Forgot your password?

Comment: Re:Display server (Score 1) 826

by Kremmy (#47751475) Attached to: Choose Your Side On the Linux Divide people themselves admit Wayland is better because those people are Wayland developers.. That's not an unbiased source by any means. It's also a source that one would expect knows what they're doing to some degree, but that's always debatable.
Personally, I have a major trust issue with Wayland. Also with SystemD. I feel my reasons are very sound, they include the fact that both of these projects have been rabidly pushed since their project inceptions. The trust issue is that I've been told these are the next big new great thing since there wasn't a line of code written. I've been told that they solve all of my problems while providing none of the solutions.
I've been computing most of my life and these projects have some serious delusions of grandeur that have frankly made me want to keep them off of my systems at all costs. Don't break the system because you think you're doing it better, do it better first.

Comment: Re:Well, (Score 1) 739

by Kremmy (#47545599) Attached to: Linus Torvalds: "GCC 4.9.0 Seems To Be Terminally Broken"
Linus has been a user of these tools for decades now. He's one of a select few who have the capability, experience, and understanding necessary to even approach the depth of the bug that they're talking about. He's one of the few who have been heavily relying on them for his entire technology career.
What I'm saying is that Linus knows his tools. He knows when his tools are need in of repair. He knows what to expect from them and by this point should have a solid understanding of their quirks.
Sometimes tools don't work how they're supposed to.
I'm inclined to believe the master woodworker when he tells me his blade is dull and needs to be sharpened.
I'm inclined to believe the master programmer when he tells me the new version of his compiler is generating incorrect code.
I'm probably a below average programmer myself, but I've hit enough depth to understand some of the context of this issue. I read through a bit and see them doing the footwork necessary to deal with the issue. That's something I don't see people doing when they're blowing smoke and being bitchy.
Linus has a bit of an attitude problem. But so does almost everyone else in a position like his, where he is the project leader for something incredibly complex. Like the local event equipment guy who can be a real ass when he's in action, but knows exactly what he is doing, how to do it, why he does it how he does it, how to explain it to his workers, and gets it done on schedule, because he's been doing it for a very long time and lives and breathes the job.

This is all beside the point, which is that you could only write your post because I previously slept with your mother.

Comment: Re:Reconcile these two sentences please. (Score 1) 502

Check out the issues that Titanfall has with audio comparing the Xbone SPU with PCs that don't have a fancy SPU. It's probably the best example I can think of right now, but there are plenty of others regarding 3D audio and onboard sound. Onboard sound quality reached par with CD quality audio with multiple channels, but there's a lot more available. The truth is that the base quality of hardware reached a point where things are 'good enough', but the tradeoffs are still very noticeable when running applications that were designed to use the advanced functionality.

Comment: Re: Is it still braindeadly single-threaded? (Score 1) 138

by Kremmy (#47425941) Attached to: Dwarf Fortress Gets Biggest Update In Years
“Traditional single-processor pathnding strategies, such as A* and its derivatives, have been long praised for their exibility. We implemented several parallel versions of such algorithms to analyze their intrinsic behavior, concluding that they have a large overhead, yield far from optimal paths, do not scale up to many cores or are cache unfriendly. In this article, we propose Parallel Ripple Search, a novel parallel pathnding algorithm that largely solves these limitations.”
I disagree with your basic assumptions.

Comment: Re: Is it still braindeadly single-threaded? (Score 1) 138

by Kremmy (#47419071) Attached to: Dwarf Fortress Gets Biggest Update In Years
I expect that the logic required to reasonably split up the pathfinding is going to easily reverse any potential savings. There is a completely disregard for the fact that in order to split the processing up, one must take the time to organize the processes such that they can be split up.

It's a little bit like rocket science. How each additional pound of fuel provides so much thrust, but the additional weight but also be accounted for. 'parallel the pathfinding' is adding fuel without accounting for the weight of the fuel.

OS/2 must die!