|
don't even bother with linux + ati. that combination leads only to tears. linux graphics:
|
#
?
Nov 15, 2014 23:38
|
|
|
|
| # ? May 13, 2024 06:39 |
|
|
Notorious b.s.d. posted:don't even bother with linux fixed lol
|
|
#
?
Nov 15, 2014 23:40
|
|
|
Smythe posted:i think starcraft 2 has a native linux client also? and all source engine games? right? the source stuff does, possibly not tf2 for some reason but the others do. they even released many of the pre-source halflife1 games in linux. i suspect lots of the small timey developers work in linux vvvv whoops, you are correct crusader_complex fucked around with this message at 10:28 on Nov 16, 2014 |
|
#
?
Nov 15, 2014 23:50
|
|
|
TF2 was one of the first games they ported to linux
|
|
#
?
Nov 15, 2014 23:55
|
|
|
my amd in linux is much improved in the last year. don't use amd's drivers though use the open source ones they're surprisingly better
|
|
#
?
Nov 15, 2014 23:55
|
|
|
Silver Alicorn posted:TF2 was one of the first games they ported to linux garbage game for a garbage is lmao
|
|
#
?
Nov 15, 2014 23:58
|
|
|
my stepdads beer posted:my amd in linux is much improved in the last year. people have been saying this to me every year for ten years my stepdads beer posted:don't use amd's drivers though use the open source ones they're surprisingly better they're less crashy but they're still hella slow
|
|
#
?
Nov 16, 2014 00:35
|
|
|
Captain Foo posted:garbage game for a garbage is lmao TF2 is a really good game.
|
|
#
?
Nov 16, 2014 01:28
|
|
|
tf2 used to be a good game
|
|
#
?
Nov 16, 2014 01:43
|
|
|
*clears throat* I never liked TF2, and preferred Fortress Forever.
|
|
#
?
Nov 16, 2014 02:20
|
|
|
trilljester posted:I tried out elementaryOS last night on my daughter's old Acer netbook. It actually runs decent on a machine with only a gig of RAM. Serious needs suiting going on there. next try haiku! you don't even need that much RAM or disk for it!
|
|
#
?
Nov 16, 2014 06:50
|
|
|
fglrx is seriously garbage and has always been. The quality of the code in the userland AMD driver looks horrible from the outside: using valgrind on a program using the AMD driver causes valgrind to complain about the large number of errors (ioctls using unintialized structures, access to unintialized memory). In some error cases, instead of reporting an error to the caller, the AMD driver will simply call exit(123) and kill the whole application. This kind of issues impacted SDL 1.3: calling XCloseDisplay caused the driver to exit. A workaround was put in place later in SDL 2.0 to avoid this problem which should have never happened in the first place. Fun fact: this bug was found while writing a minimal program that reproduce the mipmapping issue… But bugs don’t only happen on fglrx: the Windows AMD driver also has a few major bugs. AMD supports a form of client-side buffer storage that would be extremely useful for Dolphin. It is exposed via the AMD_pinned_buffer extension. Using AMD_pinned_buffer with Vertex Buffers or Uniform Buffers works perfectly, but trying to use it with Index Buffers starts rendering random polygons. Because of this issue, we had to stop using AMD_pinned_buffer for Index Buffers, leading to decreased performance for AMD users of our OpenGL backend. To this day we’re still not sure how to report fglrx bugs to AMD: we haven’t seen developers reply to bug reports on their official forums, and while there is an unofficial bug tracker for fglrx issues it does not seem to be looked at by AMD developers and keeps accumulating new issues.
|
|
#
?
Nov 16, 2014 06:55
|
|
|
to be fair, valgrind requires annotations on all system calls and ioctls to determine which parameters are in, out or in/out.
|
|
#
?
Nov 16, 2014 07:02
|
|
|
pseudorandom name posted:to be fair, valgrind requires annotations on all system calls and ioctls to determine which parameters are in, out or in/out. How else is it going to work
|
|
#
?
