Announcement

Collapse
No announcement yet.

Finally, Team Fortress 2 Benchmarks For Phoronix!

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • #16
    Originally posted by schmidtbag View Post
    There is a lot of arrogance, ignorance, and stupidity in that post. If you actually read the article (or any previous articles where Michael stated why he didn't do Source engine games), you'd realize it wasn't a matter of ignoring you.
    Yeah, those reasons where all bullshit since no Windows review site gives two shits about having to run the tests manually.

    Originally posted by Michael View Post
    After communicating with Valve and trying out the latest TF2 version on Steam, the process exiting issue has been resolved. After working around another new library path issue, the Team Fortress 2 test profile is up and running!
    About time we see some due diligence.

    Originally posted by Michael View Post
    Other gripes I have and why the Steam-based titles are seldom benchmarked and covered on Phoronix include:

    - The Steam process is hard to fully automate (100%), with different per-distribution/OS caveats and needing an account for installing the games. (This is still an issue with the current Team Fortress 2 test profile and it assumes you already have Steam installed on the system.)

    - As can be seen by other common benchmarks like Unigine and Lightsmark and the id Software games, I don't really care if a given workload is open-source or not. Open-source is obviously preferred, but quality benchmarks carry more weight. Free is really nice for ensuring the community can reproduce the tests and compare the published results against their own systems. Having to deal with account registration and other hurdles decreases the likelihood others will want to reproduce and/or use the given test(s).

    - The game payloads aren't freely redistributable and they're huge in size. As an example, Team Fortress 2 takes up about 13GB on-disk even though we just care about running a small time demo. For test systems that are formatted frequently (can be multiple times per week or even multiple times per day when swapping out lots of hardware/software components), this is wasted bandwidth unless manually backing up the files. This can also be a problem for enterprise test farms that are "behind the firewall" and may not have public Internet access. For the freely distributable files and what's provided by most of the test profiles, the Phoronix Test Suite and its related components can take care of file/download management, locally caching the files for other Phoronix Test Suite systems by using an intranet server or USB storage device, etc. It makes the entire install/setup process very streamlined rather than needing to download gigabytes of data through Steam each time.
    And now you are back to doing a whole lot of pointless whining... So those who move distros constantly don't move /home to it's own partition or better yet, drive? You can just backup the Steam folder and never have to redownload your games again.

    Originally posted by Michael View Post
    - When installing Team Fortress 2 or other games through Steam, they take care of auto-updating, which is nice for gamers but not for benchmarking.
    This is the only valid complaint in the entire article.

    Just run the damn tests. ALL of them.

    Comment


    • #17
      Originally posted by Veerappan View Post
      I had horrible performance on my 7850 when I first tried to launch Steam games in Mint 15... and then I realized that Ubuntu/Mint don't build the radeonsi gallium driver by default and that my 32-bit mesa was falling back to software rendering. My 64-bit applications didn't have the same issue because I had hand-built the Radeon SI 64-bit mesa drivers (after building LLVM with the R600 back-end of course).

      You might want to double-check that you're actually using hardware acceleration.
      Yep, you gotta add in the Xorg Edgers PPA if you want to have the current Gallium3D or Catalyst drivers.

      TF2 will run on my junked parts box with an 8800TS 320mb and an Athlon64 3500+ Clawhammer, what holds it back is the DDR266 which makes load times stupid long for some reason as I read that this is a known issue.

      Comment


      • #18
        Originally posted by Kivada View Post
        Yeah, those reasons where all bullshit since no Windows review site gives two shits about having to run the tests manually.



        About time we see some due diligence.



        And now you are back to doing a whole lot of pointless whining... So those who move distros constantly don't move /home to it's own partition or better yet, drive? You can just backup the Steam folder and never have to redownload your games again.



        This is the only valid complaint in the entire article.

        Just run the damn tests. ALL of them.
        If you're trying to quote the article, at least remove the misleading quote ID, or better yet, quote the first post in the thread. Otherwise it looks as if you're misquoting.

