yes! I did the same thing, although I never got a 486 and jumped to a amd k6-2 with Blender when it came out. Never had the math co-processor either, although it would have been nice.
My roommate in 1992 would run this on his 386DX with the 387 math coprocessor. It would take literally days to run to create a small 640x480 image.
This was in the days of DOS where you could only run one program at a time. It would run all night and then in the morning he would stop it so he could use the computer for other things. But he had some kind of Targa .tga file utility to merge the files together.
Then he compiled povray for our Sun workstations and he would split up the rendering so that each machine would render 50 lines of the image and he could merge them together with that utility.
I remember how happy he was that he could render stuff 20 times faster.
10 years ago I implemented a brainfuck interpreter that had as output an animated povray scene description with a visualisation of the brainfuck abstract machine. https://www.youtube.com/watch?v=6PIZTFrkl0w
Neat video. With some subtext to explain what the different "subthings" are for in brainfuck to stage and then do become "multiply" would make it even more "bro!"ish :-)
I also spotted your openxcom play from 6 years ago and must confess that I am playing it these days!
I love POV-Ray. I learned about it in college when I was trying to find things to do on my gaming desktop that weren’t gaming. It’s so much fun to mess with and make different things with. There’s some super impressive examples and stitching the images together is a lot like magic.
It’s honestly really satisfying to use.
I imagine a lot of people can use it. I made dice with it already.
Many years ago I asked one of my role models how he had made some pre-rendered sprites for a game, and he told me it was with POV-Ray, but that he did not recommend it because it used its scripting language to define scenes, which he was concerned would be too complicated for someone who has not done 3D work before.
I have only done 3D work as a waxing and waning hobby, but then, and to this day, the POV-Ray scripting interface seems like one of the more natural ways to define a scene to me.
In 1991 or 1992 I used POV-Ray on my Atari ST to create some title screens for some home videos. Completely gratuitous marble text infront of a glass ball on top of water type of stuff, which took all night to render, but it was fun, and crucially free. For years I'd looked enviously at Cyber Studio for the Atari ST, with its StereoTek liquid crystal shutter 3D glasses add-on, but it was just too expensive for me at the time.
Then in 1996 or 1997 I thought it would be fun to use it in a professional context at the software company I worked at, making a 3D animated GIF version of one of the product logos which I put on the web site (FWIW it looks like the 3D non-animated version is still visible on the Internet Archive Way Back Machine at https://web.archive.org/web/19971211003918/http://www.sophos... 27 years later). Although no-one had asked for it, I was still in effect getting paid to do something I used to do for fun, which felt good.
We wrote some small tools with Turbo Pascal to generate number of scene files, then rendered these to create small animations. Having a camera move over a reflective chessboard was pretty amazing.
I really liked the scripting language for defining 3D scenes; I bet today you could have an interactive UI that shows the scene in real-time as you modify the script.
It is a joke or a riddle, depending on your point of view. I used to be a high school math/science teacher, so this fits in at the intersection of those. If we have national standards for physical things like length, or mass, why not a national standard for a mathematical constant? It's just a ring with a scale on it that measures the circumference. Every now and then, a technician checks the value and records what they see in the log.
Even my fellow teachers struggled to grasp the humor of the thing. I suppose I'm a little strange.
If you once enjoyed POV, check out OpenSCAD. It's quite similar in terms of CSG concepts and primitive animation capability, but more useful for getting actual mechanical design work done. https://openscad.org/
While POV-Ray was a cool project, let's not forget how far we have come with Blender and what a great success for the Free Software movement it represents.
What I'm kind of missing with Blender and the likes is the procedural 'magic' and surprise of POV-Ray. How just typing a few lines of 'code' into a text file could produce wondrous and mind blowing images, and then just changing a couple of numbers would give you an entirely different and unexpected result. Blender makes it infinitely easier to get exactly the result you want, but sometimes that not what you actually want.
I was one of the original donors in June 2202, iirc, for Ton to buy Blender and start the Blender Foundation. I used birthday money and sent I think, $50, that's why I remember it to this day. I still use Blender, but I was using it back then to generate 3D reliefs to carve on my self-built 4x8 foot CNC router table. I wrote a script based on a shape-from-shading algorithm originally developed by NASA to get more out of all of the single-lens B&W satellite photos they had. My company was The Wooden Image. To the other poster: you could do procedural stuff in Blender early on using Python scripting, although I was using POV-Ray very early on.
POV-Ray was my first introduction to HPC clusters. In the early aughts a few of us in college participated in a summer program at Wright Patterson Air-force Base to build a MPIPOV [1] cluster out of 10 old Sun SPARCStation 20s with "Happy Meal" NICs.
We documented the process of installing Linux (Debian), configuring the network, compiling MPIPOV from source and clustering them together.
