Linked by Thom Holwerda on Sun 27th Dec 2009 14:12 UTC
KDE Since the stream of news is still pretty much dry, I figured I'd throw in something I've been meaning to talk about for a while now, but really didn't dare to: KDE4's performance. Since experiences with KDE4 seem to widely differ between people, it might be a good idea if we, together, can find a common cause among those of us having problems.
Thread beginning with comment 401368
To read all comments associated with this story, please click here.
The problem is not in KDE
by darkcoder on Mon 28th Dec 2009 05:28 UTC
Member since:

I'm not blaming KDE SC, or Linux at all. If you see closely, video chip makers have been kind of lazy lately. And many problems have affected Windows too.

1. ATI big list of unsupported cards (both Linux and Windows). Anything 2 or 3 years or older is basically unsupported no matter the platform.

2. ATI driver quality. They even release a hotfix the same day they release the official driver to fix some (last minute????) bugs. To me looks like rushing the drivers to hit a release date.

3. Nvidia first v185 driver which due to an error prevent installation of 6000/7000 family GPU's (all platforms affected).

4. Intel drivers, which gave a lot of trouble, and not so good performance for a while.

5. Hardware features unsupported for months sometimes. All makers have been in this position a couple of times.

And who we have to blame... ourselves, the users.

As long we keep buying the latest video cards as soon they left the production lines, we will have the crappy performance, incomplete features, unstability bugs related to rushed production.

Reply Score: 3

Blame everyone but Linux
by nt_jerkface on Mon 28th Dec 2009 08:11 in reply to "The problem is not in KDE"
nt_jerkface Member since:

I see you made no mention of the unstable abi issue. Maybe if the Linux devs didn't have a quasi-religious view of open source drivers the video card companies would provide better support.

Here's the kernel dev's message to hardware companies:
Open your drivers or we'll f--king break them.

Oh and I just downloaded an Ati driver for a 5 year old laptop the other day and it worked in Vista just fine. No compiling needed, just click-click-click done. Remember according to Greg KH and others having a stable abi is nonsense even though Windows, OSX and Solaris all have one.

Edited 2009-12-28 08:20 UTC

Reply Parent Score: 3

RE: Blame everyone but Linux
by lemur2 on Mon 28th Dec 2009 08:49 in reply to "Blame everyone but Linux"
lemur2 Member since:

Here's the kernel dev's message to hardware companies:
Open your drivers or we'll f--king break them.

Hardly. The kernel dev's don't necessarily want hardware companies to open their code.

The kernel developers know far more about the Linux kernel than any hardware company does.

What is wrong then with the hardware companies just providing programming specs for the hardware to the Linux kernel dev's?

Like this, for example:

This information isn't giving away AMD's IP in any way. It is just programming specs (as in ... "in order to enable this function, set this register to that value"). No other company is going to be able to make a competing chip using just this information ... but by using these programming specs the Linux kernel devs sure are (and have been) able to write a better Linux driver than the hardware companies themselves can.
"Also available is a driver known as "ati", "xf86-video-ati", "video-ati" and "radeon"."

PS: This one is the driver that has recently been included in the Linux 2.6.32 kernel.

It is a win-win. ATI/AMD get a decent 3D-capable Linux graphics driver for their chips. ATI/AMD don't have to maintain it any more. ATI/AMD's customers running Linux are happy. They win. They are henceforward far more inclined to specify ATI/AMD graphics for their machines running Linux. ATI/AMD are happy.

Everybody involved wins.

No valuable IP given away.

Edited 2009-12-28 09:06 UTC

Reply Parent Score: 5