Unpleasant install experience. HD5850 Ubuntu 10.4.

[mugginz]

yes i only have a hd4000 card to test.

but i think i buy a hd5000 card in september

I hope you wait until they release 10.8 and get some end user feedback before you buy.

I wouldn't want to see you get turned against ATI because their drivers suck when used with a 5000 series card. They probably need all the friends they can get with respect to their cards and Linux at the moment.

[Captain Sifff]

File a Bug report!

Hi Jecos,
you mentioned that you failed to boot Ubuntu but you can bring up a PCLinuxOS LiveCD with a working fglrx driver. This doesn't point to a particular Hardware issue. This sounds more like a configuration issue.
So if you used the Ubuntu repositories you either file a bug report with them, or if you tried the 10.7 script from the AMD homepage you can file a bug report at the AMD bugzilla that they misconfigured their Ubuntu packages. The Bug report 1810 does try to make the point that the driver fails to configure Radeon cards in a nForce 4 Series chipset(the MCP51 variant), but as you have an MCP55 this Bug report doesn't apply.
You can also try to go back to Ubuntu 9.10 for which I can say that catalyst 10.7 is working properly on it with a Radeon 5770 on an nForce 4 series chipset.
Good Luck!

[Captain Sifff]

Hello Quaridarium,
NVidia has released a plethora of chipsets over the years so, it's no wonder that some of the more popular chipsets work and some others don't.
To help others in deciding which configuration works you might post your board + chipset + CPU + GPU info.
If a combination works can depend on a lot of details.
Occasionaly() a chipset falls through AMDs testing and we have to tell AMD to test their driver against that particular combo(This is the hard part, but somewhen they listen).

[mugginz]

i don't care about this kind of Braindeath people

Well that would just about sum up your perspective on the matter.

If someone buys an ATI card and they have issues, it must be the end users fault.

If someone buys an nVidia card and they have issues, it's nVidia's fault.

Lets hope those who have already bought a 5000 series card that some posters on this board said were fine and a better option than an nVidia card don't have to wait for the 10.10 catalyst driver before being able to extract a good user experience from their card. For them to have to wait that long would make even more of a mockery of those who said 5000 series ATI cards were viable on Linux.

[Kano]

It is summer, no fglrx dev works, no progress. Well winter holidays must be very long at ati too

[mugginz]

users focused on closed source drivers are Guilty by definition.

And with that statement you finally loose the last bit of credibility you had as far as I'm concerned.

buy hardware for an closed source driver only like nvidia's is just stubit.

I see that you're unable or unwilling to understand that others have different requirements to you. Sometimes those requirements aren't met by the open drivers. I would hope that when you recommend ATI cards to others in the future, you only describe the functionality which is attainable from the open drivers. While this would mean that sometimes lesser hardware from nVidia driven by their blob may outperform ATI's card driven by an open driver in certain circumstances, at least you wouldn't be describing features and performance available via a mechanism that you feel is illegitimate and don't personally support.

in bridgman words the R800 mesa driver comes maybe @ August so you get full support for an hd5000 card in september.

Lets hope so but I'm not sure an ATI with open stack will compare to the alternative with a closed one in some circumstances though.

Full means clousedsource+opensource driver this is an good and solid base.

And ignores the reality that there is a range of use cases that this doesn't yet support.

in my point of view nvidia is bankruptet.

any words about nvidia is just lost of time.

It's looking more and more that this can be said of you.

I now think this view of you is warranted based on a reading of a decent amount of your posts over time. For one who is so for open drivers, you sure do have quite a closed mind.

you don't unterstand my words.

catalyst 10.9 is a good driver and 10.10 not why?
thats because 10.9 is the 3month bugfixed 10-6 catalyst
and the 10.10 is a new ('10.6' unstable) driver.
but yes this is only my point of view

I understand what you say on this particular matter. Previous similar statements that later weren't borne out in practice lead me to not value your perspective in this particular area very highly. I'm not saying these outcomes you describe are impossible or improbable, just that your assertion of them neither increases or decreases the likelihood of them in my mind. I'll continue to seek more reliable sources.

be sure over the last 2 years the open source radeon driver delivere a higher code quality than the catalyst.

this means if the r800 mesa driver comes out the hd5000 customers makes a good choise.

If you really feel this is the case I hope you recommend people not buy a 5000 series card until the r800 driver is actually delivering on this in practice.

[mugginz]

i don't care about my credibility

Clearly!

Your fokus on "credibility" is Pointles because i never ask you to have "credibility" in my person.

If you're not credible, you're not really worth listening to. Calling credibility "pointless"? It just gets better and better.

only 'evil' humans need 'credibility' to abuse and was naiv humans

This sounds pretty irrational to me, not merely wrong.

i'm not evil so i do not need credibility to abuse and was your trust in my person.

I certainly hope you're not evil.

I see that you're unable or unwilling to understand that others have different requirements to you.

Really? what is the magic requirment?

The open drivers don't support all of the functionality that fglrx does.

Sometimes those requirements aren't met by the open drivers.

this is not my word my word is its good to have a working fallback 'radeon'

its Pointles to argue in this way because i never dit that point.

my point is the hardware 'card' is ready if there are 2 Drivers funktional.