It was a thing of beauty to watch the rendering speeds increase and the blinkenlights putting on a show in that lab when we were done.
If I remember correctly they planned to take what we documented and use for a much larger cluster they were building, but never found out the specifics.
I think it was in 1994 that I posted to usenet (lost in time), offering 'render time' on my 286 for POV-Ray. They were such amazing times, yet compared to today seem so innocent.
POV-Ray was my main hobby at the time, along with the community of the Raytech BBS in the UK, and defined so much of my interests going forward, through many 3D modelling and rendering packages.
Such a huge part of my younger years, and one of the biggest influences on my life overall.
A lot of people do similar with 3D printing now (esp. over COVID to print PPE), so there is a glimmer of this time :)
Time to offer a real-world POV-Ray printing service? ;)
Having to write one for a class in college made me a master of pointers and C. Arrays within arrays rendering to Z buffers and crude matrix operators. Now all built into a nice lib and 3 calls away.
Povray clearly holds a special place for many of us. For me it was the first use (via slow dial up) to access HENSA (higher ed national software archive) in the uk for public domain software, and this by itself was mind blowing, in between dealing with my mother keeping picking the other phone up and dropping the connection.
The batch processing tools of that era for graphics were qualitatively different from the real time interactive editors. There is something to be said for the imperfect serendipity that would result. The closest thing these days, oddly, is ML training, where part of the appeal is the sense the computer is working super hard for you. Were you to recreate the same concepts on modern hardware you would do something like SDF CSG on GPUs, but it would be surprisingly interactive and so missing this surprise element.
As many here, I spent too much time with POVRay in my youth. About 8 years ago I decided to try an idea that I had in my mind and decided to install it and relearn it. I wanted to try this fractal idea made of toruses.
I cut my teeth on POV-Ray. It hasn't been updated in some time but being able to code scenes has a lot of advantages.
For product design I use OpenSCAD. Maybe POV-Ray made this style of design popular.
A couple weeks ago I decided to use Blender for some 3D print modeling. I needed to make a wheel cap for my son, who had broken his. Although I love Blender I was disappointed when I wanted to update some things later in the design phase. Blender has some destructive editing that caused me grief. Perhaps you could avoid this with some additional mastery of the tool.
It's been a long time since I have looked at POV ray and my knowledge of it is woefully out of date, but does POV ray currently make use of hardware acceleration or is it still CPU bound?
By their FAQ: CPU, FPU, Bus speed, and Memory bound - in decreasing order of relevance.
Will POV-Ray render faster if I buy the latest and fastest 3D videocard?
3D-cards are not designed for raytracing. They read polygon meshes and then scanline-render them. Scanline rendering has very little, if anything, to do with raytracing. 3D-cards can't calculate typical features of raytracing as reflections etc. The algorithms used in 3D-cards have nothing to do with raytracing.
POV-Ray 3.7.0 (released 6 November 2013) is the current official version for all platforms.
There are significant internal changes in this version due to the introduction of SMP support.
That's a rather old statement. Nvidia and Apple GPUs have hardware-accelerated raytracing now. But even without specific raytracing features, lots of renderers use GPU compute for some of the raytracing workflow.
Things have changed a bit since that was written. Of course nowadays Graphics cards do permit somewhat arbitrary code to run, and can also be used by a ray tracing engine. Of course said engine has to be written to utilize them.
For instance, the Cycles [1] engine in Blender.
If you're into ray tracing as a hobby you have to play with it at least once! Cycles does probabilistic rendering and can handle tricky things like caustics.
POV-Ray itself dates back to the early 90s (and was based on code from the 80s). The FAQ in question was last updated 2013, and the paragraph about graphics cards is probably older than that.
To add some anecdote, I remember doing a short animation with a Star Wars A-Wing Fighter (the model for which I downloaded, not made myself). I added light sources in the engines for that "glow" effect.
Unfortunately, I messed up some of the geometry in the animation, so while the A-Wing was rolling to one side, the two lights were rolling to the other side ;)
Which I of course only found out after a day or so, as renders were so slow (and this was something the size of a poststamp...)
long gone unfortunately, though not that fancy, just made some text in moray and applied some materials, in a semi transparent box or sphere, render. The reaction was always wow, I always chuckled at the time.
Probably 25 or so years ago I created my first render ever, in POV-Ray of a LEGO character's head using a hypercube.
I only had a ruler to use for measurements of the head shape and the face and it took me quite a while, but it came out better than any other LEGO render I ever saw during that time. I was quite proud of it for such a simple thing.
I think I remember playing with POV-Ray (along with Yafaray) as alternative renderers for Blender back in the early 2000s, not long after Blender had been ported to OS X. At that point I had been making simple scenes in Blender for a while but had grown tired of the limitations of its internal renderer, most notably its inability to render caustics and bounced light, and so was trying my luck with other renderers.
