Goodbye ATI

[Ansla]

First of all, thanks to M$ and Intel, 99% of laptops (at least in Portugal) have hybrid c****y muxless graphics. So, if you want performance on a laptop, you're forced to use either nVidia Optimus or ATI BACON.

No, that is true only if you want both performance and an Intel CPU.

That was my first obstacle on Linux, as nobody could ask me if ATI muxless was supported or not on Linux

You wanted to say nobody could answer? I have no interest in switchable graphics, but from reading the comments here on Phoronix it looks like AMD is the only one even trying to support them on Linux, and it actually works.

(and I didn't want to change the distro I was using (Arch x86_64)).

What's the distribution got to do with the drivers? It can only affect how easy it would be to install/configure those drivers.

So I chose a laptop with nVidia optimus (as it seems I can make it work more or less properly with bumblebee+bbswitch (I'm going to rant about this later...)).

It sounds like you enjoy a good chalenge

Secondly, my scientific apps (MATLAB, for instance) use specifically CUDA and not OpenCL (I think it's a shame, but that's developers fault too...).

Well, you could either blame the developers or blame yourself for choosing to use it, but not AMD.

[8bitbubsy]

I finally switched to NVidia (I ordered a GTX 560 Ti). [blah blah childish ranting]

You create a thread to tell people that you're parting with ATi products..? You mean absolutely nothing in general, no offense, but it's true. No one cares about your daily life inquiries except for your friends and family, and you're 1 out of millions of ATi users. Your ego level is too high, but so is mine for registering here just to post this. I guess we're kind of even. Okay, this is creepy, gotta go.

[Lasse.]

For some reason I keep buying cheap computers like my Acer 722 that have ATI video. It's always a struggle too.

I think that if enough people tell them "good bye" they will make a better driver.

Doubt that, AMD clearly isn't interested in their Linux-userbase. Releases of documentation is nice, but it feels more like AMD is trying to make others do THEIR work. It's their hardware afterall.

Linux-userbase is nothing compared to Windows gaming-community which is the biggest income for AMD when it comes to Radeons.

Generally AMD-hardware works ok in Linux if 2D is ONLY thing you need, but that's not the case with many users. I switched to Nvidia GTX560Ti from RadeonHD4850 with my computer-upgrade in august 2011 and it sure made things lots easier.

Also, dunno what's with certain person's claims & absurd lies in this topic. Looking the world through AMD-glasses (or with any fanboy-glasses) isn't very healthy.

[homerhomer]

Good for you

For some reason I keep buying cheap computers like my Acer 722 that have ATI video. It's always a struggle too.

I think that if enough people tell them "good bye" they will make a better driver.

[evolution]

[Rant] Another one that joined the green "Club"...

I know this thread is a bit dead now, but I've read it quite some time ago and I'd like to share my 2c.

After almost 15 years using AMD/ATI cards (on both Windows and Linux OS's), now I'm stepping out too... My laptop with a Radeon HD 2600 burned and I urgently needed a new laptop.

First of all, thanks to M$ and Intel, 99% of laptops (at least in Portugal) have hybrid c****y muxless graphics. So, if you want performance on a laptop, you're forced to use either nVidia Optimus or ATI BACON. That was my first obstacle on Linux, as nobody could ask me if ATI muxless was supported or not on Linux (and I didn't want to change the distro I was using (Arch x86_64)). So I chose a laptop with nVidia optimus (as it seems I can make it work more or less properly with bumblebee+bbswitch (I'm going to rant about this later...)).

Secondly, my scientific apps (MATLAB, for instance) use specifically CUDA and not OpenCL (I think it's a shame, but that's developers fault too...).

Finally, and now my "rant":

a) 6 years later and we still don't have proper support using FOSS drivers or Catalyst. It's very basic at best. Power management is AWFUL (on desktops that's not a problem but that kills a laptop's battery life). 3D performance is still BAD (I don't know how nouveau has managed to catch up radeon in terms of 3D performance, is it due to ATI graphics architecture being much more difficult to program?). 2D acceleration, although works for basic compositing and such, I STILL can't use it for decoding basic things such as Full-HD videos on Linux (XV s***s btw). On lower-power systems such as ultrabooks, netbooks and HTPCs (with AMD APUs, OC), that should've been AMD's nr.1 priority on Linux, as a lot of people use HTPCs and netbooks for watching videos, without caring about codecs. Even on such c***y systems that use GPUs such as Intel HD2xxx/HD3xxx, I've better 2D support than any of ATI current solutions. I won't blame Bridgman and other ATI/AMD/Xorg Linux developers (such as Alex Duscher or agf5d), as I recognize their hard work for trying to make ATI/AMD hardware useful on Linux, (and I really know how hard must be be programming os ASICs), but they're doing something almost "impossible".