Nov 16, 2014 07:05
|
|
|
Cocoa Crispies posted:How else is it going to work my point was that valgrind complains about fglrx because it doesn't have any of the annotations, not because fglrx is necessarily reading uninitialized memory otoh, http://richg42.blogspot.com/2014/05/the-truth-on-opengl-driver-quality.html is still as amusing as ever
|
|
#
?
Nov 16, 2014 07:06
|
|
|
Yeah, mesa fails the "clean valgrind" test too. The rest of fglrx is terrible. And then don't get me started about Mali. I started working on a Mali-based product in the last month, and I've already reported around 12 Mali bugs to ARM.
|
|
#
?
Nov 16, 2014 08:09
|
|
|
to amd's credit, they are taking measures to address their lovely drivers and they are doing it in a way that mostly placates the rms-style "NO CLOSED DRIVERS EVER  " sperglords. unfortunately the new drivers will only work on the R9 285 and newer cards so if you have a 7000 series or an 'old' R7/R9 card you are still screwed.it's too bad that amd's software sucks and maxwell is kicking their teeth in on the hardware side because they are actually fairly non-lovely as a company when it comes to publishing hardware specs and generally being less scummy about things compared to nvidia. Sudo Echo posted:But bugs don’t only happen on fglrx: the Windows AMD driver also has a few major bugs. AMD supports a form of client-side buffer storage that would be extremely useful for Dolphin. It is exposed via the AMD_pinned_buffer extension. Using AMD_pinned_buffer with Vertex Buffers or Uniform Buffers works perfectly, but trying to use it with Index Buffers starts rendering random polygons. Because of this issue, we had to stop using AMD_pinned_buffer for Index Buffers, leading to decreased performance for AMD users of our OpenGL backend. I liked how amd's opengl driver guy gave a talk at gdc about using multidraw indirect when their drivers still didn't have support for it, more than a year after the gl 4.3 spec was released. even mesa had support months before them. then, when the driver that did support it was released a month or so later, actually calling glMultiDrawArrays caused the driver to crash. The_Franz fucked around with this message at 16:33 on Nov 16, 2014 |
|
#
?
Nov 16, 2014 16:22
|
|
|
The_Franz posted:it's too bad that amd's software sucks and maxwell is kicking their teeth in on the hardware side because they are actually fairly non-lovely as a company when it comes to publishing hardware specs and generally being less scummy about things compared to nvidia. they behave this way because they're getting their asses kicked by nvidia it's always cheaper to publish information than to improve your proprietary driver on a platform where you have no market share
|
|
#
?
Nov 16, 2014 18:57
|
|
|
Suspicious Dish posted:Read the bug comment. Some people don't use workspaces to separate activities. you had a ux problem so you asked a bunch of what are probably the worst possible ux opinion havers what they'd like and so now you have a garbage option that satisfies nobody instead of doing one thing properly lunix in a nutshell
|
|
#
?
Nov 17, 2014 08:08
|
|
|
I just switched it so alt tab switches between windows in a single workspace like god intended. Problem solved.
|
|
#
?
Nov 17, 2014 08:29
|
|
|
I would straight up murder a man to have the multimonitor workspace behaviour shamelessly copied from how OSX has done it since mavericks.
|
|
#
?
Nov 17, 2014 16:53
|
|
|
just switched my work machine from a 2009 i7 920 with 18GB ddr3 ram running fedora to a 2008 mac pro with dual xeons and 32GB ddr2 fbdimm with osx it's a downgrade on hardware but holy poo poo is the desktop better, so much for linux on the desktop
|
|
#
?
Nov 17, 2014 18:24
|
|
|
with two more systemd-related resignations from debian over ian jackson, even bruce perens has joined in
|
|
#
?
Nov 17, 2014 18:39
|
|
|
Progressive JPEG posted:with two more systemd-related resignations from debian over ian jackson, even bruce perens has joined in that russ allbery letter is pretty good -- he goes out of his way to be reasonable and polite
|
|
#
?
Nov 17, 2014 19:17
|
|
|
idk what systemd is but it must be good if its making greybeards quit Linux.
|
|
#
?
