Jump to content

Is using Houdini's viewport anti-aliasing free?


magneto

Recommended Posts

Hi,

 

I normally set it off, so I get sharp lines but for curved surfaces, it looks really nice, especially with 16x. Is this an expensive operation? I remember reading that GFX cards do it almost freely?

 

I also didn't notice any difference between 16x vs 32x. Would 32x be more useful for 4K displays?

 

 

Thanks :)

Link to comment
Share on other sites

Its certainly not free - try HQ lighting and increasing the AA settings. You'll see the fps drop, at least on a Gtx980, but it's absolutely fine most of the time to leave it on.

 

Your card caps at 16 most likely so 32 looks the same -btw you can MMB on that slider and get 65,000x+ AA samples!

Link to comment
Share on other sites

Thanks Marty. I noticed a difference when I set to 32x, tiny but still. Your card doesn't crash with 65k samples? I don't know how that would be possible. I imagine the excess is clamped? Or maybe Mark left the upper value unlocked to future proof Houdini to year 2100 and beyond :)

Link to comment
Share on other sites

I can't find it now but there is quite a bit of research into faster AA, especially in games. I'm wondering whether we can use that tech in DCC apps.

 

Anyways the Gtx 980 is almost a year old tech now so starting to look at the next lot of cards, Pascal might be good but I'm thinking the performance boost is mainly touted for Deep-learning, lower bit depth stuff, though the 3d memory that AMD or Nvidia have will be good for sims I'm thinking. Vulcan too, NextGL, is probably going to provide us with a lot of boost!

Link to comment
Share on other sites

Up to 4x AA is nearly free in Normal Quality lighting mode. I find 16x AA to be a good "high quality" mode; beyond that the quality improvement is hardly noticeable. 16x AA, at least on Nvidia, is 4 color samples (fragment shader run) and 12 depth-only samples (rasterizer output only)

 

High Quality lighting is a different story. It is clamped at 8 samples, because the data per sample is significant. It runs into a VRAM bottleneck fairly quickly, simply due to all the deep pixel lookups.

 

And no, you don't get 65K samples if you use the XCF slider :) It gets clamped at whatever the maximum samples is for a 2D MS texture, which is reflected by the range of the AA slider.

  • Like 1
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...