AMD A10-7850K Kaveri: Windows 8.1 vs. Ubuntu Linux

[smitty3268]

Open Arena openGL ES

You've said this before, and were wrong then and wrong now.

Open Arena uses desktop OpenGL. Not ES.

I know it's been ported to ES in the past, but the standard build that Michael uses is just plain old GL.

[ua=42]

Ah. Thank you. I'll fix that in my copy paste response.

[nightmarex]

Are the Catalyst drivers really that bad in terms of system stability? I've been wanting to slap together a mini-PC and was considering using an AMD APU.

Not really, micro stutters(really apparent in L4D2) are about the worst part of AMD drivers. It tends to smooth out over the course of a game but damn it can be annoying. I have had issue with plugging in new monitors and having catalyst freeze as well. as a HTPC down clocking works fine as well so it shouldn't be a bad choice. If you're going for gaming keep in mind that kaveri is good for dota 2 @ max http://www.hardwareheaven.com/review...8x-dota-2.html but IDK what you had in mind exactly. full HD gaming is still a discrete card thing at the moment... sadly. Maybe next gen I gen ditch my 7870.

[[deXter]]

I'm surprised no one's mentioned testing using the open-source radeon drivers yet?

Curious to see how the 7850K would perform under kernel 3.13 and the latest Mesa 10.1.

[Ericg]

I'm surprised no one's mentioned testing using the open-source radeon drivers yet?

Curious to see how the 7850K would perform under kernel 3.13 and the latest Mesa 10.1.

On Kaveri? Michael says in one of the other articles that latest -git code is broken on Kaveri

[ForkedPython]

Well that's sort of the issue with many visitors on phoronix - they treat this website as a place primarily for gaming and hardware recommendations, when that's really just a side objective (and yes, I know the title says otherwise but the title really needs to be changed to suit the content). There really needs to be an entirely separate website dedicated to benchmarking games and hardware in a consumerist perspective; phoronix seems to largely target enthusiasts, admins, and developers. I see no problem in how phoronix runs tests, and I think taking a consumerist approach would taint the website.


I've considered starting a website myself for strictly linux gaming and hardware reviews, but considering I don't have the time or money to pursue such a thing, I dismissed the idea pretty quickly. Besides, you'll never satisfy everyone. Linux users are hard to please - they're generally arrogant, fussy, and nitpicky.

I agree entirley. I wish someone would start a dedicated benchmark website (i dont have the funds atm) even if it just relied on people loaning cards. It sounds ridiculous but until AMD get flak over poor gaming results then they would happily sit until 2050 doing nothing about the game profiles/patches. Of course SteamOS brings them some embarassment when it starts to take off and people at conventions ask questions as to why AMD isnt in the race to provide the best PC gaming experience (given that there flat on their ass now with regards to powerful CPU's and can only cater for low end laptops and HTPC's with their APU's ((dont forget intel's are pretty great AND have better linux drivers and they will catch up and walk past AMD soon enough in this arena too))

Perhaps it really is time for some opensource GPU hardware ?

[Herem]

Perhaps it really is time for some opensource GPU hardware ?

Intel can't get close to the performance of AMD and NVidia with their considerable R&D budget. Do you really think an opensource GPU project would ever be successful without billions of investment?

[GreatEmerald]

Are the Catalyst drivers really that bad in terms of system stability? I've been wanting to slap together a mini-PC and was considering using an AMD APU.

If you don't care about demanding games, then I don't see why you shouldn't (just use the FOSS drivers).

[kokoko3k]

No they don't, there is that thing called optimization, you know? Often drivers have to be optimizied for specific app to work properly.

I strongly disagree, what you call "optimizations", i call "bad patches and workarounds" built around a particular game to make it run better but that often leads to other (previously) working fine games, to run slower.
Definitely a bad marketing pratice.

[wargames]

I strongly disagree, what you call "optimizations", i call "bad patches and workarounds" built around a particular game to make it run better but that often leads to other (previously) working fine games, to run slower.
Definitely a bad marketing pratice.

It does not necessarily lead to other games running slower...
They use something like: if (name_of_the_running_app == "foo") then apply_optimizations_for_foo()

And yes, this actually happens