        Also, the size is an issue. For those who don't have it installed, it would be a hell to download the whole thing just to run a benchmark. When doing a cross-distro test, even more so. It's not even that good of a benchmark, too, as TF2 is already quite outdated.

        Comment


        • #19
          Originally posted by GreatEmerald View Post
          If you're trying to quote the article, at least remove the misleading quote ID, or better yet, quote the first post in the thread. Otherwise it looks as if you're misquoting.

          Also, the size is an issue. For those who don't have it installed, it would be a hell to download the whole thing just to run a benchmark. When doing a cross-distro test, even more so. It's not even that good of a benchmark, too, as TF2 is already quite outdated.
          I agree it's not a great benchmark, but it's about 100 times better than anything else Michael has ever run.

          I agree with Kivada - screw automated benchmarking if it means we don't get any good benchmarks. Prioritize what's actually important.

          Actually, i suspect that is what michael has done, and these unspecified 3rd party companies that are paying him to develop PTS care about that, while phoronix readers get ignored because we aren't funding him.

          Comment


          • #20
            Originally posted by Veerappan View Post
            I had horrible performance on my 7850 when I first tried to launch Steam games in Mint 15... and then I realized that Ubuntu/Mint don't build the radeonsi gallium driver by default and that my 32-bit mesa was falling back to software rendering.
            Wait, so radeonsi can render source games now?

            Comment


            • #21
              Originally posted by smitty3268 View Post
              I agree it's not a great benchmark, but it's about 100 times better than anything else Michael has ever run.

              I agree with Kivada - screw automated benchmarking if it means we don't get any good benchmarks. Prioritize what's actually important.

              Actually, i suspect that is what michael has done, and these unspecified 3rd party companies that are paying him to develop PTS care about that, while phoronix readers get ignored because we aren't funding him.
              100 times better? Isn't Xonotic demanding enough for you? Stop hating games just because they're FOSS.

              Comment


              • #22
                Originally posted by GreatEmerald View Post
                If you're trying to quote the article, at least remove the misleading quote ID, or better yet, quote the first post in the thread. Otherwise it looks as if you're misquoting.

                Also, the size is an issue. For those who don't have it installed, it would be a hell to download the whole thing just to run a benchmark. When doing a cross-distro test, even more so. It's not even that good of a benchmark, too, as TF2 is already quite outdated.
                How is it misleading? Larabel wrote and posted the article, the Phoronix news bot didn't.

                And no, you only have to download it once and HDD space is stupid cheap. Furthermore you can move the Steam folder to any other distro just fine, even better, just keep your /home directory on it's own partition or another drive, meaning that you can move installs without having to restore your stuff from backups. At most all you need to do is run the "Verify Local Game Cache" test on the game installs.

                Originally posted by Calinou View Post
                100 times better? Isn't Xonotic demanding enough for you? Stop hating games just because they're FOSS.
                It's a poorly optimized and laid out extension of Quake3 being stretched far beyond where it should have dropped in favor of a newer engine. This is the problem with all of the OSS shooter benchmarks.

                Compare Xonotic to the HL2:Lost Coast benchmark or any of the recent Valve benchmarks. If you aren't testing several different engines you aren't getting a feel for what the GPU drivers are good at.

                Comment


                • #23
                  Originally posted by synaptix View Post
                  319.32 is a Linux only driver. :P Are you referring to latest Windows 7 WHQL 320.49 maybe?
                  lol sorry. Yes 320.49 is my Win driver.

                  Comment


                  • #24
                    Originally posted by Kivada View Post
                    It's a poorly optimized and laid out extension of Quake3 being stretched far beyond where it should have dropped in favor of a newer engine. This is the problem with all of the OSS shooter benchmarks.

                    Compare Xonotic to the HL2:Lost Coast benchmark or any of the recent Valve benchmarks. If you aren't testing several different engines you aren't getting a feel for what the GPU drivers are good at.
                    Can back it up with facts or links?