But that presupposes that 'radeon' is viable as a fallback. It may well be for some, but not for all unfortunately.

But,... are you saying they're both functional? They may both enable a frame buffer and other bits, but each of them don't seem to enable the full potential of the 5000 series cards in a reliable way just at the present though hopefully this will be addressed shortly. I do ask that you hold your comments to the same standard when discussing nVidia's products that you do when discussing ATI's. I ask, but I may not get.

If you believe nVidia or Inel should work to a certain level out of the box with a certain level of end user input, yet continuously give ATI a "free pass" when it doesn't meet the same standards you slam nVidia for not meeting, then that lessons the value of your comments in my view.

I would hope that when you recommend ATI cards to others in the future, you only describe the functionality which is attainable from the open drivers.

why should i do this? if this was my position I would only recommend intel hardware because 'intel' hardware only delivers only opensource drivers on linux

but opensource was a joke@intel in the past without spec's

and the driver in general only makes clear that linux allways the loser.

so i fuck on intel.

I'm saying that if you feel that closed drivers are illegitimate, then how can you consider functionality derived from ATI cards via fglrx, a closed driver, as legitimate. Surely using an ATI card with fglrx is just as "evil" as using an nVidia card with its blob in your mind.

While this would mean that sometimes lesser hardware from nVidia driven by their blob may outperform ATI's card driven by an open driver in certain circumstances, at least you wouldn't be describing features and performance available via a mechanism that you feel is illegitimate and don't personally support.

? sorry my english fails here

You seem to feel that closed drivers are evil.
Surely any functionality that's enabled via closed drivers is off the table in your mind if the driver to enable it is off the table. You seem to be saying people shouldn't use closed drivers. If they're not using a closed driver, the functionality that's exclusively available via that closed driver isn't available without using one.

Lets hope so but I'm not sure an ATI with open stack will compare to the alternative with a closed one in some circumstances though.

in my point of view its importand to seek a second opinion and the radeon driver is the second opinion.

???

End users have two choices. Open or closed driver. If you rule out the "evil" closed driver, that leaves you with the open one. An ATI card with the open driver, while superior in hardware, when compared to a slower nVidia card running its blob, may actually perform slower and lack functionality when compared to the nVidia option.

and nvidia ignores the soul of all linux dev's and users and in my point of view the linux users/devs will never accept this point.

That issue is a bit more complex than you make out there. While I hear (and usually ignore) those kinds of comments from Windows fabois, I would hope a long term Linux user would be more broadly across the various aspects of this issue and not try to simplify it beyond what is fair.

It's looking more and more that this can be said of you.
I now think this view of you is warranted based on a reading of a decent amount of your posts over time. For one who is so for open drivers, you sure do have quite a closed mind.

should i really do an Statement about that ?

Ultimately that's up to you.

I understand what you say on this particular matter. Previous similar statements that later weren't borne out in practice lead me to not value your perspective in this particular area very highly. I'm not saying these outcomes you describe are impossible or improbable, just that your assertion of them neither increases or decreases the likelihood of them in my mind. I'll continue to seek more reliable sources

yes... and sorry my english fails here its very hard for me to folling you here.

yes you always should seek a second opinion and prove this by own practise.

Basically that I place little value on your predictions.

If you really feel this is the case I hope you recommend people not buy a 5000 series card until the r800 driver is actually delivering on this in practice.

feeling? i buy an card for the opensource driver back in 2007 ;-)

yes it was a feeling about a good driver back in 2007 ;-)

If you think that in the future an open driver and 5000 series card will outperform in reliability, functionality and speed an nVidia card and blob combination then you should be clear that this is what you think, but is not yet reality. I would hate a new Linux user to read Phoronix forums and be lead to an inaccurate idea that they can today buy a 5000 series card and take it home, plug it in, and use it with the open driver and enjoy the type of outcomes they've been used to via Windows or be equal to an nVidia card and blob combo on Linux. All I ask for is accuracy. It's fine to outline what you get with an nVidia card, warts and all, but when evangelising for ATI you shouldn't forget the warts there as well.

[Dandel]

Thanks! I'll follow up on that bugzilla when i have time.

The bug report is a bit off... that's for nforce 4 chips... the system you have is noforce 5... so that requires a separate bug report...

Possible Title:
Nforce 5 series fails with radeon 5000 series graphics card

Commands you need to provide the list of pci devices...

lspci -nn

and in the bug report include the Processor Model, motherboard brand and model, kernel version and then the distribution.

as for log files, the dmesg log, and xorg logs should also be included.

[FunkyRider]

People just don't listen, get the fsck rid of your 5850 and buy a GTX 460 you will never be happier. Using a HD5000 in Linux now is like poking a chair leg into your ares!

[Dandel]

FunkyRider, That type of logic is what leads to bugs taking forever to get fixed. Honestly the main problems with the ati driver are the lack of bug reporters that create good bug reports. I know that the ati developers sometimes take more than a month to even confirm bugs. Although, once the bug can be confirmed, and reproduced, usually the fix will be created. Once the fix is created, the time for release can range anywhere from a month (next driver) to three months.