Nov 17, 2014 19:19
|
|
|
ahmeni posted:you had a ux problem so you asked a bunch of what are probably the worst possible ux opinion havers what they'd like and so now you have a garbage option that satisfies nobody instead of doing one thing properly
|
|
#
?
Nov 17, 2014 19:21
|
|
|
Shaggar posted:idk what systemd is but it must be good if its making greybeards quit Linux. http://www.freedesktop.org/wiki/Software/systemd/ quote:Spelling aaaaaaaaaaaaaaaaaaaaaaaaaaaa
|
|
#
?
Nov 17, 2014 19:21
|
|
|
Progressive JPEG posted:with two more systemd-related resignations from debian over ian jackson, even bruce perens has joined in if i had a nickel for every time bruce perens "quit" debian i might retire
|
|
#
?
Nov 17, 2014 19:25
|
|
|
ahmeni posted:you had a ux problem so you asked a bunch of what are probably the worst possible ux opinion havers what they'd like and so now you have a garbage option that satisfies nobody instead of doing one thing properly a gui that does one thing properly is appealing as a design concept, but it doesn't satisfy the users who actually exist
|
|
#
?
Nov 17, 2014 19:26
|
|
|
Notorious b.s.d. posted:a gui that does one thing properly is appealing as a design concept, but it doesn't satisfy the users who actually exist the usual gnome response is to declare this a bug in the users
|
|
#
?
Nov 17, 2014 19:30
|
|
|
carry on then posted:http://www.freedesktop.org/wiki/Software/systemd/ "cache invalidation and naming things"
|
|
#
?
Nov 17, 2014 19:32
|
|
|
crusader_complex posted:i suspect lots of the small timey developers work in linux
|
|
#
?
Nov 17, 2014 19:34
|
|
|
carry on then posted:http://www.freedesktop.org/wiki/Software/systemd/ lmfao
|
|
#
?
Nov 17, 2014 19:37
|
|
|
Progressive JPEG posted:with two more systemd-related resignations from debian over ian jackson, even bruce perens has joined in and here's a third one now jeez
|
|
#
?
Nov 17, 2014 22:03
|
|
|
lmfao
|
|
#
?
Nov 17, 2014 22:05
|
|
|
ahahaha Bruce Perens is a huge babby
|
|
#
?
Nov 17, 2014 22:07
|
|
|
Notorious b.s.d. posted:a gui that does one thing properly is appealing as a design concept, but it doesn't satisfy the users who actually exist you wouldn't make a change to kernel scheduling based on how a bunch of graphic designers think, you'd take their input on responsiveness as part of the design process but develop something that works based on its own merits this is the Software Engineer Problem because engineers think that any non code issue is trivial so you get piss ui
|
|
#
?
Nov 17, 2014 22:24
|
|
|
I do a little bit of UI stuff at work and i try to reject as many direct suggestions from my users as possible, understand pain points instead and bring a spiked out UI to a UX and it results in way happier people all around because I know my limitations (and am colourblind lol)
|
|
#
?
Nov 17, 2014 22:28
|
|
|
ahmeni posted:you wouldn't make a change to kernel scheduling based on how a bunch of graphic designers think, you'd take their input on responsiveness as part of the design process but develop something that works based on its own merits kernel scheduling isn't a user interface. i can objectively measure how well a kernel schedules processes without ever asking a user about his preferences for a ui, to get any meaningful information, you have to listen to users. real users. if your users are sperglord loving systems administrators, make an interface for sperglord loving systems administrators. imagining some other audience you wish you had won't produce a good UI for the audience you actually have i mean, gnome 3 exists. that is pretty much all i have to say to refute any argument that mentions designers: "gnome 3" Notorious b.s.d. fucked around with this message at 22:36 on Nov 17, 2014 |
|
#
?
Nov 17, 2014 22:34
|
|
|
|
| # ? May 13, 2024 06:39 |
|
|
I write a UI and it's for sperglords, and they get exactly what they deserve.
|
|
#
?
Nov 17, 2014 22:38
|
|