Bad quality of antialiasing in Rhino 6

Hi Steve,

The TestLineSmoothing makes is singnificantly worse over here.
I, too, would love to see better AA in viewport like V5 had, so far V6 is still behind in that aspect.

–jarek

Another observation: with GPU Tesselation OFF, AA looks much better (combined with WireThickness @ 0.6 it is as good as V5).

Did you adjust the wire thickness scale? That will have an effect on TestLineSmoothing.

I’m not going to be able to further improve this look until I can find a computer/monitor combination that looks bad.

Yes, tried both with 0.6 and 1 wire thickness… No luck with any of these settings.

Hi @stevebaer, I wanted to let you know that over here the AA got better with the recent V6 builds (I got yesterday’s one). Not using the TestLineSmoothing, that actually makes things worse, but in general it improved. I did not change anything in the system or video drivers, so I assume it’s something you did on your end.

–jarek

That’s great to hear. I have been trying my hardest to get the pixels to light up just right, but when it comes to single pixel wide wires it is really hard. Hopefully it will look good for others (I know there are still complaints being made about the latest work on discourse).

Hi @stevebaer,

Big HELLO from Germany :slight_smile:

Is there any chance to get the antialiasing of Rhino 6 to the quality of Rhino 5?
It is a kind of disappointing buying a newer version but having worse antialiasing and therefor still working with Rhino 5.

Best regards and have a nice day.

Holger

I’m not going to promise this as I haven’t been able to satisfy this issue for everyone yet. Technically, we can’t use the same technique as we did in Rhino 5 as we have moved beyond legacy fixed function OpenGL and are now drawing everything with a shader based approach. Using the newer technology has improved display in many areas, but the AA of wires is something that we are still trying to improve.

Is this because legacy OpenGL AA was a “black box” so that you need to second guess and reverse engineer the shader code for it? i.e: no open source to start from?

Ive noticed recently how rubbish the AA is to in V6. I take it nothing has been done to rectify this problem since last year?

Okay, please let me dig up this 1 year old thread. :grinning:

Technically, we can’t use the same technique as we did in Rhino 5 as we have moved beyond legacy fixed function OpenGL and are now drawing everything with a shader based approach.

Could you explain this a little bit further? Are there absolutely no ‘legacy’ OpenGL calls (which invoke the fixed function pipeline) in the Rhino 6 codebase anymore?

Wow, I didn’t even try the fixed function OpenGL pipeline. Rhino is old and I am young!

@felipegutierrezduque What do you mean? You didn’t try Rhino V5?

@stevebaer what I found out is that if the display mode is adjusting the output gamma to 2.2, the AA looks very similar to v5. Is this a coincidence or…?
please download and inspect at 1:1 scale

I didnt create OpenGL apps with the fixed pipeline.