[|it|Nexu]

mitsubishi, on 08 August 2010 - 11:29 PM, said:

part of it can be done with /set
e.g.
/set q quit
/vstr q

/set srv connect x.x.x.x:port
/vstr srv

I already do that actually, but it's a hassle when like for connecting to let say same ip but different port or different pass...word. Need to rewrite that whole line....
The purpose would be to reduce the amount of typing... you're a linux user yourself, so you know what i mean

This post has been edited by Nexu: 09 August 2010 - 12:30 AM

[mitsubishi]

Ye, it's only partly there.

versus666, on 08 August 2010 - 08:10 PM, said:

BUT the game is like VERY delayed

Can you try the latest version?

The vars involved are hopefully less likely to overflow.
I wouldn't be surprised if it makes no change though in your case.

[versus666]

mitsubishi, on 08 August 2010 - 11:29 PM, said:

What is your processor?

hm, that's quite severe. There are reports that the windows version of the tools employed to support it may be erratic on multicores but nothing like that, the quirks should at most be like small 'fronts and backs' in time.

C2D 2.66GHz, so it's more than enough horsepower to run UrT at constant 125FPS (even thought it seems much more GPU dependent with bloom etc...). With or without bloom the MB is bugged, sadly.

(I'll add a screenshot if you want)

EDIT : Ok, I'll try with the newest version.

This post has been edited by versus666: 09 August 2010 - 12:49 AM

[mitsubishi]

oh the reply is for microGranularity, I'm not giving much hope in debugging motion blur considering there's not even support by ATI for the OpenGL 1.5 feature that base it.

This post has been edited by mitsubishi: 09 August 2010 - 12:53 AM

[|it|Nexu]

mitsubishi, on 09 August 2010 - 12:52 AM, said:

oh the reply is for microGranularity, I'm not giving much hope in debugging motion blur considering there's not even support by ATI for the OpenGL 1.5 feature that base it.

Did you ever found the source for q3mme part that did the motionblur?

[mitsubishi]

No, their forum reports implies he never released it and then he's gone.
--
The general idea here is most modern approaches are via GLSL or/and VBO (that 'bumpy' has support for) so ATI and others don't bother much for features that do similar things.

This post has been edited by mitsubishi: 09 August 2010 - 01:38 AM

[versus666]

Mitsubishi: The artefacts from MB are bigger, like having a monochromatic tetris on the screen ^^
MicroGranularity seems corrected with the new version, even if the FPSgraph show some 'waves' ( bigger and bigger yellow spikes, then lower and lower one) time to time. Bearable.
Well done ! ^^

[|it|Nexu]

mitsubishi, on 09 August 2010 - 01:36 AM, said:

No, their forum reports implies he never released it and then he's gone.

I found 2 version of the source on my PC. Uploaded them for you:
http://nexu.pwnz.org....33.4.1-src.zip
http://nexu.pwnz.org...1.52_source.zip

[mitsubishi]

versus666, on 09 August 2010 - 01:41 AM, said:

MicroGranularity seems corrected with the new version

Wow, really!? That's surprising. The variables involved are BIG and calculations with them need care to not create overflows. Hopefully a few corrections were enough.

Nexu, on 09 August 2010 - 01:43 AM, said:

http://nexu.pwnz.org...1.52_source.zip

thanks. I'll look into it.

[mitsubishi]

From a first look, apparently he takes screenshots and manipulates images. That's definitely different from doing it on OpenGL [and not possible to see effects in-game]. It also seems quite a maze to understand for me. I wonder if it's easy to just include his file (he has most of his stuff in the same file), that'd be cool. oh he has heavily modified the client all over the place; it'd be an undertaking.
And he doesn't use .avi support, the base source is older.
Not very encouraging to me. But it might be inspirational under right circumstances if a particular feature is needed.

This post has been edited by mitsubishi: 09 August 2010 - 02:24 AM