b) I'm also going to rant about Linux users' mentality because I simply don't understand this. A company that has fulfilled to support its hardware by releasing documentation to linux developers, why there's almost no one supporting / helping them after 6 years? Furthermore, when things go wrong why are 99% of the time ATI/AMD developers the ones to blame? To support my argument, I'll talk about the sorry situation of muxless hybrid graphics + Proprietary Drivers: I know that AMD/ATI has been trying to support muxless ATI+Intel hybrid cards with Catalyst. If it was AMD/ATI that wouldn't support (hybrid graphics) I bet no one would try to help them make it work on Linux. But in nVidia's case, no! That's always someone's else fault... Not nVidia's fault for not listening to its users... And the biggest irony is that community has done a great work trying to support nVidia optimus with both nouveau and nVidia's proprietary driver. And where's someone trying to make things work, for instance, in my distro of choice (Arch Linux) (or other distos), when ATI proprietary drivers already give us some kind of support to AMD+Intel muxless laptops? (Vi0L0, I think you should see Chakra's PKGBUILDs for using a recent Catalyst with PowerXPress on a Arch-based distro). It seems Linux users mentality didn't almost change albeit ATI committed support for providing documentation to the Linux community, and that s**ks.

Summarizing, these situations made me buy a Mid-to-High-End laptop with a nVidia optimus GFX (GeForce 555M). I'm really sad to abandon ATI's "ship", but there were some constraints that forced me to do it.

My 2 cents, and sorry if I was harsh.
3V0LUT10N

p.s.: I wish the best of luck for ATI/AMD open-source driver developers...

[RealNC]

Indeed. Good thing the hardware doesn't suck ;-)

[curaga]

That goes both ways, drivers are worth nothing if the hardware sucks

[RealNC]

Hardware is worth nothing if the drivers suck

[DaemonFC]

I don't count open source drivers because they don't exist, even the gallium r600 driver has closed firmware. About 1.4ghz radeon, that was with ice cooling, the true maximum clock is 1.1ghz: http://www.tomshardware.co.uk/AMD-Ra...ews-37352.html Radeon7970 is only 20% faster than gtx580: http://www.xbitlabs.com/articles/gra...c_6.html#sect3 The end gtx600 specs are 1536(64bit) cores @1.8+ghz, in 28nm, 3.5Btrans, 200+-w: http://www.techpowerup.com/162500/GK...Explained.html and http://www.techpowerup.com/162504/NV...-Detailed.html How a card 3.6x faster than gtx580 and 3x faster than radeon7970 will lose to someone? Even the 1/3cores 130box mini Kepler will be magnificent!

Nouveau has free and open source firmware. They asked Nvidia if they could distribute theirs and Nvidia said no, so the Nouveau project came up with their own.

I guess it's an odd example of a company being openly hostile to free and open source software where some good came out of it.

I doubt we'll ever be able to use most of our devices without blob firmware. Sometimes it's hard to convince a company to allow their blobs to be redistributed. So many distributions like Debian and Fedora give up and try to rationalize blob firmware as being somehow different than proprietary software so that they can look like they are not violating their own policies.

[artivision]

I don't count open source drivers because they don't exist, even the gallium r600 driver has closed firmware. About 1.4ghz radeon, that was with ice cooling, the true maximum clock is 1.1ghz: http://www.tomshardware.co.uk/AMD-Ra...ews-37352.html Radeon7970 is only 20% faster than gtx580: http://www.xbitlabs.com/articles/gra...c_6.html#sect3 The end gtx600 specs are 1536(64bit) cores @1.8+ghz, in 28nm, 3.5Btrans, 200+-w: http://www.techpowerup.com/162500/GK...Explained.html and http://www.techpowerup.com/162504/NV...-Detailed.html How a card 3.6x faster than gtx580 and 3x faster than radeon7970 will lose to someone? Even the 1/3cores 130box mini Kepler will be magnificent!