Never did get that far with either. If I recall, the problem with POV-Ray was getting the Blender file translated correctly for POV-Ray to render right, and while Yafray didn’t have the translation issue it was too slow to practically use on my little 400Mhz iMac G3.
It never even crossed my mind back then to directly write code for POV-Ray. At that point, in my teenage mind 3D was something you did with GUI software packages like Strata 3D and Blender.
My first ever contact with programming back in 8th grade some 20 years ago. My then-teacher's website about PoV-Ray is still online, in all its late 90s goodness (in German): http://asti.vistecprivat.de/index.html
I remember playing around with that like... 20 to 25-ish years ago.
Back then only pre-rendered cut-scenes could come close to having graphics like that, and I dreamed of the day when games actually look like that realtime.
Kinda ironic that now that games actually do look like that (if not even better), I prefer to play retro pixel-art indie games, which actually put their focus on gameplay instead of graphics.
While I can't know what the parent-parent post meant, POV Ray is around for over 30 years and was popular in the early 2000's on the internet as well, and is fun to play around with, I remember people posting renderings on forums like gamedev.net etc...
I really appreciate seeing things that are named after the Internet as part of the exuberance of being able to do or share them on a worldwide network. Naming something after the Internet itself is probably most common in the 1990s.
The POV-ray syntax with its wild mix of curly and whitespace state description driven by a C-preprocessor lookalike successfully primed my mind for making the jump from Basic and Pascal to C (and eventually to other curly languages). Eternally grateful, also to the IRTC that made me spend enough time with the syntax to have a strong learning effect.
Depends what you mean? If you want a modern high performance ray tracer with support for all the latest hardware and implementing all the cutting edge ray tracing research, there is Embree and OptiX. They don't however come with their own scene description language, which is what made POV-Ray so popular in the first place.
Nothing, except it's thousands (literally) of times slower than alternatives because it uses primitives instead of meshes and therefore doesn't lend itself well to GPU parallelization
I did my modelling in radiance, an architectural lighting/illumination framework. it used a very "unixy" text file format to specify objects as functions in xyz space.
Overall I thin POV-Ray overtook it.
Going from 512x512 to 1024x1024 was a quadratic explosion in time to render in radiance. Never worked out why.
This is an amazing project and I'm glad it exists. My college capstone project was building a 3d scanner, and we were able to use POV-Ray to create a repeatable test environment for our algorithms. Wish I had an excuse to play around with it nowadays.
Wow, POV-Ray, what a blast from the past. I thought it might be fun to see how fast a scene renders today vs. back in the 90s with my Pentium-75, but apparently they've decided to integrate Boost so it takes forever to compile.
Edit: My old scripts are no longer compatible. :(
Edit2: The -MV option is your friend. :)
Yeah, modern machines are fast.
Trace Time: 0 hours 0 minutes 0 seconds (0.126 seconds)
I seem to recall this render taking a couple of minutes on my old machine.
I spent a lot of time with POV-Ray. Back in ~1992 I had the IBM C++ Compiler for OS/2 and spent a day or so tweaking the source to get it to compile. Fun times.
Honestly delighted that this still exists. Like others, I used to leave my computer running overnight to render. Dating myself, but it was a big deal in the BBS scene in the UK in the early nineties - so many files and techniques being swapped.
This is what I thought the internet wiuld become, back in the 1990s, just tons of projects like Povray. Not addeictive dark pattern trillionaires, see the recent Eli Gray article posted on HN how big tech enables link fraud for example. This was before FAANG existed as it does now. Apple was a struggling PC maker and Microsoft was the evil empire being challenged by a Finnish college student. Sun and SGI and DEC were big tech and they made you know… actual tech. It was this crazy dream that it would last like that forever, that the internet would just be one huge BBS of Povray style people, makers and users, who were interested in art and science above all else. Nobody was thinking you could contract third parties of poor people to screen out death videos on your BBS so they would have PTSD but you could make billions. Nobody was thinking to clickfarm children.
As I have aged (almost 50 now) I have realised that greed will turn any good thing dark without vigilance, and then who watches the watchers. Eternal spring perpetually. A culture which prioritises individual profits over the commons will destroy anything not regulated.
As I have said before on this forum, we need a way to price in the whole lifecycle of manufacturing through to waste and punishments for behaviour like Purdue and the opioid crisis should be the loss of all wealth generated by such dark behaviour.
of course this is also unworkable, but I am not personally certain what is workable without a secular moral revolution.
Or we could move away from pricing everything and the neoliberal obsession to turn everything into a market, and restructure our economy to provide everyone with their basic material needs, regardless of how much income they have.
even if we restructure everything to meet human basic needs, pricing will still be part of the equation. finite resources means value has to be attached to materials.