                    Comment


                    • #25
                      Originally posted by Ramiliez View Post
                      Can back it up with facts or links?
                      Run both with maxed out detail settings with 4xAA & 16xAF. HL2:LC looks much better, has more going on and isn't built on the ancient IdTech engine like every OSS shooter that isn't Cube or Saurbraten. For everything else, look it up your own damn self.

                      Comment


                      • #26
                        Originally posted by Kivada View Post
                        Run both with maxed out detail settings with 4xAA & 16xAF. HL2:LC looks much better, has more going on and isn't built on the ancient IdTech engine like every OSS shooter that isn't Cube or Saurbraten. For everything else, look it up your own damn self.
                        Even FUD from elanthis has much better information value than this. Actually his statements contain a lot of salvageable info.
                        Since you claim to be knowledgeable about the topic i naively assumed that you can back it with something not just with OOH Half-Life Tech Demo haz betterz texturez

                        Comment


                        • #27
                          Originally posted by Calinou View Post
                          100 times better? Isn't Xonotic demanding enough for you? Stop hating games just because they're FOSS.
                          It has nothing to do with FOSS.

                          Count how many people play TF2, and count how many people play Xanotic. 100 times was probably conservative.

                          It's also based off a semi-modern game engine rather than quake 3, so it's likely far more representative of other games people are playing, too.

                          Unigine tests are hard on the gpu, but they still suck as GPU tests because no real games use that engine. It's only semi-decent because it's one of the only tough tests michael runs that stresses the hardware. Xanotic is somewhere in between - some people are actually using the engine, and it's somewhat stressful, but still mostly sucks on both accounts.
                          Last edited by smitty3268; 07-18-2013, 06:25 PM.

                          Comment


                          • #28
                            big, Big, props for the work!

                            Now we're only missing frame latencies (micro stutter and the like) and PTS would be truly feature complete.

                            Edit: I was too quick to judge: http://www.phoronix.com/scan.php?pag...tem&px=MTQxNDI

                            Michael, I salute you. Two articles in one day, I don't know which is better news
                            Last edited by tomato; 07-18-2013, 06:38 PM.

                            Comment


                            • #29
                              Originally posted by Ramiliez View Post
                              Even FUD from elanthis has much better information value than this. Actually his statements contain a lot of salvageable info.
                              Since you claim to be knowledgeable about the topic i naively assumed that you can back it with something not just with OOH Half-Life Tech Demo haz betterz texturez
                              So you think that the OSS games currently use for benchmarking look as good as professional class games that have sold in the millions? So you think a poorly doctored up original Quake engine can compete with what Valve did years ago now.

                              Theres a reason nobody outside of Linus and BSD users has ever given a shit about the OSS games even though they are free, they instead flock to freeware or paid games, they are not and have not ever been good and the only people that played them save for the captive community of Linux and BSD users who basically had no other choice.

                              So again, look it up yourself.

                              Comment


                              • #30
                                Originally posted by Kivada View Post
                                Yep, you gotta add in the Xorg Edgers PPA if you want to have the current Gallium3D or Catalyst drivers.

                                TF2 will run on my junked parts box with an 8800TS 320mb and an Athlon64 3500+ Clawhammer, what holds it back is the DDR266 which makes load times stupid long for some reason as I read that this is a known issue.
                                Xorg Edgers PPA does not compile mesa/gallium3d with GLAMOR/s3tc so no radeonsi. It only carries fglrx which is completely broken beyond all repair on radeonsi. Oibaf PPA carries updated stuff but it's also broken on radeonsi and source games cannot be launched without crashing or spewing a ton of errors due to missing files/packages (and even if you manage to figure out all the files you need, it still explodes as soon as you launch a map). There is another ppa which has glamor drivers for radeonsi without bleeding edge stuff, but it doesn't have s3tc texture support so no source games.

                                In the end, people with HD 7xxx cards like myself are stuck until something is done.

                                Comment

                                Working...
                                X