New deferred rendering demo
- Thread starter Humus
- Start date
I get an error upon application launch - dxgi.dll not found.
Or if I read a little closer I'd realize this is a DX10 demo
Or if I read a little closer I'd realize this is a DX10 demo
Awesome! I did a deferred shading demo with hundreds of Christmas lights a few years ago and I was nothing but impressed with the technique. [Edit: seems I still have some images online.] With G80/R600-class hardware now, I'm pretty sure I'd use it for any "real game/app" moving forward
Great demo of the technique: runs super-fast and I like your GS idea...
Great demo of the technique: runs super-fast and I like your GS idea...
Are you still at AMD, or are you somewhere new entirely?Fine so far. Always nice to try something new.
Very nice! Runs beautifully on my 8800 GTS. DR has been my topic of interest for the past few months or so, and I always like seeing new demos and research. I really like how you used the geometry shader, I may have to shamelessly *borrow* that when I eventually upgrade my renderer to DX10.
Betanumerical
Veteran
I get a can't find server error at the address given to download the demo, anyone wanna host it for me?.
Dont worry it works now.
Dont worry it works now.
Last edited by a moderator:
Did you see Independance Day? Humus demos upload the ebil payload to all Nvidia cards.
But he has to pay a royalty to Apple each time for not using a Mac to do it with.What's a "2900guy" doing with a GTX anyway?
More seriously, my GTX reacts just fine. The only thing than can make a whine on a graphics card is the fan.
I started at Avalanche Studios in the beginning of September.
Actually, sometimes the electronics can whine by itself. I don't know why, but I've had that happen to me too. Even my Matrox G400 back in the day that didn't have any fan occasionally made some noise. When things like that happens the pitch tends to follow the framerate. Some old CRT monitors are also known to whine at certain resolutions and refreshrates.
Oh, right I remember you mentioning that now that I hear the name again
Not necessarily.
Almost all contemporary electronics uses switched voltage regulators. They are very compact and very efficient in bringing down a high voltage (say, 12V) to 1.2 or so volt. In a different configuration, they can also do the opposite.
See here for a fairly standard configuration.
Anyway, switched regulators will charge a discharge coils and capacitors at a very high speed.
coils + current -> magnetic field -> vibrating inductor wires -> buzzing sound, which will vary depending on the amount of current.
Those inductors can actually quite small. (Maybe coil is the wrong word for this?) Not entirely sure about this, but I believe the size can be made smaller when the switching frequency is higher?I thought they eliminated the coils & now use digital whatever they are which shouldn't vibrate
One way or another, you need to mechanism to store and release electric charge. For a switched regulator, a capacitor alone is not going to do it.
Why? Surely for decreasing the voltage, all one needs to do is square wave chop the input supply (at a suitably high frequency) and put in adequate low pass filtering on the output. A capacitor will surely do that job for you.
When increasing the voltage an inductor would be handy but, IIRC not absolutely necessary.
Thanks! The particle stream looks a bit chopped up though in that video, but I suppose that is from the video capture causing unstable framerates.
Yeah if this was taken in fraps you really want to leave "sync" enabled while capturing or else you can easily get the "propellor rotating backwards" effect from sampling.