I've seen a fair amount of users use browser benchmarks to measure browser performance. However I don't understand why browser benchmarks would be that useful. A fair amount of them (e.g Octane, JetStream, and Kraken) use numbers with no (seemingly) assigned meaning and the explanation of their benchmarks feel vague to me. It just seems way more reliable to check memory (the core program, extensions, and HTML/JavaScript objects themselves), CPU and GPU usage, and the source code to compare performance. Not all browsers function the same, however, so it's really not a good idea to compare them directly anyway.

EDIT: I meant browser benchmark sites, not browser benchmark themselves. I need to stop mistyping titles…

It's right here. I just couldn't find the link yesterday.

no.

Oh ok… Can some one else explain then?

Optimize Gas Passer (if it was unbanned) and Dragon's Fury plays.

lol I love how that post got a high number of upvotes and then just got downvoted to oblivion sometime later.

Always -dxlevel 100. DXLevel 81 runs an older inefficient pipeline meant for older systems. As such you get higher frame rates, but your game becomes more unstable and likely get worse frame pacing. DXLevel 9 conversely uses a more modern pipeline meant for newer systems, so you get more stability and better frame times, but also "lower" frame rates and dx 9 might run worse on older systems. High frame rates of course don't always equal better performance, because it's impossible to actually tell if every frame is actually rendering properly, so frame time is a better metric to use in this situation. [1][2]

TL;DR Just let the game set the dxlevel. A lot better than forcing it to values.

Right can you explain to me with specific details and references what's wrong about my guide? I can't even see anything wrong with it (aside from the anime reference, that just made me cringe).

Wait I was? I never even noticed until you said it.

Well yeah not me obviously, but there are other people who are experts when getting PCs (like Setsul). That's where I get some hardware buying tips from. Also I really worded that above quote badly; I should have just said I got the hardware tips sections from other people to begin with.

I can't I'm too addicting to posting. Some one lock this thread at some point.

Not all of it tho. I just took the objective performance facts from mastercomfig. The Hardware buying tips section comes from me alone, culminated with some horror stories of shitty computers from crappy commercial stores (cough cough Best Buy).

The steam browser uses an old version of chromium, where DNS would be used, thus DNS technically improves performance, but it's the steam browser so it doesn't make a difference anyway. Also I consider general OS optimizations to be good, regardless if they help with TF2 performance.

Also what's wrong with me using frame time?

Isn't Scout bad at team fights in general? (Without Mad Milk of course.)

I'm attracted to the Gas Passer. I'm not sure why, but I think it's the fact that the Gas Passer Gas has gravity, so you can make a slow falling wall and force people to panic. It works great with Dragon's Fury cuz that weapon deals direct damage and allows the pyro to quickly charge his Gas Passer, without the need for a taunt that puts him at risk (phlog).

So anyway can someone explain in full detail what's wrong with this guide?