Value and price are not the same thing. Money is a form of access control which limits the availability of scarce resources to those people and organisations which have enough funds. It also collapses the value of all things to a single dimension. There are other ways to decide how much of something is produced and how it is distributed, for example participatory economics.
Money isn't just a form of rationing, it creates the scarcity it claims to solve.
A neoliberal economy wastes talent and skill in much the same way an ICE wastes most of the energy from the gas it burns. Vested interests clog up the engine and keep it from running cleanly and efficiently.
This doesn't just create pollution of all kinds - physical, social, political, and ecological - which makes the environment a very unpleasant space for most humans.
It also puts a hard cap on the maximum speed, which is nowhere close to what's possible.
pricing is the exercise of determining value, even in a non monetary economy (think barter or contribution) you still need to price the value of materials and time. obviously money creates secondary effects which are not related to value due to arbitrages and other effects, but I do not believe you can have finite resources without determining value, which is what I meant by pricing.
We still haven't found a better resource allocation model than pricing.
We have tried central planning, and it resulted in horrendous living standards (as compared to the western world), queues all-night-long that you had to wait in if you wanted to buy bread in the morning, "if you're not stealing from your employer, you're stealing from your family" being adopted as a common proverb, and the whole system basically running (for some definition of running) on bribes, favors and theft. Communism finally fell around '89 in most of Eastern Europe, and we're still recovering.
Perhaps you could solve some of these points with computer-aided optimization and dystopian AI-powered mass surveillance, but is that really what we want?
In my view, the problem isn't capitalism, the problem is the government trying to fix capitalism, but instead making it much harder for small competitors to emerge, effectively causing almost-government-mandated monopolies.
Think about what industries are complained about most in America, and how regulated those industries are. You can't just lay fiber, make medications or help patients without going through a regulatory minefield, mostly for good reasons, but this is why the big providers of these services aren't outcompeted by smaller ones. There's a reason why the mostly-unregulated big tech is considered to be one of the most trustworthy industries among most (non ideologically motivated) consumers, far surpassing any political party.
Capitalism is sometimes bad, central planning is worse, but heavily regulated capitalism is the worst of them all.
A key problem with unfettered capitalism is the tragedy of the commons. If left alone rogue/selfish actors will destroy that which belongs to all of us and is required to live (see nature). How do you propose to solve this without "benign" interference?
If your property generates air pollution, noise pollution, smells, unclean water, radio interference etc, you get taxed and/or have to offset the effects (e.g. by planting trees).
You can do this with very simple, straightforward regulation, in a way that is very easy to understand, doesn't require an army of lawyers to follow and doesn't advantage or disadvantage large companies.
The temptation of exceptions, exceptions to exceptions, and exceptions to exceptions to exceptions might be too much for governments to stomach, though.
Very well said. I miss the early days of the internet. In addition: Event though the net was not very safe at the time (no/poor encryption, no much monitoring by officials etc.) it kind of felt differently. Many users where at least aware of the dangers of speaking with strangers and downloading things but today it's somehow of more deceiving. It looks nice and shiny on the outside but full of traps. Maybe because the 90s web was less "shiny" on the outside, it was less deceiving?
Would you really go back to the internet as it was in the 90s? The current one may have problems (to say the least) but it also has miracles.
I’d go back for nostalgia, but not for practical purposes. Even in the 2000s it was much harder to find information. Wikipedia didn’t launch until 2001, and wasn’t useful till long after.
Usenet in the early 1990's is still superior to modern forums. Threads could go on for years, the newsreader programs automatically marked posts as read and would only show you new posts in the thread. Compared to today it is hard to find what is new and then discussions die after a day on places like Hacker News and Reddit after it is no longer a top post on the page or subreddit.
Sometimes I wonder, what is current / next thing that's like the PC / BBS / early internet scene of the 90s, with such a rich ecosystem of innovation, hobbyists, open source / shareware, where one or two people in a garage have as much of a chance of changing the direction as any entrenched company?
At least from the outside, the bitcoin scene of the early-mid 2010s looked like that - although there was plenty of dumb hype about the "product" itself, there was also of opportunity for innovation with mining, exchanges, and trading setups.
I too wish there was someone cutting their way through the Internet jungle to find the Scene, and report back to the rest of us. Pockets of vibrant, virtuous, inventive community about. They are, rightfully, protective of their hard-earned turf. We probably need not look far to find one nearby, just keep an open mind about the tech being used. It could be a WhatsApp group, an old-school bulletin board, a Mastodon hashtag. Your own genuine engagement and empathetic attention are the entry ticket. Go thrive!
I began with simple spheres and cubes and gradually progressed to more intricate shapes and textures. Here are the results:
https://github.com/susam/pov25
The source code is in the "src/" directory. The rendered images are included in the README (scroll down to see them). I hope you like them!