 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 18/08/2014 09:19 AM, Stephen wrote:
> On 18/08/2014 07:58, scott wrote:
>> The balls moving like that made me remember back to those computer demos
>> in the early 90s, they always seemed to have at least one section with
>> 10 or 20 balls moving about in interesting patterns. Of course back then
>> the balls were usually just 2D sprites and the screen resolution was
>> 320x200 or something ;-)
>
> That is the one. I am sure Andrew posted an animation of that years ago
> as a homage.
About 2 years ago, in fact. ;-)
I first saw this trick on the SAM Coupe (no, I cannot find any mention
of it on Google). I and my siblings were astounded that the computer
could animate millions of balls like that. (In the demo, it runs as long
as you leave it, with a ball counter in the corner constantly counting
up. When you press a key, it switches to the next pattern.)
The "trick" is that it's not actually sprites at all. I saw a demo like
this on the Amiga, using Bliz Basic. And it turns out, all it does is
this: There are 16 framebuffers. It draws just ONE ball onto the next
framebuffer, and then switches that to be the current one. That's all it
does. It's only rendering 1 ball per frame, but because each framebuffer
still has all the previously rendered balls on it, and there are 16
framebuffers constantly switching, it LOOKS LIKE the computer is
actually animating millions of balls, all at once!
I, of course, REALLY AM animating several hundred spheres, in actual 3D,
with full global illumination. At about one frame per hour. :-P
Makes you think, though... Remember that browser-based GPU unbiased
renderer demo a while back? If I could somehow utilise that, the entire
animation could be done in mere *hours*...!
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 18-8-2014 20:24, Orchid Win7 v1 wrote:
> Makes you think, though... Remember that browser-based GPU unbiased
> renderer demo a while back? If I could somehow utilise that, the entire
> animation could be done in mere *hours*...!
but where would be the fun in that?
--
Everytime the IT department forbids something that a researcher deems
necessary for her work there will be another hole in the firewall.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am Sat, 16 Aug 2014 18:17:45 +0100 schrieb Orchid Win7 v1:
> On 28/07/2014 05:20 PM, scott wrote:
>> And are you planning to make the animation public?
>
> https://www.youtube.com/watch?v=8s7I80H96CM
>
> Worth it, eh?
Looks a bit like the old school demos (not that they were that fancy with
water, but some had shiny spheres with shadows and some were flying over
wobbly surface (albeit in a different scene - not at the same time)).
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am Mon, 18 Aug 2014 07:58:53 +0100 schrieb scott:
>>> And are you planning to make the animation public?
>>
>> https://www.youtube.com/watch?v=8s7I80H96CM
>>
>> Worth it, eh?
>
> Nice water surface effect! The balls moving like that made me remember
> back to those computer demos in the early 90s, they always seemed to
> have at least one section with 10 or 20 balls moving about in
> interesting patterns. Of course back then the balls were usually just 2D
> sprites and the screen resolution was 320x200 or something ;-)
Oh... seems I was not the only one with that association :)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> Makes you think, though... Remember that browser-based GPU unbiased
> renderer demo a while back? If I could somehow utilise that, the entire
> animation could be done in mere *hours*...!
Plenty of examples on shadertoy.com of real-time GI in the browser ;-)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 8/19/2014 1:27 AM, scott wrote:
>> Makes you think, though... Remember that browser-based GPU unbiased
>> renderer demo a while back? If I could somehow utilise that, the entire
>> animation could be done in mere *hours*...!
>
> Plenty of examples on shadertoy.com of real-time GI in the browser ;-)
Uh.. Just don't expect Firefox to hand that site well. Something about
the way Firefox threads its content causes it to slog to a total crawl,
until all the shaders fully load. They work, they just.. lag the shit
out of it, while loading up. I presume this is still true in the latest
release, since I haven't been stupid enough to try going there recently. lol
--
Commander Vimes: "You take a bunch of people who don't seem any
different from you and me, but when you add them all together you get
this sort of huge raving maniac with national borders and an anthem."
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
>> Plenty of examples on shadertoy.com of real-time GI in the browser ;-)
>
> Uh.. Just don't expect Firefox to hand that site well. Something about
> the way Firefox threads its content causes it to slog to a total crawl,
> until all the shaders fully load. They work, they just.. lag the shit
> out of it, while loading up. I presume this is still true in the latest
> release, since I haven't been stupid enough to try going there recently.
> lol
It's similar in Chrome here, the first time I open the main page it
takes about 4-5 seconds to fully load, and if I go "back" to the main
page after that it still takes a second or two to populate all the
shader previews.
But the most impressive thing is that while watching a shader run, you
can modify the code on the right and hit play to run it.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 21/08/2014 01:12 AM, Patrick Elliott wrote:
> Uh.. Just don't expect Firefox to hand that site well.
Does anybody remember when Firefox was the browser of choice, because it
was so much faster than everything else? These days, Firefox seems to
take *forever* to do anything...
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 21-8-2014 9:43, Orchid Win7 v1 wrote:
> On 21/08/2014 01:12 AM, Patrick Elliott wrote:
>> Uh.. Just don't expect Firefox to hand that site well.
>
> Does anybody remember when Firefox was the browser of choice, because it
> was so much faster than everything else? These days, Firefox seems to
> take *forever* to do anything...
What is /forever/ though? from above I understand 4-5 seconds to be
/unacceptable/ ;-) To tell the truth, I have no complains about firefox.
Thomas
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 8/21/2014 12:08 AM, scott wrote:
>>> Plenty of examples on shadertoy.com of real-time GI in the browser ;-)
>>
>> Uh.. Just don't expect Firefox to hand that site well. Something about
>> the way Firefox threads its content causes it to slog to a total crawl,
>> until all the shaders fully load. They work, they just.. lag the shit
>> out of it, while loading up. I presume this is still true in the latest
>> release, since I haven't been stupid enough to try going there recently.
>> lol
>
> It's similar in Chrome here, the first time I open the main page it
> takes about 4-5 seconds to fully load, and if I go "back" to the main
> page after that it still takes a second or two to populate all the
> shader previews.
>
> But the most impressive thing is that while watching a shader run, you
> can modify the code on the right and hit play to run it.
>
Umm. No, when I try it in Firefox its more like 4-5 minutes. There is a
lot trying to load through that page, and.. Chrome takes a bit of time
to do it, but even loading just "one" shader, on its own page, in
Firefox lags like hell. Figure.. 20-40 second "per" shader, and the main
page has like 12 on it (just checked), so, at minimum it ends up taking
2 minutes minutes, or worse. But, no Chrome handles the page **way**
better than Firefox does.
--
Commander Vimes: "You take a bunch of people who don't seem any
different from you and me, but when you add them all together you get
this sort of huge raving maniac with national borders and an anthem."
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
