Finally, Team Fortress 2 Benchmarks For Phoronix!

[smitty3268]

There isn't a single ounce of logic in what you just said. That was an incredibly stupid thing to say.

It's like talking to a 2 year old. He actually thinks he is being incredibly smart here, and that we're the ones who can't comprehend his brilliance.

[duby229]

Timedemos run the same each subsequent run...

Which is not the same as gameplay. Subsequent runs of actual gameplay will not be exactly the same. Timedemo's therefore are not representative of gameplay. It really is that simple.

[duby229]

It's like talking to a 2 year old. He actually thinks he is being incredibly smart here, and that we're the ones who can't comprehend his brilliance.

You are the one that can't comprehend that gameplay is not exactly repeatable... That repeating a timedemo breaks the entire purpose of benchmarking a game. It isnt brilliance. It's simply common sense.

[smitty3268]

You are the one that can't comprehend that gameplay is not exactly repeatable...

No, I understand that just fine. The thing you don't seem to understand is that it's a meaningless point. It would be like me pointing out the sky is blue - yes, yes, that's true.

That repeating a timedemo breaks the entire purpose of benchmarking a game.

It certainly doesn't for me, or 99% of other people. I'm willing to believe that different people have different purposes behind benchmarking, so i won't ignore this argument out of hand, but it seems very weak to me. I still don't understand what you think the purpose of benchmarking is, to be honest.

It isnt brilliance. It's simply common sense.

It's certainly something.

[mmstick]

Which is not the same as gameplay. Subsequent runs of actual gameplay will not be exactly the same. Timedemo's therefore are not representative of gameplay. It really is that simple.

Either you're really stupid and lack all common sense, or you're just trolling...

[duby229]

No, I understand that just fine. The thing you don't seem to understand is that it's a meaningless point. It would be like me pointing out the sky is blue - yes, yes, that's true.


It certainly doesn't for me, or 99% of other people. I'm willing to believe that different people have different purposes behind benchmarking, so i won't ignore this argument out of hand, but it seems very weak to me. I still don't understand what you think the purpose of benchmarking is, to be honest.


It's certainly something.

I'm not trying to say that benchmarks shouldnt be run in controlled environments with precompiled binaries distributed to the client using preset configurations that are easily repeatable. What I am trying to say is that I think it's a great idea to be able to benchmark my own binaries in my environment that I set up. If I have a game installed, I think it would be kick ass cool to be able to fire it up and benchmark it while I play through some of it.

EDIT: And I'm damn sure I'm not alone in this desire... I think it could easily be done to have the mesa drivers report statistics during runtime, which could then be displayed using PTSs kick ass graphing abilities.

EDIT2: Plus thermal sensors and load. Lets face it PTS is awesome.

[mmstick]

I'm not trying to say that benchmarks shouldnt be run in controlled environments with precompiled binaries distributed to the client using preset configurations that are easily repeatable. What I am trying to say is that I think it's a great idea to be able to benchmark my own binaries in my environment that I set up. If I have a game installed, I think it would be kick ass cool to be able to fire it up and benchmark it while I play through some of it.

EDIT: And I'm damn sure I'm not alone in this desire... I think it could easily be done to have the mesa drivers report statistics during runtime, which could then be displayed using PTSs kick ass graphing abilities.

TF2. cl_showfps 2
Done, all the statistics you want but without the BS you claim is needed. It won't matter how many games you play in TF2, your performance will not vary greatly at all. I get over 300+FPS in Windows at all times regardless of what is happening in the game.

[LinuxGamer]

TF2. cl_showfps 2
Done, all the statistics you want but without the BS you claim is needed. It won't matter how many games you play in TF2, your performance will not vary greatly at all. I get over 300+FPS in Windows at all times regardless of what is happening in the game.

how do you get 300? i cap at 299?

[mmstick]

how do you get 300? i cap at 299?

fps_max lets you set a custom frame limit in Source engine. Mine hovers around 350-450.